Opinion Monitor for Development Policy 2024. Public Support and Criticism in the Context of Multiple Crises and New Guiding Principles

Abstract

The Opinion Monitor for Development Policy 2024 covers two main topic areas. The first part focuses on key attitudes of the German population towards development policy and development cooperation (DC) over the course of time. The stability of attitudes towards DC is also analysed for the first time. The second part addresses current global developments and new policy guiding principles. It sheds light on the attitudes of the German population towards development cooperation in connection with armed conflicts and economic crises as well as general attitudes towards foreign and security policy. It also explores the knowledge and attitudes with regard to feminist development policy. The report concludes with implications for development policy strategy and communication work.

Full text

1
PUBLIC SUPPORT AND
CRITICISM IN THE CONTEXT
OF MULTIPLE CRISES AND
NEW GUIDING PRINCIPLES
Opinion Monitor for Development Policy 2024
2024

The Opinion Monitor for Development Policy 2024 covers two main topic
areas. The rst part focuses on key attitudes of the German population
towards development policy and development cooperation (DC) over the
course of time. Among other things, general support for state DC and for the
current development cooperation/ODA expenditure, as well as for various
development cooperation actors and motives, is examined. The stability
of attitudes towards DC is also analysed for the rst time. The second part
addresses current global developments and new policy guiding principles.
It sheds light on the attitudes of the German population towards DC in
connection with armed conicts and economic crises as well as general
attitudes towards foreign and security policy, and analyses what factors
have an impact on the preferences for partner countries for German DC.
Italso explores the knowledge and attitudes of the German population with
regard to feminist development policy. Alongside the contents of feminist
development policy and the “feminist” label, the focus here is on support for
the guiding principle in the context of global crises. The report concludes
with implications for development policy strategy and communications.

PUBLIC SUPPORT AND
CRITICISM IN THE CONTEXT
OF MULTIPLE CRISES AND
NEW GUIDING PRINCIPLES
Opinion Monitor for Development Policy 2024
2024

IMPRINT
Authors
Dr Martin Bruder
Dr Alexandra Gödderz
Dr Sebastian H. Schneider
Dr Helge Zille
Responsible team leader
Dr Sebastian H. Schneider
Responsible head of department
Dr Martin Bruder
Layout
Zlatka Dimitrova, Katharina Mayer, DEval
Editing
Silvia Richter, mediamondi, Berlin
Picture credits
Illustration Cover: FAZIT Communication GmbH
Translation
KERN AG, Bonn
Bibliographical reference
Schneider, S.H., A. Gödderz, H. Zille and M. Bruder (2024),
Opinion Monitor for Development Policy 2024. Public Support
and Criticism in the Context of Multiple Crises and New Guiding
Principles, German Institute for Development Evaluation
(DEval), Bonn.
Printing
Bonifatius, Paderborn
© German Institute
for Development Evaluation (DEval),2024
ISBN 978-3-96126-227-4 (PDF)
Published by
German Institute for
Development Evaluation (DEval)
Fritz-Schäer-Straße 26
D-53113 Bonn
Tel: +49 (0)228 33 69 07-0
E-Mail: info@DEval.org
www.DEval.org
The German Institute for Development Evaluation (DEval)
is mandated by the German Federal Ministry for Economic
Cooperation and Development (BMZ) to independently
analyse and assess German development interventions.
The Institute’s evaluation reports contribute to the
transparency of development results and provide policymakers
with evidence and lessons learned based on which they
can shape and improve their development policies.
In its “Opinion Monitor for Development Policy” series,
DEvalregularly draws up analyses on the attitude of the
German population towards development policy and
cooperation as well as global sustainable development.
This report can also be downloaded as a PDF le
from the DEval website:
https://www.deval.org/en/publications

ACKNOWLEDGEMENTS
The DEval Opinion Monitor for Development Policy 2024
was produced with the assistance of numerous people and
organisations. We would like to thank them for their support.
We particularly wish to thank the reference group, whose
members actively supported the preparation of the study
concept and the report with their specialist input. The
following organisations were represented in the reference
group: the Federal Ministry for Economic Cooperation
and Development (BMZ), the Deutsche Gesellschaft für
internationale Zusammenarbeit (GIZ), KfW Development
Bank, CARE Deutschland e. V., Germanwatch e. V., Terre des
Hommes e.V., the Association of German Development and
Humanitarian Aid Non-Governmental Organisations (VENRO)
and WorldUniversity Service – Deutsches Komitee e. V.
We would like to thank Dr Markus Steinbrecher (Bundeswehr
Centre of Military History and Social Sciences, Potsdam) for
his excellent content-related, methodological and practical
suggestions as an external peer reviewer for the project.
Many thanks to Prof. Jennifer Hudson (University College
London), Prof. David Hudson (University of Birmingham),
Dr Paolo Morini (King’s College London) and Dr Soomin Oh
(University College London) for providing the Development
Engagement Lab (DEL) data.
We wish to thank the team from Respondi/Bilendi (Cologne)
for collecting the data. Thanks go to Rahkakavee Baskaranand
Friederike Bauer (&eect, Berlin) as well as to Jonas Volle (Univer-
sity of Magdeburg) for their assistance in analysing the data.
Finally, we would like to thank our colleagues at DEval,
whoprovided the Opinion Monitor team with dedicated support
in every stage of the project and thus made an important
contribution to the study. We would particularly like to mention
Prof. Jörg Faust and Heike Steckhan, who accompanied the
report as internal peer reviewers, as well as Simon Becker,
ZlatkaDimitrova, JohnDzaba, Merle Gmeineder, HannaLink,
Katharina Mayer, Caroline Orth, Carla Vasco Páez and
JelanaVajen.

vii
Executive Summary
EXECUTIVE SUMMARY
Background
As a result of numerous global crises and the lack of progress
in achieving the United Nations’ sustainability goals, German
development policy is being challenged more than almost ever
before. At the same time, many partner countries are becoming
more fragile, and the number of states ruled by an autocratic
regime is increasing. In the context of domestic developments
such as the strained economic and budgetary situation,
the general public in Germany is increasingly questioning
development policy. These challenges come at a time when
the German Federal Government is establishing new focal
points in its international engagement, for example with the
guiding principle of a feminist foreign and development policy
presented in 2023 (see Chapter 1: Development policy context
and objectives of the report).
Question
In light of these multiple crises and the new guiding principle
of development policy, the Opinion Monitor for Development
Policy 2024 provides important information for policy
communications, education and strategy by
1. providing a detailed picture of the attitudes of the
Germanpopulation towards development policy and
howthese have changed over time, selectively addressing
dierences between population groups,
2. examining attitudes towards development policy in the
context of general attitudes towards foreign and security
policy,
3. studying attitudes towards development policy in the
context of foreign and security policy crises and the
waragainst Ukraine with the associated challenges and
4. touching upon the population’s knowledge, understanding
and perceptions of and attitudes towards the guiding
principle of feminist development policy and its contents.
Data sources
The study uses a wide range of survey data collected in 2023 and
2024. In addition, secondary data collected by the AidAttitudes
Tracker (AAT) and the Development Engagement Lab (DEL) from 2013
onwards is analysed and time series of relevant attitude indicators
are updated (see Chapter 2: Data sources and methodology).
Attitudes towards development policy and
development cooperation over the course of time
Since the beginning of 2022, the respondents’ support for
development cooperation (DC) has declined. In January 2024,
only 47 percent of respondents advocated DC expenditure
remaining the same or increasing – a drop of 21 percentage
points. Decreasing support can be observed across the
entire political spectrum. The fact that the German Federal
Government provides DC and humanitarian aid in general is
met with a high level of support – from 63 and 72 percent of
respondents respectively, although this gure has declined
slightly since January 2023. In comparison with other policy
areas, many respondents see DC as an area in which they would
be most willing to cut funding (21% of respondents).
Even though the majority of respondents express a general
interest in the topic area of development policy/DC (57%),
only 34 percent feel well informed about it. In October 2023,
30 percent of respondents were aware of the Sustainable
Development Goals (SDGs) of the 2030 Agenda, but only
9percent knew precisely what these involve. A majority of over
60 percent had never heard of the SDGs.
“Doing good with modest resources” (46%) and “combating
the causes of ight” (43%) remain the most convincing motives
for DC in the eyes of the respondents. Other motives such as
“moral obligation”, “combating climate change” and “preventing
pandemics” have lost some support over the course of time.
According to the respondents, the European Union (49%) and
multilateral organisations (48%) are primarily responsible for
implementing DC, followed by the governments of partner
countries (41%) and the German Federal Government (35%).
The most frequently endorsed goals of DC are improvements
to water, sanitary and hygiene (WASH; 56%), education (47%),
health care (42%), and food and agriculture (41%).
The majority (58%) of respondents have a moderately positive,
but relatively unstable attitude towards DC. Only 19percent
can be identied as stable supporters, while 18 percent are
stable opponents.

viii Executive Summary
Respondents consider their self-ecacy in the eld of
development policy – i.e. the feeling that they can achieve
something themselves in terms of development policy – to be
low. Since the beginning of 2022, the feeling of self-ecacy
has dropped from 3.3 to 2.7 (on a scale from 0 = “can achieve
absolutely nothing” to 10 = “can achieve a lot”). The respondents
also continue to have a low opinion of the eectiveness of
German DC. Around 18 percent rate it as “very eective”, while
around 32 percent consider it to be “absolutely ineective”. Here
too, the assessment is more negative than at the beginning of
2022 (see Chapter 3: Attitudes towards development policy and
development cooperation over the course of time).
Attitudes towards development policy inthecontext of ...
… foreign and security policy
In the area of foreign and security policy, many respondents
most strongly endorse cooperation in international institutions
as well as diplomacy and striving for consensus (attitude
dimension: “cooperative internationalism”). The attitude that
Germany should keep out of international aairs received
a similarly high level of support (dimension: “isolationism”).
At the same time, fewer respondents agreed with a foreign
and security policy focusing on military might and deterrent
potential (dimension: “militant internationalism”). There is even
less support for a foreign and security policy centred around
assisting countries in the Global South (dimension: “global
justice”). This attitude dimension is conceptually closest to
general support for DC.
Respondents who advocate cooperative internationalism and
global justice are also more likely to be in favour of increasing
the DC expenditure, concrete DC interventions, development
policy aid for Ukraine and a feminist development policy.
However, respondents who support an isolationist approach
reject such interventions (see Section 4.2: Attitudes towards
development policy in the context of attitudes towards foreign
and security policy).
… security policy and economic challenges
Against the backdrop of security policy and economic
challenges, respondents show a high level of support for
international engagement in emergency situations. In this
context, they endorse measures to assist Ukraine more strongly
than measures to assist other countries.
Since the Russian war of aggression against Ukraine began,
respondents have had a more negative view of their future
personal and the future national economic situation. This goes
along with a decline in support for development cooperation.
Furthermore, experimental ndings show that a reference
to the tense budgetary situation in Germany has a negative
impact on support for DC across various demographic, socio-
economic and political characteristics.
If respondents in a survey experiment are informed about the war
in Ukraine, a war that is visible in the media and geographically
close, they express greater support for DC than those who have
been informed about the war in Yemen, which is less visible in
the media and geographically further away. This greater level of
support is primarily because the respondents feel more aected
by the situation in Ukraine, because they perceive the security
threat caused by this situation to be greater and because they
consider the people living there to be more similar to themselves.
When asked about specic countries, participants prefer countries
in sub-Saharan Africa as partners for German DC, whereas the
preference for a cooperation with China and countries on the
Arabian peninsula and in North Africa is lower. A cooperation
with Ukraine receives much greater support than in 2020.
In a comparison of hypothetical country proles, geopolitical
factors such as a close partnership with China or support for the
Russian war of aggression against Ukraine lead to a clear rejection
of partner countries. By contrast, the respondents prioritise
countries from which Germany obtains energy. Thenumber of
refugees coming to Germany from a potential partner country

ix
Executive Summary
has no eect on whether the respondents support DC with
this country. In addition, needs-based characteristics of
the countries, such as vulnerability to the consequences of
climate change or the level of poverty, play an important role.
Furthermore, respondents are more likely to reject countries
with a predominantly Muslim population as partner countries
(see Section 4.3 Attitudes towards development cooperation
in the context of security policy and economic challenges).
… feminist development policy
An increasing number of people are familiar with the term
“feminist development policy”. In January 2024, for example,
a narrow majority of respondents (52%) had heard of or read
about feminist development policy; 27 percent conrmed
that they knew what the term meant. This is an increase of 15
percentage points since June 2022. Even though the respondents
understand the main features of a feminist orientation,
they are largely unaware of more specic contents, such as
intersectional, post-colonial or power-critical approaches.
Whereas 59 percent of respondents still supported feminist
development policy in January 2023, this fell to 52 percent
in January 2024. In comparison with other focuses, such as a
“human rights-based” or “peace-promoting” development
policy, fewer respondents are in favour of a feminist development
policy. Concerning the goals of a feminist development policy
described in the strategy drawn up by the BMZ, respondents
primarily agree with the overarching goal of strengthening the
“rights, resources and representation of women”. However,
fewer respondents support the specic provision of nancial
resources for projects promoting equal rights.
The opinions of the supporters of dierent parties vary more
when the “feminist” label is used for development policy than
when the same feminist contents are presented without the label.
Around 28 percent of respondents are stable supporters
of feminist development policy. A total of 63 percent are
moderately unstable or ambivalent supporters, while 19 percent
can be described as stable opponents.
In times of global crises, feminist development policy is
more likely to be supported if it is justied on normative
or instrumental grounds than if no additional justication
is provided (see Section 4.4 The attitude of the German
population towards feminist development policy).
Implications for development policy strategy, communications and education
Implication 1: Support for development cooperation is waning. In this situation, development policy actors should look into
possible changes to their strategies and design policy communications and education proactively to meet future challenges.
Implication 2: A large proportion of the population has moderately positive, but unstable attitudes towards development policy.
Development policy actors should actively seek communication to maintain the existing development policy consensus.
Implication 3: The increasing interconnectedness between DC and geopolitical and security policy aspects is leading to new
areas of tension in development policy; here it is important to nd a new balance of interests and to communicate this
appropriately.
Implication 4: The polarising eect of the term “feminist development policy” requires careful consideration of whether and
how the label is used; contents that are capable of achieving consensus should be implemented and conveyed to the general
public (see Chapter 5: Implications).

CONTENT
Imprint iv
Acknowledgements v
Executive Summary vii
Content x
Abbreviations and acronyms xiv
1. Development policy context
and objectives ofthe report 1
2. Data sources and methodology 4
2.1 Aid Attitudes Tracker and
DevelopmentEngagement Lab 5
2.2 DEval surveys 6
2.3 Data sources for the analysis of Twitter/X posts 9
3. Attitudes towards development policy
and development cooperation
over the course of time 10
3.1 How strong is the respondents’
support fordevelopment policy and
developmentcooperation? 12
3.1.1 Support among the respondents for DC in Germany
has declined since the beginning of 2022 12
3.1.2 Support for development cooperation/ODA
expenditure is falling across the entire
politicalspectrum 14
3.1.3 In a cross-country comparison, respondents in
Germany show the lowest level of public support
for the current development cooperation/ODA
expenditure for the rst time 14
3.1.4 In spite of reservations about the current
development cooperation/ODA expenditure,
respon dents endorse DC and humanitarian aid
inprinciple 15
3.1.5 In comparison with other policy areas, respondents
see DC (and support for the unemployed) as the
areas with the biggest potential for reducing
expenditure 16
3.2 How interested are the respondents in development
policy/DC and how well informed do they feel? 18
3.2.1 The majority of respondents show an interest
indevelopment policy/DC, but only a small
proportion feel well informed 18
3.2.2 The SDGs remain little known
torespondentsinGermany 19
3.2.3 Development policy/DC and humanitarian aid
arementioned more frequently on Twitter/X 20
3.3 What are the respondents’ attitudes towards
DCmotives, actors and objectives? 22
3.3.1 “Doing good with modest resources”
and“combating the causes of ight”
remainthemost convincing motives for DC 22
3.3.2 Respondents most frequently consider
supranational and multilateral organisations
toberesponsible for putting DC into practice 23
3.3.3 The most frequently endorsed goals of DC
areWASH, education, health care, food
andagriculture 25
3.4 How stable are the attitudes towards
developmentpolicy and DC? 26
3.4.1 Many respondents are condent in their
attitudetowards DC and nd the topic relevant;
onlya few feel well informed. 27
3.4.2 Attitude types: the majority has a moderately
positive, but unstable attitude towards DC 30
3.4.3 Attitude strength as an important factor
forunderstanding uctuations in attitudes
towardsDC 33
3.5 How do the respondents rate their self-ecacy
andhow eective do they consider other
development policy actors to be? 34
3.6 How eective do respondents rate German DC? 35
3.7 How do respondents engage
indevelopmentpolicy? 36

3.8 Conclusion: German development policy
isatacrossroads 38
4. Development policy attitudes in the
contextofsecurity policy challenges
andfeminist development policy 40
4.1 Overview of the chapter 41
4.2 Attitudes towards development policy in the context
of attitudes towards foreign and security policy 42
4.2.1 Respondents showed most support forthe concepts
of cooperative internationalism and iso-lationism
and least for the concept of global justice 44
4.2.2 Respondents who advocate cooperative and
militantinternationalism also support various
development interventions more strongly 46
4.3 Attitudes towards development cooperation
in thecontext of security policy and
economicchallenges 48
4.3.1 How should Germany engage in international
crisesand emergency situations? 50
4.3.2 Economic crises and the German population’s
support for DC 53
4.3.3 How stable is support for DC in the face
ofmilitaryconicts and economic worries? 57
4.3.4 Which countries are preferred as
partnercountriesfor German DC in the
contextofgeopolitical tensions? 60
4.3.5 Conclusion: support for international engagement
is high, but could wane as a result ofongoing
economic challenges 66
4.4 The attitude of the German population
towardsfeminist development policy 67
4.4.1 What does the German population understand by
feminist development policy? 69
4.4.2 Does the population support feminist development
policy and its content? 73
4.4.3 How stable are the attitudes
towardsfeministdevelopment policy? 80
4.4.4 Feminist development policy
intimesofglobalcrises 85
4.4.5 Conclusion: a challenging label with contents
thatlend themselves to a consensus 88
5. Implications 90
6. Literature 96
7. Annex 108
7.1 Data sources 109
7.2 Supplementary analyses and gures 111
7.3 Study schedule 116
7.4 Study team and contributors 116

Figures
Figure 1 Development Cooperation: support and
assessment of eectiveness (2019–2024) 13
Figure 2 Cross-country comparison of public
supportfor the current development
cooperation/ODA expenditure 15
Figure 3 Support for the German Federal Government
carrying out DC and humanitarian aid 16
Figure 4 Support for expenditure in dierent
policyareas 17
Figure 5 Interest and knowledge in the area
of development policy/development
cooperation 18
Figure 6 Familiarity with the Sustainable Development
Goals in Germany (2015–2023) 20
Figure 7 Posts relating to development policy/
development cooperation and humanitarian
aid on Twitter/X (2019–2024) 21
Figure 8 Convincing motives for DC 23
Figure 9 Actors in the area of
developmentcooperation 24
Figure 10 Objectives of German development
cooperation: “target” v. “actual” 25
Figure 11 Score assigned to the characteristics
ofattitude strength 28
Figure 12 Shares of respondents according to their
score on the factors of attitude strength 30
Figure 13 Proportion of respondents assigned to
thefour attitude types (in percent) 31
Figure 14 The proles of the four attitude types 32
Figure 15 Average support for development
cooperation by attitude type 33
Figure 16 Eectiveness rating of various actors
indevelopment policy 35
Figure 17 Assessment of the eectiveness of
development cooperation (2013–2024) 36
Figure 18 Development engagement over
thecourseoftime 37
Figure 19 Attitude towards foreign and security
policyby party identication 45
Figure 20 Relationship between attitudes towards
foreign and security policy and attitudes
towards development cooperation 47
Figure 21 Support for Germany’s engagement in
dierent crisis and emergency situations 50
Figure 22 Approval of support for Ukraine and other
countries experiencing acute crises 52
Figure 23 Assessment of the economic situation
inGermany and of the respondent’s
ownhousehold (2019–2023) 54
Figure 24 Inuence of the assessment of the
economicsituation on support for DC 56
Figure 25 Eect of information on the budget freeze
and on the wars in Yemen/Ukraine on
supportfor development cooperation 58
Figure 26 Eect of information about the wars
in Ukraine/Yemen on support for
developmentcooperation 59
Figure 27 Eect of information on the choice
ofpotential partner countries for DC 63
Figure 28 Endorsement of possible
partnercountriesforDC 65
Figure 29 Feminist development policy/
foreign policy on Twitter/X (2021–2024) 70
Figure 30 Familiarity with feminist development
policy over the course of time 71
Figure 31 Target groups of feminist development policy
according to the German population 72
Figure 32 Support for a feminist development policy
over the course of time 73
Figure 33 Support for dierent focuses
ofdevelopmentpolicy 74
Figure 34 Support for the objectives by support
forfeminist development policy 75
Figure 35 Factors that predict support for
feministdevelopment policy 76
Figure 36 Support for Development cooperation/
ODAexpenditure by experimental group 78
Figure 37 Support by experimental group
andpartyidentication 79

Figure 38 Characteristics and factors of attitude
strength by type of development policy 81
Figure 39 Factors of attitude strength by type of
development policy (share of respondents) 82
Figure 40 Share of respondents assigned to
the fourattitude types for feminist
developmentpolicy (in percent) 83
Figure 41 The proles of the four attitude types
regarding feminist development policy 84
Figure 42 Support for feminist development policy
depending on the justication 87
Figure 43 Support for feminist development policy
byjustication and gender 88
Figure 44 LayoutSupport for development cooperation/
ODA expenditure by party voted for in the
last Bundestag elections 111
Figure 45 Policy area in which respondents would
bemost willing to cut expenditure 112
Figure 46 Preferred policy area for cuts by
partyidentication 113
Figure 47 Strength of the correlation between
theassessment of the economic situation
and support for development cooperation
indierent years 114
Figure 48 Strength of the correlation between
theassessment of the economic situation
andpublic support forthecurrent
development cooperation/ODA
expenditurein dierent years 115
Boxes
Box 1 Sampling error 6
Box 2 Recording gender and political orientation 7
Box 3 Use of online surveys in social research 8
Box 4 Statistical signicance 9
Box 5 Exploratory factor analysis 29
Box 6 Latent prole analysis (LPA) 30
Box 7 Structure of the conjoint experiment 61
Box 8 Limitations of development policy
communications 93
Tables
Table 1 Surveys: Aid Attitudes Tracker and
Development Engagement Lab 109
Table 2 Respondi/Bilendi surveys 110
Table 3 Study schedule 116
Table 4 Study team and contributors 116

ABBREVIATIONS AND ACRONYMS
3R+D
Rights, representation and
resources of women and
marginalised groups + diversity
AAT
Aid Attitudes Tracker
ACLED
Armed Conict Location
andEvent Data Project
AfD
Alternative für Deutschland
(Alternative for Germany)
AMCE
Average marginal
componenteect
BDI
Bundesverband der Deutschen
Industrie (Federation of German
Industries)
BMGF
Bill & Melinda Gates Foundation
BMZ
Bundesministerium für
wirtschaftliche Zusammenarbeit
und Entwicklung (Federal Ministry
for Economic Cooperation and
Development)
BVerfGE
Bundesverfassungsgericht
(FederalConstitutional Court)
CDU
Christlich Demokratische
Union Deutschlands (Christian
Democratic Union of Germany)
CEDAW
Convention on the Elimination
of All Forms of Discrimination
Against Women
CSU
Christlich-Soziale Union in Bayern
(Christian Social Union in Bavaria)
DAC
Development Assistance
Committee
DE
Germany
DEL
Development Engagement Lab
DEval
Deutsches Evaluierungsinstitut
der Entwicklungszusammenarbeit
(German Institute for
Development Evaluation)
ESEM
Exploratory structural
equationmodel
ESOMAR
European Society for Opinion
andMarket Research
EU
European Union
FC
Financial cooperation
FDP
Freie Demokratische Partei
Deutschlands (Free Democratic
Party of Germany)
FM
Federal minister
HIIK
Heidelberg Institut für
Internationale Koniktforschung
(Heidelberg Institute for
International Conict Research)
IPBES
Intergovernmental Science-
Policy Platform on Biodiversity
andEcosystem Services
IPCC
Intergovernmental Panel
onClimate Change
ISO
International Organization
forStandardization
LGBTIQ+
Lesbian, gay, bisexual,
trans,intersex, queer and
othergender identities
LPA
Latent prole analysis
NATO
North Atlantic Treaty
Organization
NGO
Non-governmental organisation
ODA
Ocial development assistance
OECD
Organisation for Economic
Co-operation and Development
OSF
Open Science Framework

SDG
Sustainable Development Goal
SIPRI
Stockholm International Peace
Research Institute
SPD
Sozialdemokratische Partei
Deutschlands (Social Democratic
Party of Germany)
TC
Technical cooperation
UN
United Nations
UNDP
United Nations Development
Programme
UNESCO
United Nations Educational,
Scientic and Cultural
Organization
WASH
Water, sanitary and hygiene
WFP
World Food Programme
ZMSBw
Zentrum für Militärgeschichte
und Sozialwissenschaften der
Bundeswehr (Bundeswehr
Centreof Military History
andSocial Sciences)

1.
DEVELOPMENT POLICY
CONTEXT AND OBJECTIVES
OFTHE REPORT

Development policy context and objectives ofthe report2
As a result of multiple global crises and the lack of progress
in achieving the United Nations’ sustainability goals,
German development policy is being challenged more than
almost ever before. Implementation of the 2030 Agenda and
its 17Sustainable Development Goals (SDGs; United Nations,
2015) is faltering and is unlikely to be achieved by 2030
(UnitedNations, 2023a). Climate change is also moving forward
(IPCC, 2023), biodiversity is declining (IPBES, 2019), the food
supply is critical in many regions (World Food Programme,
2022), and gender equality is stagnating (UN Women, 2023).
At the same time, development policy is facing changing
contexts in the partner countries. In many countries in
the Global North and South, democracy and the rule of law
are being eroded and nations are becoming autocracies
(Bertelsmann Stiftung, 2024; Freedom House, 2023; Nord et
al., 2024). This is also aecting partner countries of German
development cooperation (DC). In this context, state fragility
is also relevant and can put development progress at risk
(Faust et al., 2023; Wencker and Verspohl, 2019). The same
applies to an increase in armed conicts (ACLED, 2024). In
the foreseeable future, DCwill predictably have to deal with
the Russian war of aggression against Ukraine in particular –
involving high costs when it comes to tackling the individual
and societal consequences of the war (see, for example, BMZ,
2022, Grävingholt et al., 2023).
In this complex situation, German development policy
faces additional challenges as a result of domestic policy
developments, especially the strained economic and
budgetary situation. When the war against Ukraine began in
February 2022, sanctions were imposed on Russia by Germany
and the European Union (EU). This caused energy and food
prices to increase considerably (e.g. OECD, 2022). In addition,
in November 2023, the Federal Constitutional Court ruled that
the planned reclassication of budgetary funds for tackling
the coronavirus pandemic as funds for protecting the climate
was unconstitutional (Federal Constitutional Court, 2023).
This decision meant that savings were needed in the federal
budget. As a consequence of these developments, the use of
government funds is becoming an increasingly contentious
issue. Even development policy has not escaped this, as
the public debate about cycle paths in Peru in January 2024
demonstrated (Tagesschau, 2024). In this context, Germany’s
short-term domestic interests are becoming increasingly
prominent. For example, at the beginning of 2024, the
Federation of German Industries (BDI) repeated its demand
for development policy to be geared more strongly towards
Germany’s economic interests (BDI,2024). This demand is part
of an increasing polarisation of political discourse that reaches
the very centre of society and also aects development policy
(Herold et al., 2023).
All these challenges come at a time when the German Federal
Government elected in 2021 has formulated a feminist
foreign and development policy as a guiding principle. The
aim of this guiding principle, which was set out in the coalition
agreement (SPD, Alliance 90/The Greens and FDP, 2021)
for foreign policy and has been carried over to development
policy, is to increase the rights, representation and resources
of women and marginalised groups and promote diversity
(“3R+D” formula; Federal Foreign Oce, 2023a; BMZ, 2023a).
However, this has met with criticism from some politicians
and in some sections of society (e.g. Sassenhagen et al., 2023;
Schneider et al., 2024b). This could erode the hitherto relatively
broad consensus on development policy.

3Development policy context and objectives ofthe report
Especially against the backdrop of these multiple crises, it is
important for development policy actors to concern themselves
with the attitude of the German population. Thisis where the
Opinion Monitor for Development Policy 2024 comes in, as it
1. provides a detailed picture of the attitudes of the
German population towards development policy and how
these have developed over time, selectively addressing
dierences between population groups;
2. examines attitudes towards development policy in the
context of general attitudes towards foreign and security
policy;
3. studies attitudes towards development policy and DC
in the context of foreign and security policy crises and
the war against Ukraine as well as the resulting nancial
challenges;
4. touches upon the population’s knowledge, understanding
and perceptions of and attitudes towards the guiding
principle of feminist development policy and its contents.
To shed light on these aspects, a vast array of survey data
collected in 2023 and 2024 is used. In addition, as in previous
reports in DEval’s Opinion Monitor for Development Policy
series, secondary data collected by the Aid Attitudes Tracker
(AAT) and the Development Engagement Lab (DEL) from 2013
onwards is analysed and time series of relevant attitude
indicators are updated.
The report comprises two analytical parts: in Chapter 3,
selected attitude indicators are presented over the course of
time and the strength of attitudes towards DC is analysed.
Alongside established indicators such as basic support for
DC or the assessment of its eectiveness, the general level of
interest in development policy/DC and the extent to which
the German population considers itself to be informed about
this policy area are examined. Furthermore, psychological
research on attitude strength (e.g. Krosnick and Petty, 1995,
Vogel et al., 2016) is picked up and the strength of the attitudes
of the German population towards development policy/DC is
explored. The analysis of attitude strength should provide an
indication of how volatile the citizens’ DC-related attitudes are
in an intensied debate of development policy.
Chapter 4 focuses on detailed analyses of three aspects:
(1)the relationship between attitudes towards development
policy and the attitudes of the German population towards
foreign and security policy in general, (2) attitudes towards
development policy/DC in the context of security policy
challenges and economic bottlenecks, and (3) attitudes
towards feminist development policy. In the reports in the
Opinion Monitor series, attitudes towards development policy
have so far been considered in isolation. Owing to its close
connection to other areas of action within the sphere of foreign
policy, a holistic approach is appropriate (see, for example,
the German Federal Government’s national security strategy;
Federal Foreign Oce, 2023b). Two of the most dening
development policy topics of the last two years – DC in the
context of the war against Ukraine and feminist development
policy – have hardly been addressed in the literature to
date (Sassenhagen et al., 2023; Schneider et al., 2024b;
Zilleetal.,2023) and are explored in depth in this study.
The report concludes in Chapter 5 with implications for the
German Federal Government’s development policy strategy
and for communications and education.

2.
DATA SOURCES AND
METHODOLOGY

5Data sources and methodology
The Opinion Monitor 2024 is essentially based on data
from the Aid Attitudes Tracker and its follow-on project,
the Development Engagement Lab, as well as data that the
Opinion Monitor team has collected itself. In addition,
selective use is made of social media data from Twitter/X.
2.1 Aid Attitudes Tracker and
DevelopmentEngagement Lab
The Aid Attitudes Tracker (AAT) is a long-term cross-country
comparative study in which online surveys were conducted
every six months from 2013 to 2018 in Germany, Great
Britain, France and the USA. The study was carried out in a
panel design. This means that the same people, around 6,000
in total, were asked about their knowledge, their attitudes and
their engagement with regard to DC and other relevant topics
relating to development policy in each of the ten survey waves.
The participants were over the age of 18 at the time of the
survey and resided in the respective country, regardless of their
nationality or registered place of residence. The panel study
was funded by the Bill & Melinda Gates Foundation (BMGF)
and conducted by researchers from the University of Texas in
Dallas, the University of Birmingham and University College
London. The data was collected by the British market and
opinion research institute YouGov, which randomly selected
the participants from the institute’s pool of respondents (online
access panel). Various measures were taken during the survey
to guarantee the data quality, including eliminating extremely
quick or one-sided answers (YouGov, 2017).
Since 2019, the data collection work of the AAT has been
continued under the name Development Engagement Lab
(DEL) at University College London and the University of
Birmingham. It is still funded by the BMGF and carried out by
YouGov. Important changes are:
1. The data is now collected annually – instead of semi-
annually – and the number of questions in the panel survey
has been greatly reduced (N still ≈ 6,000 per country and
wave; data collection in September/October).
2. Some of the questions have been reformulated and the
order of the questions has been changed.
3. Cross-sectional surveys have been introduced as additional
forms of data collection.1
The dierences between the two surveys are improvements
to the survey design but mean that time series cannot simply
be updated. Shorter, less frequent and more comprehensible
surveys of changing samples are supposed to reduce survey
fatigue as well as learning eects over time. At the same time,
changes to the wording and to the order of the questions
can have an impact on the response behaviour (for example
question-order eects, e.g. Oldendick, 2008; Schuman and
Presser, 1996). Therefore, a seamless link between AAT and
DEL data is not possible. When interpreting time series, it is
important to take into account that changes in the results
could to some extent be traced back to these adjustments in
the methodology.
Unless stated otherwise, survey weighting is used in analyses
of AAT and DEL data. This ensures that the distribution of
important sociodemographic and political characteristics in
the sample corresponds to the distribution in the population
as a whole. An overview of all the AAT/DEL surveys used can be
found in Table 1 in the Annex.
1 The DEL tracker records the same key attitude and behavioural characteristics of the population (N ≈ 1,000) twice a year (January and June), while the DEL sandbox surveys
(generally in February and May) address current issues. Both are based on cross-sectional surveys with a new random sample each time.

Data sources and methodology6
Box 1 Sampling error
The shares and mean values of the analyses presented in the report show statistical uncertainties because of the use of
samples. For example, if 20 percent of 1,000 respondents in a ctional sample state that they have donated to a development
cooperation organisation in the past year, there is a 95-percent probability that the percentage of the entire population
lies between 17.5 and 22.5 percent (95% condence interval).2 If the sample size is increased to 5,000 respondents, there is
a 95-percent probability that the actual value lies between 18.8 and 21.1 percent. Generally, in the case of a sample of 1,000
respondents, for a dichotomous characteristic such as “yes/no” or “for/against” it is assumed that the sampling error amounts
to +/-3 percentage points (Erikson and Tedin, 2011, p. 30-31). High shares around 50 percent exhibit broader condence
intervals than small ones. To avoid overloading the text and diagrams, the descriptive representations of percentages in this
report usually do not contain any condence intervals, but always specify the number of observations.
2.2 DEval surveys
As well as using the AAT and DEL data, the Opinion Monitor
team conducts a regular DEval tracking survey and its own
survey experiments. The DEval tracking survey records key
indicators relating to attitudes, knowledge and engagement
in the area of development policy/DC twice a year in January
and June/July. To ensure that the respondents understand and
are able to answer the presented questions, the questionnaire
was developed in collaboration with experts from GESIS
– Leibniz Institute for the Social Sciences, and a cognitive
online pretest was conducted (Schick et al., 2022). The data is
collected using quota samples of around 2,000 people from the
online access panel of the opinion research institute Respondi/
Bilendi.3 These samples are representative of the German
population up to the age of 75 with respect to gender, level of
education, age and federal state. Various measures are taken
to ensure the quality of the data. For example, respondents
who repeatedly give atypical responses are eliminated from the
panel. Participants who complete a survey exceptionally quickly
or exhibit atypical response patterns (e.g. so-called straight
liners, i.e. respondents who always select an identical response
category across several groups of questions) are also removed
from the nal data set. Because of the stratied quota sample,
the results are analysed without survey weighting.
2 The box has been taken from the glossary of the Opinion Monitor for Development Policy 2022 (Schneider et al., 2022, p. xx) with minor adjustments.
3 Respondi/Bilendi meets the standards of the International Organization for Standardization (ISO) and the European Society for Opinion and Market Research (ESOMAR)
for online surveys.

Box 2 Recording gender and political orientation
In some places in the report, the relationship between attitude indicators and various sociodemographic and political
characteristics is examined. Standard predictors such as age, gender, level of education, region (Eastv. West Germany) and
indicators for recording political orientation are generally used for this. Two explanations are necessary here:
In the analyses, the relationship with gender is studied on the basis of a binary variable. In all the surveys commissioned
by DEval, the respondents are given the opportunity to specify their gender on the basis of the response categories “male”,
“female”, “non-binary” or “prefer not to say”. Here it is not specied whether the respondents are being asked for their social
gender or their biological gender. Therefore, the answer relates to how the respondents prefer to self-identify. As only a very
small number of respondents (less than 10 people out of 2,000 respondents) in all the surveys uses the category “non-binary”
and such a small category is hardly (statistically) informative, the analyses only distinguish between people who self-identify
as “male” (“men”) or “female” (“women”).
As in the previous Opinion Monitor studies (Schneider and Gleser, 2018; Schneider et al., 2019, 2021a, 2022), the classic left-
right scale with the end points “left” (scale value 0) and “right” (scale value 10) is used to measure political orientation.4
To make the analyses more tangible and make non-linear connections across the corresponding attitude indicators visible,
for descriptive purposes the scale is divided into ve groups (see Vehrkamp and Merkel, 2019, p. 30–31): left (0–2), centre-left
(3–4), centre (5), centre-right (6–7) and right (8–10).
The scale records fundamental notions regarding social interaction and politics (Erikson and Tedin, 2011, p. 72–73). “Left”
essentially stands for welfare-state benets, the regulation of the economy and progressive social policy, regarding for
example immigration, same-sex marriage and gender equality. In turn, “right” is associated with alean state, liberal economic
policy and conservative social policy. This includes, for example, traditional gender and marriage models as well as restrictions
on immigration. Accordingly, where respondents see themselves on the scale from left to right correlates with their attitudes
towards numerous political issues. This also includes attitudes towards DC (see, for example, Bodenstein and Faust, 2017;
Milner and Tingley, 2013; Schneider and Gleser, 2018)
However, the scale is subject to criticism because of the scope for interpretation. Studies show that the interpretation of
the scale end points varies in dierent population groups (Bauer et al., 2017; Jankowski et al. 2023; Zuell and Scholz, 2019).
Under certain circumstances, this may distort content-based conclusions. Moreover, several authors suggest measuring
political orientation (or ideology) based on a socio-economic dimension and a socio-political dimension (Evans et al., 1996;
Heathet al., 1994; for a current application, see, for example, Jankowski et al., 2019). Despite these limitations, the left-right
scale provides a simple, but useful tool for recording political orientation.
In addition, party identication is selectively also used as an alternative indicator. This indicator makes it possible to focus
on dierences along political party lines or even on lines of conict within the population.
In recent years, experimental designs such as those used in
this report have been increasingly applied in the study of
attitudes, including in the context of development policy
(see, for example, Bayram and Thomson, 2022; Eger et al.,
2023; Scotto et al., 2017). Their advantage lies in their ability
to reliably determine causal relationships. This is done by
randomly dividing the participants into two groups of equal
sizes and giving them dierent information (treatments).
They are then asked about their attitudes or other relevant
characteristics (outcomes; for an overview, see Gaines et
al., 2007; Mutz, 2011). This procedure makes it possible
to establish what the eects of the respective treatment
are, as other possible inuences can be eliminated by the
random assignment to groups. As well as the eect of the
treatment across all participants, it is also possible to study
whether the impact of the information varies depending on
4 The explanation was already used in a box in the Opinion Monitor 2022 (Schneider et al., 2022, p. 11) and has been adapted and expanded for this report.

Data sources and methodology8
particular characteristics of the respondents (moderation
analysis; see MacKinnon et al, 2012). For example, a test can
be carried out to ascertain whether people with aparticularly
negative attitude towards DC are less convinced by
information about successful development interventions
than those with a more neutral attitude (cf., for example,
theliterature on motivated reasoning; Taber and Lodge, 2006).
The experiments conducted follow the good scientic
practice of studies – for example through preregistration
– and avoid distortions by always drawing new samples
of respondents where possible. All the survey experiments
presented in the report, with the exception of the conjoint
experiment in Section 4.3.4 were preregistered with the Open
Science Framework (OSF). Apreregistration is used for planning
research studies in advance and determining which hypotheses,
methods and analyses are used. This promotes the transparency
and reproducibility of scientic studies (Nosek et al., 2018).
To prevent potential distortions that may arise as a result of
treatments in dierent survey experiments within a single
survey inuencing one another (Gaines et al., 2007; Transue
et al., 2009), the experiments were carried out separately in
dierent surveys. An overview of all the data collections carried
out by the study team can be found in Table 2 in the Annex.
Box 3 Use of online surveys in social research
In social research, it is becoming increasingly dicult to generate samples for data collection over the telephone, in person
or by post. However, online access panels also involve challenges. Sample surveys conducted over the telephone, in person
or by post are increasingly suering from low response rates. In online access panels, on the other hand, the participants must
register for the pool of respondents themselves. This raises the question of whether the population as a whole is represented.
In particular, people with no internet access are unable to participate in online surveys, and the people registered in the
panels may dier systematically from the general population (Baker et al., 2010; Cornesse et al., 2020; Jerit and Barabas, 2023;
Kohlerand Post, 2023). However, a current study from the USA shows that quota systems and quality management make
it possible to come close to the gold standard of a genuine random sample when population characteristics are specied
– especially in comparison with pure convenience samples without any quality assurance, monitoring of participation or
aquota system (Stagnaro et al., 2024). Online surveys also have the advantage that texts (or even audiovisual contents) can be
integrated without any problems, for example for survey experiments. Another point in favour of surveys that are conducted
without ahuman interviewer (unlike telephone or in-person surveys) is that the respondents may answer more openly and
may be less inuenced by social desirability (e.g. Holbrook and Krosnick, 2010; Kreuter et al., 2008; Tourangeau and Yan, 2007).
Incontrast to estimates of shares and mean values, multivariate analyses and experimental designs are not aected by the
aforementioned challenges, or are only aected to a lesser extent (Baker et al., 2010).
All in all, it is assumed in this report that the data collected using an online access panel is an approximate estimate of the
actual attitudes in the German population. Nevertheless, it must be assumed that this estimate is less accurate than when
agenuine random sample is used. Therefore, the expression “representative of the German population” is not used.

9Data sources and methodology
2.3 Data sources for the analysis of Twitter/X posts
In order to be able to look not only at public opinion, but
also at the public discussion or the information environment
of the German population in the area of development
policy and DC, data collected by the short message service
X (known as Twitter until June 2023) is accessed and the
frequency with which the topic is mentioned on the platform
is studied. Twitter/X is accessed via the media monitoring
software produced by Meltwater, which provides an interface
to the platform. This interface is used to identify and download
relevant posts (approx. 1.4 million posts in the period from
1May 2019 to 29 February 2024). Since 2019, a search term and
account list has been used for this. This list is regularly reviewed
to ensure that it is up to date. The number of thematically
relevant posts per day is calculated for the analysis. Posts may
relate to both development policy/DC and humanitarian aid,
as many organisations are active in both elds, making a clear
delineation between them impossible. In addition, a separate
search has been conducted for the topic “feminist foreign and
development policy” since September 2021 (around 69,000
identied posts by 29 February 2024).
Please note that Twitter/X is suitable for an analysis of
public online debates, but not for analyses of public opinion.
Firstly, only a small section of the population uses the platform
(2023: 8% of the German population over the age of 14; see
Koch, 2023); secondly, the users are not representative of the
population as a whole (e.g. Jungherr, 2019; Koch, 2023).
Furthermore, as on all social media platforms, there is a risk
that automated accounts (known as bots) can create or share
posts on Twitter/X. Especially in the case of controversial
political topics, there is a high probability of political actors
attempting to use such bots to inuence public opinion (e.g.
Martini et al., 2021). However, a review proves challenging, as
Twitter/X closed the public interface that was accessible for
research on 31 March 2023. A manual review of the data set
using the tool Botometer X (https://botometer.osome.iu.edu;
see also Yang et al., 2020) revealed that it is very unlikely that
the 20 most active accounts are bots. However, the presence of
posts by bots in the data set cannot be ruled out. It must also
be borne in mind that the use of the platform and therefore the
potential reach of posts declined as a result of the acquisition
of Twitter/X by the entrepreneur Elon Musk in October 2022
(e.g. Carr, 2023). The extent to which Twitter/X will continue to
be suitable for broadly mapping out public online debates in
the future must be reviewed or weighed up anew in each case.
Box 4 Statistical signicance
To check whether a relationship or dierence is statistically relevant, i.e. signicant, signicance tests are used. If the p-value
falls below the critical threshold of 0.05 often used in social sciences, then a relationship or dierence is considered to be
signicant (Bryman, 2016, p. 347). The p-value is the probability of observing the relationship found in the sample or an even
stronger relationship if the null hypothesis was true, in other words if there was no relationship (Goodman, 2008, p.136).
To put it simply, there is only a low probability that the result is due to chance, and the null hypothesis can be rejected.
In the case of p-values below 0.10 (i.e. 10%), a relationship is shown too, but the greater uncertainty regarding the rejection of
the null hypothesis is pointed out. Finally, in the case of p-values above 0.10, the null hypothesis is upheld.5
5 The box was taken from the glossary in the Opinion Monitor 2022 (Schneider et al., 2022).

3.
ATTITUDES TOWARDS
DEVELOPMENT POLICY AND
DEVELOPMENT COOPERATION
OVER THE COURSE OF TIME

are in favour
of maintaining
or increasing
DC/ODA
expenditure
are in favour
of reducing
DC/ODA
expenditure
Support is falling
across the entire
political spectrum
Government engagement
in development cooperation
is generally endorsed
Respondents continue
to rate the eectiveness
of DC as low
endorse an
engagement
in DC
are interested
in DC “Combating global
epidemics”
“Moral obligation”
“Combating
climate change”
“Combating the
causes of ight”
“Doing good with
modest resources”
Most frequently selected:
Stable opponents
Stable supporters
Ambivalent
supporters
Unstable moderate
supporters
feel well informed are not familiar with the
Sustainable Development
Goals (SDGs)
endorse an
engagement in
humanitarian aid
At a
glance
Support for development cooperation
Development cooperation in comparison
with other policy areas
Interest and feeling informed Motives for development cooperation
Self-ecacy Attitude strength
 
 
 
 
   
 
 
 
 
 
 
   
 
 
21 percentage points
less than in 2022
declining since
June 2022
24 percentage points
more than in 2022
For which of the following policy areas
would you be most willing to cut expenditure?
most frequently selected
Unemployment benets %
Development cooperation %
Economic promotion %
least frequently selected
Health  %
Internal security  %
Education  %
The share of the gross
national income spent on
development is greatly
overestimated
Respondents rate their
self-ecacy in the eld of
development policy as low
In January 2024, they assigned
it the lowest value to date since
September 2019, with a mean
value of 2.7 (on a scale of 0 to 10)
OPINION MONITOR FOR
DEVELOPMENT POLICY 2024
Attitudes towards development policy
and development cooperation

Attitudes towards development policy and development cooperation over the course of time 12
In nancial crises, political decision-makers often see
potential for savings in development cooperation
(e.g.Heinrich et al., 2016). The population’s support for DC/
ODA expenditure can also fall in crisis situations. In the winter
of 2023/24, acorresponding discussion can also be observed
in Germany (e.g. Tagesschau, 2024). The criticism expressed
was aimed at individual DC interventions and contrasted these
with the frequent lack of funds for interventions that benet
the population in Germany. Such criticism, which sometimes
uses a populist “us against them” rationale can in turn reduce
public support for development cooperation (see, for example,
Bayram and Thomson, 2022; Bayram et al., 2024).
Against this backdrop, this chapter analyses public opinion
of development policy and DC in order to provide a strong
empirical foundation for development policy strategy
and communications. The analysis maps out established
attitude indicators such as general support for development
cooperation, support for the current DC/ODA expenditure
(Section 3.1), the assessment of the eectiveness of DC
(Section 3.6) as well as self-ecacy with regard to development
policy (Section 3.5) and, nally, development engagement
(Section 3.7). In addition, newindicators in the form of the level
of interest and knowledge in the area of development policy/
DC (Section 3.2) and questions regarding actors and objectives
in DC (Section 3.3) are mapped out. The stability of the attitudes
towards DC is also examined so that conclusions can be drawn
about the potential impact of information (Section 3.4).
3.1 How strong is the respondents’ support for
development policy and development cooperation?
3.1.1 Support among the respondents for DC in
Germany has declined since the beginning of 2022
Since January 2022, support for the German Federal
Government’s current DC/ODA expenditure has fallen
steadily. In January 2024, 47 percent of respondents (dark blue
line in Figure 1) indicated that the current expenditure should
be increased or at least maintained. In January 2022, this share
was still 68 percent.6
General support for development cooperation is also
declining substantially. Ever since 2013, the Opinion Monitor
(Schneider and Gleser, 2018; Schneider et al., 2022) has
observed the general support for development cooperation
among the German population (see yellow line in Figure 1). In
January 2024, 27 percent were in favour of generous support for
the countries in the Global South; this is 9 percentage points
less than in January 2022.
The group that considers DC to be eective or very eective
remains largely stable. However, the group that regards
DC as ineective is growing. In January 2024, 19 percent of
respondents rated DC as “(very) eective” (light blue line
in Figure 1); in January 2022, this was slightly higher at 22
percent. However, the share of respondents who consider DC
to be “(absolutely) ineective” has risen from around 21 percent
before the start of the war against Ukraine (February 2022) to
around 32 percent in January 2024 (+ 11 percentage points).7
Most of the remaining people selected a middle response
category (approx. 39%).
6 Both the data from the DEL tracker and the data from the subsequent DEval tracking survey were collected at a time when the construction of cycle paths in Peru was already
being criticised in the wake of protests by the agricultural sector in Germany (DEL tracker: 8– 13 January 2024; DEval tracking survey: 10–17 January 2024).
7 A complete representation of the time series, including all the response categories, can be found in Figure 17 in this chapter. The time series for the remaining variables from
Figure 1 can be found in Section 1.2 of the online appendix.

13Attitudes towards development policy and development cooperation over the course of time
Figure 1 Development Cooperation: support and assessment of eectiveness (2019–2024)
64
34
21
68
38
24
65
29
23
67
33
22
68
34
21
68
36
22
66
33
22
68
36
22
61
38
26
59
34
24
58
36
23
59
30
20
55
30
20
47
27
19
1. Coronavirus
lockdown
SPD, Green, FDP
coalition treaty
Beginning of
Russian attack
on Ukraine
Hamas
attack
on Israel
BVerfGE
ruling on
the federal
budget
0
20
40
60
80
100
9/19
12/19
3/20 6/20 9/20
12/20
3/21 6/21 9/21
12/21
3/22 6/22 9/22
12/22
3/23 6/23 9/23
12/23
3/24
Date of survey
Percentage of respondents
Public support for the current DC/ODA expenditure
[% don't change/increase]
Support for DC
[% generous support]
Effec�veness of DC
[% (very) effec�ve]
Source: DEval, own visualisation. Source of the data: DEL tracker and DEL panel 2019 to 2024. N tracker ≈ 1,00 0. N panel ≈ 6,000. DEL tracker surveys take place in
January and June, DEL panel surveys in September/October. Weighted data.

Attitudes towards development policy and development cooperation over the course of time 14
Overall, these gures indicate that support for DC in Germany,
after remaining high for a long time, is waning against the
backdrop of multiple crises – for example the repercussions
of the coronavirus pandemic, the deteriorating nancial
situation since the war against Ukraine began in February
2022 and the public discussion on the federal budget. Since
the beginning of the 1980s, a large majority of the German
population (70% and more) has considered it to be important
to support people in the Global South (see the data from the
Eurobarometer since 1983 in Schneider et al., 2022, p.9). This
support remained high in the 1990s and 2000s (e.g.Fransman
and Solignac Lecomte, 2004, p. 3; Stern, 1998, p.4) and did not
fall even during the so-called European refugee crisis of 2015
(Schneider and Gleser, 2018) and the coronavirus pandemic,
which raged from 2020 (Schneider et al., 2022). However,
asubstantial decline in support for DC can be observed since
January 2022.
Over the course of time, public support for the current
development cooperation/ODA expenditure is decreasing
across the entire political spectrum. The demand for a reduction
in development cooperation/ODA expenditure is not only
increasing among AfD-voters; it also meets with greater support
among some of the voters for the coalition parties, the SPD,
Alliance 90/The Greens and the FDP, in January 2024 than in
2022 (see Figure 44). Among CDU/CSU voters too, the number
of responses in the category “reduce slightly/a great deal” has
increased gradually since January 2022. It is only among people
who voted for The Left that no clear trend can be observed.9 This
nding suggests that support for DC interventions is dropping
in the wake of a growing threat perception as a result of global
conicts, a deteriorating budgetary situation and associated
consequences for the population, and media debates on
government expenditure. Therefore, the relationship between
the perception of the economic situation and support for DC is
addressed in more detail in Section 4.3.
3.1.2 Support for development cooperation/
ODA expenditure is falling across the entire
politicalspectrum
A breakdown by party preference shows that respondents
who vote for parties on the right of the political spectrum
support development cooperation/ODA expenditure
less than those who vote for parties on the left of the
political spectrum. Among those who vote for parties
that are categorised as being on the left politically
(Alliance 90/TheGreens, TheLeft,SPD), amajority is in favour
of maintaining or even increasing development cooperation/
ODA expenditure (51–73%); those who vote for centre-right
parties (CDU/CSU and FDP) fall just short of a majority in
favour of maintaining or increasing the expenditure, with
48 percent in favour of this in each case (see Figure 44 in
the Annex). Most of the AfD voters surveyed (71%) want
development cooperation/ODA expenditure to be reduced.8
3.1.3 In a cross-country comparison,
respondents in Germany show the lowest level
of public support for the current development
cooperation/ODA expenditure for the rst time
In Germany, public support for the current DC/ODA expen-
diture has noticeably been falling in comparison with France,
Great Britain and the USA for around two years. Whereas
Germany demonstrated the highest level of support until
January 2022, the DEL survey in January 2024 shows the lowest
support out of all four of the countries studied for the rst time
at 47 percent (see Figure 2). At 59 percent, the highest level of
support in January 2024 can be observed in France. This pattern is
primarily interesting because the formerly higher level of support
for development cooperation in Germany than in the Anglo-
Saxon countries was explained in the research by a dierent way
of organising the welfare state and a dierent redistribution of
funds by the government (Schneider et al., 2022, p. 13).
8 In all the surveys up to and including the DEL panel wave 9/2021, the party that the respondents voted for in the 2017 Bundestag elections was used for the breakdown; in all
the surveys from January 2022 onwards, it was the party voted for in the 2021 elections.
9 A similar pattern can be found when the ve waves of the DEL panel are used to break down public support for the current development cooperation/ODA expenditure on
the basis of political orientation on the left-right scale. Here too, public support for the current development cooperation/ODA expenditure is falling in the groups on the left,
centre-left, centre, centre-right and right (Figure 3 in the online appendix). In the political centre and on the right of the political spectrum, however, the decline in support is
greater. Furthermore, a similar picture emerges with regard to party identication (Figure 4 in the online appendix).

15Attitudes towards development policy and development cooperation over the course of time
Figure 2 Cross-country comparison of public support for the current development cooperation/ODA expenditure
64
68
65 67 68 68 66 68
61
59 58
58
55
47
66
70
65 66 67 67
64
67
66 65
62
65
60 58
46 47
45 44
44
53 53 53
56
50 48
51 51 52
60
59
54 61
53
54 57
56
54
59
54
54 56
57
0
20
40
60
80
100
9/19
12/19
3/20
6/20
9/20
12/20
3/21
6/21
9/21
12/21
3/22
6/22
9/22
12/22
3/23
6/23
9/23
12/23
3/24
Date of survey
Percentage of respondents
Germany
France
Great Britain
USA
Source: DEval, own visualisation. Source of the data: DEL tracker and DEL panel 2019 -2024 . Weighted data. The question was: “From its total budget of approximately
EUR 360 billion, the German Federal Government currently provides X. X percent – EUR XX.X billion – (note: the gures are updated each year) to poor countries
for development cooperation. Do you think that the government should increase or decrease the amount of money that it spends on development cooperation?”
Theresponse options were: “increase a great deal”, “increase somewhat ”, “don’t change”, “decrease somewhat”, “decrease a great deal”. The categories “increase”
and“don’t change” were conated for the gure.
3.1.4 In spite of reservations about the current
development cooperation/ODA expenditure,
respon dents endorse DC and humanitarian aid
inprinciple
The DEval tracking surveys ask about support for DC and
humanitarian aid separately. There is reason to assume that
the population does not make a distinction between DC and
humanitarian aid or has humanitarian aid in mind when
“development aid” or DC is mentioned (e.g. Riddell, 2007, p. 112).
For this reason, the Opinion Monitor surveys have examined
measures in these two areas separately since January 2023.
Both humanitarian aid and DC experience a high level of
support as elds of action for foreign and development policy
– but humanitarian aid more than DC. As Section Figure 3
shows, around 63 percent of respondents in January 2024
agreed that Germany should provide assistance to countries
in the Global South in the form of DC. For humanitarian
aid, it was 72 percent. Support is declining slightly in both
areas – by around 8 percentage points in comparison with
January 2023 for DC, and by just under 4 percentage points for
humanitarian aid. Against the backdrop of the public debate
surrounding the expenditure of the Federal Foreign Oce
and the BMZ, the majority of the population appears not to

Attitudes towards development policy and development cooperation over the course of time 16
consider it to be up for discussion whether DC interventions
(or humanitarian aid interventions) should be carried out.
Instead, the prevailing question seems to be how high the
nancial expenditure should be, and possibly what objectives
should be pursued.
Respondents who endorse the German Federal Government
carrying out DC measures also advocate humanitarian aid
and vice versa. The data shows a strong positive correlation
between support for the two elds of action.10 This suggests that
respondents do not really dierentiate between the two areas.
Figure 3 Support for the German Federal Government carrying out DC and humanitarian aid
11
71
14
5
12
70
14
4
15
63
16
5
8
76
12
4
8
77
11
4
11
72
12
4
Development cooperation
Humanitarian aid
1/2023 6/2023 1/2024 1/2023 6/2023 1/2024
0
25
50
75
100
Date of survey
Percentage of respondents
Agree
Neither agree nor disagree
Do not agree
Don’t know
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2023, July 2023 and January 2024. N for each
survey ≈ 2,000. The seven-point response scale was condensed for the visualisation. The question was: “Thinking about the extent to which you endorse dierent
kinds of support for developing countries, please tell us to what extent you agree or disagree with the following statements. 1. [humanitarian aid] In the context of
emergency aid or humanitarian aid, the German Federal Government should provide short-term assistance in emergency situations (e.g. natural disasters or famines)
in developing countries. 2. [development cooperation] In the context of development cooperation, the German Federal Government should work with developing
countries to reduce poverty and hunger in the medium and long term and to create ecient health care, education and economic systems in these countries.”
3.1.5 In comparison with other policy
areas, respondents see DC (and support for
the unemployed) as the areas with the biggest
potential for reducing expenditure
Besides support for the unemployed, DC is the policy area
in which respondents would most heavily cut government
expenditure. In January 2024, around 30 percent pleaded for
a reduction of the expenditure, while just under 28 percent
declared themselves in favour of an increase and approximately
37 percent advocated maintaining the existing expenditure
(see Figure 4).11 Fewer respondents endorse providing more
funds just to suppor t the unemployed than support an increase
in the development cooperation/ODA expenditure. Greater
expenditure on education, pensions and health care receives
a particularly high level of support. More than 70 percent of
respondents supported an increase in the expenditure in each
of these areas.
10 Pearson‘s r 0.66; p < 0.001. Source of the data: DEval tracking January 2024 (N = 2,101).
11 The dierences in comparison with the gures from the DEL survey in January that are presented in sections 3.1.1 and 3.1.3 can largely be explained by the fact that the specic
expenditure is included in the DEL question. Mentioning a specic sum in the billions of euros could give rise to stronger reservations with regard to DC.

17Attitudes towards development policy and development cooperation over the course of time
Figure 4 Support for expenditure in dierent policy areas
72 20 4 4
70 22 5 3
71 22 4 3
51 30 16 4
61 29 6 4
55 32 9 4
46 31 19 4
63 26 7 4
42 37 16 5
28 37 30 6
24 39 33 4
Unemployment benefits
Development cooperation
Economic promotion
Environmental protection
Defence
Digitalisation
Internal security
Transport
Health care
Pensions
Education
0 25 50 75 100
Percentage of respondents
Increase Remain the same Decrease Don’t know
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2024 (N = 2,101). The ve-category response scale was
condensed for the visualisation. The question was: “Do you think that government expenditure on the following policy areas should increase, decrease or remain the
same? When answering the question, please remember that the government budget is limited.” The question was taken from the population survey that was conducted
in 2022 by the Bundeswehr Centre of Military History and Social Sciences (ZMSBw) (Graf, 2022, p. 5).
When respondents were asked, in a direct comparison,
tomention a policy area in which they would be most willing
to make savings, DC and support for the unemployed were
also selected most frequently. 21 percent of respondents
declared themselves in favour of a reduction of expenditure
in the area of DC – and just as many supported a reduction in
the support given to the unemployed (Figure 45 in the Annex).12
This means that in this kind of survey too, DC comes far behind
policy areas such as “education”, “internal security”, “health
care” and “pensions”, which have a direct impact on the lives of
the German citizens or directly address their needs.
The proportion of mentions given to the policy area of DC
varies with party identication. Among sympathisers of
theAfD, the Sahra Wagenknecht Alliance (BSW), the CDU/CSU,
theFDP, the Free Voters and the SPD and among people who do
not feel aliated to a party, DC consistently has the highest or
second highest share of mentions (see Figure 46 in the Annex).
This means that these people see DC as one of the two areas
in which they would be most willing to make savings. It is only
among supporters of the Greens and The Left that DC is named
proportionately less frequently. Their most frequent responses are
the areas of “defence” and “promoting industry and the economy”.
12 In comparison with the results of the ZMSBw survey in June/July 2022 (Graf, 2022, p. 5; a representative telephone survey of the German-speaking population in private
households over the age of 16), the gures from January 2024 indicated a more critical attitude towards development cooperation/ODA expenditure.

Attitudes towards development policy and development cooperation over the course of time 18
3.2 How interested are the respondents in
development policy/DC and how well
informed do they feel?
In order to be able to better assess whether additional
information is relevant to the attitude of the German population
towards development policy, the DEval tracking studies have
started asking, since January 2023, about the respondents’
interest in development policy/DC and how well informed they
feel in this policy area. These indicators are particularly important
for development policy communications and education.
3.2.1 The majority of respondents show an
interest in development policy/DC, but
only a small proportion feel well informed
The majority of respondents are interested in development
policy/DC. This can be seen in the left-hand bar chart in
Figure 5. For example, 57 percent of respondents in January
2024 indicated that they are interested in the topic area.
However, 25 percent are not interested, and 16 percent selected
the response “neither interested nor uninterested”. No clear
trend can be observed across the three survey waves.
Figure 5 Interest and knowledge in the area of development policy/development cooperation
58
23
16
2
55
24
19
2
57
25
16
2
0
25
50
75
100
1/2023 6/2023 1/2024
Date of survey
Percentage of respondents
a
a
a
a
Interested
Neither interested
nor uninterested
Not interested
Don’t know
Interest in development policy/DC
24
37
35
4
27
32
38
3
26
34
36
4
0
25
50
75
100
1/2023 6/2023 1/2024
Date of survey
Percentage of respondents
Well informed
Neither well informed
nor poorly informed
Poorly informed
Don’t know
Feeling informed about development policy/DC
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2023, July 2023 and January 2024. N for each survey
≈ 2,000. The seven-point response scales were condensed for the visualisation. The questions were: “ How interested are you in topics relating to development policy,
development cooperation and global poverty?” “ How knowledgeable or well informed do you think you are on topics relating to development policy, development
cooperation and global poverty?”
A much smaller proportion of respondents feels well informed
about the topic area of development policy/DC. In January
2024, 34 percent classed themselves as (very/relatively) well
informed (right-hand diagram in Figure 5). 26 percent feel
neither well informed nor poorly informed, 36 percent see
themselves as (very/relatively) poorly informed. It is evident
that respondents feel slightly less well informed in the most
recent two survey waves compared to January 2023.

19Attitudes towards development policy and development cooperation over the course of time
In addition, there is a relationship between interest
in development policy/DC and how well informed the
respondents feel about this topic area: the more interested
a respondent is in the topic, the better informed they feel.13 It
is probable that a greater level of interest also leads to greater
eort to nd out about the topic. Conversely, however, more
information could also contribute towards an emerging or
growing interest in development policy/DC.
Interest and knowledge in the area of development policy/
DC are accompanied by increased knowledge in this area.
People who are interested in development policy/DC and feel
well informed are much more likely to have heard of the SDGs,
to be familiar with the United Nations 1.5°C climate target and
to indicate that they have heard of feminist development policy
than people for whom this is not the case. 14
Irrespective of the respondents’ interest and how well informed
they feel, the share of German’s gross national income that is
spent on development is greatly over-estimated. Those who
class themselves as very interested or well informed are not
able to assess the share substantially more accurately than
other respondents. For example, people who are interested in
development policy assume that the share allocated to ocial
development assistance (ODA) is around 9.5 percent; people
who are not interested assume that it is 10.7 percent. Those
who describe themselves as well informed estimate the share
at 9.6 percent; those who state that they are poorly informed at
10.5 percent. These estimates are considerably above the actual
quota of 0.79 percent in 2023 (OECD, 2024). The fact that the
expenditure is estimated as being considerably higher than is
actually the case is well known from the available literature
(Milner and Tingley, 2013, Scotto et al., 2017) and has already
been established for Germany (e.g. Schneider and Gleser, 2018, p.
33-34). However, it is interesting that this also applies to people
who consider themselves to be well informed and interested.15
There is a relationship between the formal level of education
and both interest in development policy/DC and the
subjective level of information; a relationship also exists
between interest in the topic and age and political orientation.
The higher the school-leaving qualication of the respondents,
the more pronounced their interest in DC and the greater their
reported knowledge (see also the ndings relating to political
knowledge in Germany, e.g. Tausendpfund, 2020). Younger and
older people are more interested in development policy/DC
than middle-aged people; this also applies to people who place
themselves on the left of the political spectrum (in comparison
with the political centre or right).16
With respect to how well-informed people feel, there are
also slight dierences depending on political orientation
and gender. People who place themselves on the political
fringes tend to feel better informed than people who position
themselves in the political centre. Furthermore, women feel
less well informed about development policy/DC than men.17
3.2.2 The SDGs remain little known
torespondents in Germany
Since they were adopted in 2015, familiarity with the United
Nations’ Sustainable Development Goals (SDGs) has
remained persistently low – the “People’s Agenda” has not
yet reached the wider population.18 At the half-way point of
the 2030 Agenda in October 2023, 9 percent of respondents
stated that they were familiar with the SDGs and knew what
they involved (see Figure 6). 21 percent are familiar with the
SDGs, but do not know what they involve. A clear majority of
63percent have not heard of the SDGs yet (for further details,
see Schneider et al., 2024a).
13 The Pearson correlation coecient is r = 0.39, with a signicance value of p < 0.001. Source of the data: DEval tracking January 2024 (N = 2,101).
14 All the analyses in this section are documented in Section 1.3 of the online appendix.
15 It is also interesting that the share is overestimated even in an online survey, in which the respondents, at least theoretically, have the option of looking up the answers to
knowledge-based questions on the internet. Therefore, this nding indicates that a large proportion of the respondents did not consult the internet for knowledge-based questions.
16 The results of the regression analyses can be found in Table 12 in the online appendix.
17 Only people who identify either as “female” or as “male” were included in the analysis, as the group of people identifying as “non-binary” was too small for a meaningful
analysis. For a detailed explanation of how gender information was collected in the surveys behind this report, see Box 2.
18 This particularly applies when we consider that the SDGs are better known to people with a higher level of formal education and a strong interest in politics (for details,
seeSchneider et al., 2024a).

Attitudes towards development policy and development cooperation over the course of time 20
Figure 6 Familiarity with the Sustainable Development Goals in Germany (2015–2023)
8
23
57
12
9
23
55
13
10
23
57
10
9
24
54
13
7
22
60
11
7
21
63
9
8
21
63
8
8
21
63
8
9
21
63
7
0
20
40
60
80
100
12/2015 7/2017 12/2017 7/2018 9/2019 9/2020 9/2021 10/2022 10/2023
Date of survey
Percentage of respondents
Yes, and I know what they are
Yes, but I don’t really know what they are
No
Don’t know
Source: DEval, own visualisation. Until July 2018 AAT panel, from September 2019 DEL panel. For each survey N ≈ 6,000. Weighted data. The question was: “Have you
ever heard or read about the United Nations’ Sustainable Development Goals?”
3.2.3 Development policy/DC and humanitarian
aid are mentioned more frequently on Twitter/X
The analysis of development policy contents on the short
message service X (known as Twitter until June 2023) makes it
possible to form an impression of what topics are relevant at a
particular point in time in the public debate on (development)
policy. The information environment also plays a role in
shaping the attitudes of the German population. The platform
Twitter/X in particular picks up on aspects of public discussions,
but can also have an inuence on them (e.g. Jungherr, 2019).
Therefore, Figure 7 updates the time series from past Opinion
Monitor reports.19 Alternative short message services do not
yet have the same reach. With regard to other social media,
messenger services such as WhatsApp, Facebook Messenger
or Telegram provide a poorer access to data, whereas other
platforms such as Facebook, Instagram, YouTube or TikTok
are harder to analyse because of their photo- and video-based
contents. In the future, the various platforms and their relevance
to the questions studied in the Opinion Monitor – and their
feasibility for analysis – will have to be constantly re-assessed.
19 For the limitations of Twitter/X analyses, see Section 2.3.

21Attitudes towards development policy and development cooperation over the course of time
Figure 7 Posts relating to development policy/development cooperation and humanitarian aid on Twitter/X (2019–2024)
Campaign
#entwicklungwirkt
Oxfam study on social inequality
World Bank study:
development aid
to tax havens
Moria
refugee camp
Oxfam study CO2emissions
Doctors Without Borders:
situation in Moria
refugee camp
UN: blockade on
aid in Syria
Evacuation of
local employees
in Afghanistan
Oxfam study on social inequality
GIZ
local employees
in Afghanistan
Russian
war of aggression
in Ukraine
Criticism
of DC/ODA
expenditure
Oxfam
study
on inequality
Economic
budgetary situation
in Germany
Hamas
terror attack on
Israel/
DC &
humanitarian aid
to Gaza
Criticism
of DC/ODA
expenditure
Farmers’
protest/
in criticism of
DC/ODA
expenditure
0
2.500
5.000
7.500
10.000
4/19 6/19 8/19 10/19 12/19 2/20 4/20 6/20 8/20 10/20 12/20 2/21 4/21 6/21 8/21 10/21 12/21 2/22 4/22 6/22 8/22 10/22 12/22 2/23 4/23 6/23 8/23 10/23 12/23 2/24 4/24
Date
Number of tweets per day
Source: DEval, own visualisation. Source of the data: around 1.4 million posts that were downloaded between 1 May 2019 and 29 February 202 4 using the analysis tool Meltwater, with the help of a
search term list. The search term list is documented in Box 1 in the online appendix. The vertical dotted lines mark the periods presented in the Opinion Monitor for 2021 or 2022.

Attitudes towards development policy and development cooperation over the course of time 22
For the new period under consideration from January 2022 to
October 2023, the number of posts relating to development
policy/DC and/or humanitarian aid observed each day is
similar to the previous period, apart from a few uctuations.
This “basic level” is interrupted by sporadic upward uctuations.
A rst such uctuation can be seen in January 2022. This consists
primarily of posts on how the German Federal Government and
government DC actors were handling local Afghan employees
after the Taliban took power in August 2021. It is important
to bear in mind that individual accounts have generated
alarge number of posts. From February 2022, the time series
reects the Russian war of aggression against Ukraine. In this
context, the posts primarily discussed the humanitarian aid
interventions undertaken by the German Federal Government.
In November 2022, a surge of posts explicitly criticising DC/
ODA expenditure was observed for the rst time. This aspect
is picked up again in August/September 2023 in the course of
discussions about the economic and budgetary situation in
Germany, also reected in spikes in the time series. The attack
on Israel by the Islamist terror organisation Hamas on 7 October
2023 and the subsequent Israeli military operation in Gaza led
to a considerable increase in the average number of posts per
day. The government is criticised for supporting an Islamist
terror organisation with DC and humanitarian aid in Gaza and
also for providing too little support for the people in Gaza.
With the German Federal Constitutional Court’s ruling on the
federal budget in November 2023 and the farmers’ protests
in January 2024, the time series spikes strikingly high. These
posts reect the criticism for DC/ODA expenditure in politics
and the media that pointedly questions the expenditure on cycle
paths in Peru and gender trainings in Rwanda, for example, and
contrasts this expenditure with a lack of funds for agriculture in
Germany (see, for example, Tagesschau, 2024).
On the whole, the increasingly critical tone of communications
on Twitter/X since autumn 2023 corresponds to a drop in
support for DC/ODA expenditure. However, the question of
whether the discussion of development policy/DC in the media
and politics is the reason for this decline is unresolved. It is
also conceivable that an increasingly negative perception of the
respondent’s own personal or the national economic situation
leads to conicts over the distribution of resources (see, for
example, Heinrich et al., 2016; Kobayashi et al., 2021) and thus
to reduced support for development cooperation. Therefore,
the relationship between the perception of the economy and
support for development cooperation is addressed separately
in Section 4.3.2. Another reason could be ashift in priorities
with respect to the responsibilities of the state because of
astronger sense of threat in the German population as a result
of the war against Ukraine (e.g. Graf, 2024, p. 6) – atopic that is
dealt with in Section 4.3.3. Furthermore, the choice of afeminist
guiding principle for German foreign and development policy
could have led to a polarised discourse and to the rejection of
current development policy in some circles (Sassenhagen et al.,
2023). This is examined in more detail in Section 4.4. Finally,
the response of development policy communications to the
expressed criticism may not have been sucient.
3.3 What are the respondents’ attitudes towards
DC motives, actors and objectives?
Especially in economically challenging times, it is important
to assess not only the extent to which general support for
development cooperation is changing, but also what attitudes
exist towards the various motives, actors and objectives in the
area of DC. For example, it is conceivable that the population
thinks that DC should focus more strongly on national
interests and that it is primarily seen as the responsibility
of non-governmental organisations (NGOs) or companies.
Thebelief that DC should be restricted more to the basic needs
of the people in the partner countries could also be prevalent.
3.3.1 “Doing good with modest resources”
and“combating the causes of ight” remain
themost convincing motives for DC
“Doing good with modest resources” and “combating the
causes of ight” are still the most convincing motives for DC
in the eyes of the respondents; the remaining motives follow
a long way behind. As can be seen in Figure 8, 46 percent of
respondents in January 2024 named “doing good with modest
resources” and 43 percent “combating the causes of ight” as

23Attitudes towards development policy and development cooperation over the course of time
the most convincing motive for DC. All the other motives were
named by just 25 to 29 percent of the respondents.
When the motives are considered over the course of
time, the proportion of respondents mentioning the
motives “moral obligation”, “combating terrorism” and
“combating global epidemics” has fallen slightly since the
survey in January 2022. “Combating climate change” is also
mentioned proportionately less frequently in the surveys in
October 2023 and January 2024 than in the previous surveys
(minus5percentage points; see Figure 5 in Section 1.2 of the
online appendix).
These results can be interpreted as a decline in the
persuasiveness of the narrative that DC can be used as a means
of overcoming global challenges. The fall in the proportion of
respondents mentioning the motive “moral obligation” can also
be seen as an indication that the German population is putting
aside the interests of the countries in the Global South in light
of economic challenges in their own country.20
Figure 8 Convincing motives for DC
29
46
43
29
25
27
29
Economic promotion
Combating terrorism
Combating climate change
Combating global epidemics
Moral obligation
Combating the causes of flight
Doing good with modest resources
0 20 40 60 80 100
Percentage of respondents
Source: DEval, own visualisation. Source of the data: DEL tracking, January 2024. N = 1,021. Weighted data. The question was: “Here are several possible arguments in
favour of development cooperation. Which of these arguments do you personally nd particularly convincing? (Please select all applicable answers.)”.
3.3.2 Respondents most frequently consider
supranational and multilateral organisations to
beresponsible for putting DC into practice
An important aspect is the question of which actors the
population considers to be responsible for putting the DC
measures into practice. Even though governments of the
donor states provide a large share of the nancial resources
for DC, DC interventions can also be implemented by civil
society organisations and private companies. This can also be
done by the governments from the partner countries as well
as local civil society organisations and private companies.
Other actors are multilateral and international organisations
such as the European Union (EU) and the United Nations (UN).
In population surveys on DC, however, it is often only the
government that is mentioned as a responsible actor.
20 The DEL panel has also found that the percentage of respondents agreeing with the statement “Countries like Germany should provide more funding for development
cooperation because it is morally the right thing to do” has decreased since 2022 (see Figure 9 in the online appendix).

Attitudes towards development policy and development cooperation over the course of time 24
Figure 9 Actors in the area of development cooperation
32
35
18
48
21
31
9
49
10
41
25
21
11
Towns and communities (DE)
Federal states
None of the above
Companies (DE)
Political foundations (DE)
Companies (developing countries)
NGOs (developing countries)
Church actors (DE)
German Federal Government
NGOs (DE)
Governments (developing countries)
Multilateral actors (e.g. UNDP)
European Union
0 20 40 60 80 100
Percentage of respondents
Source: DEval, own visualisation; source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2024. N = 2,101. The question was: “Development
cooperation can be carried out by dierent actors and organisations. Please name up to 5 actors and organisations that you believe are the best t for carrying out
development cooperation.”
The respondents most frequently consider multilateral
actors (e.g. the United Nations Development Programme,
UNDP) and the EU to be responsible for carrying out DC
measures. The proportion of respondents mentioning these
actors is around 48 to 52 percent across the three survey
dates (see Figure 6 in the online appendix and Figure 9 for
January 2024). These are followed by the governments of
the partner countries, which were mentioned by around 40
to 42 percent of the respondents in all the surveys. Then,
after a clear gap, come the German Federal Government,
civil society organisations (NGOs) and church actors, which
were mentioned by around 31 to 37 percent. The proportion
of respondents mentioning the German Federal Government
has fallen across the three surveys, whereas the proportion
mentioning NGOs and church actors has remained the same.
Here too, it can be assumed that the strained budgetary
situation has an impact on this decline. Only a small fraction
of respondents – around 7 to 11 percent – considered the
German federal states or the towns and communities to be
responsible for DC interventions. This share has proved to be
stable across the three surveys to date.
The high support for the EU and multilateral actors could
indicate that supranational and multilateral actors are
deemed to have a greater inuence than individual national
actors. It could also imply that the respondents advocate
dividing the expenditure of resources and the responsibility
between several DC donor states.

25Attitudes towards development policy and development cooperation over the course of time
3.3.3 The most frequently endorsed goals of
DCare WASH, education, health care, and food
andagriculture.
Overall, the data suggests that respondents advocate
a traditional approach to development cooperation with
afocus on basic needs and particularly approve of classic
DC sectors (education, food, health care). On the other
hand, more progressive goals such as protecting the climate,
women‘s rights and gender equality, or protecting minorities
receive much less support.
Figure 10 Objectives of German development cooperation: “target” v. “actual”
47
42
22
56
24
22
7
16
41
25
25
38
12
28
16
34
35
17
44
21
18
14
16
28
18
28
45
23
24
18
Infrastructure
Refugee assistance
Energy supply
Protecting minorities
Family planning
Social welfare
Democracy and the rule of law
Economic promotion
Climate and environmental protection
Women’s rights and gender equality
Humanitarian aid
Food and agriculture
Health
Education
Water, sanitary and hygiene
0 20 40 60 80 100
Percentage of respondents
Actual
Targe t
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2024. N = 2,050. The questions were: “In your opinion,
what objectives should German development cooperation contribute towards? Please select up to 5 objectives.” (“target”). “And which objectives do you think German
development cooperation is actually contributing towards? Please select up to ve objectives.” (“actual”).

Attitudes towards development policy and development cooperation over the course of time 26
In concrete terms, respondents believe that German
development cooperation should focus primarily on
improvements in the areas of “water, sanitary and hygiene”
(WASH), “education”, “health care” and “food and agriculture”
(blue bar (“target”) in Figure 10). “WASH” was selected by 56
percent of respondents, “education” by 47 percent, “health
care” by 42percent and “food and agriculture” by 41 percent.
After these objectives, which focus on basic needs, comes
“humanitarian aid” at 38 percent. Most of the other objectives –
including “women‘s rights and gender equality”, “protecting the
climate and the environment” and “promoting democracy and
the rule of law” – were selected by around 20 to 30 percent of
respondents. The DC objectives endorsed by fewest respondents
are “creating infrastructure” and “helping refugees”. Across all
the goals, no conspicuous upward or downward trends can be
seen over the course of the three survey waves.21 With respect
to these opinions, however, it should be noted that traditional
and progressive goals for DC are not mutually exclusive but
can inuence one another. For example, improvements in the
education sector could also promote gender equality.
In the subsequent question – which objectives do the
respondents think that German development cooperation
is actually contributing towards (“actual”) – approximately
the same objectives were mentioned most frequently.
Respondents most frequently mention “Humanitarian aid” as
the area to which they believe the German Federal Government
is contributing (45%), followed by WASH (44%) and health care
(35%; see Figure 10, yellow bar).
A comparison of the objectives that the German Federal
Government should pursue in the area of development
cooperation (“target”) with the objectives to which the
respondents believe it to actually contributes (“actual”)
reveals some major dierences. For example, respondents are
most likely to believe that the German Federal Government is
contributing less than it should to the objectives of “WASH”,
“education”, “food and agriculture”, “health care” and “promoting
the economy” (dierence between “target” and “actual”
greater than 5 percentage points in each case). In the case of
“humanitarian aid”, “helping refugees” and “infrastructure”, it
21 A visualisation of the objectives over the course of time can be found in Figures 7 and 8 in the online appendix.
22 A sample calculation to illustrate this: if ten people are surveyed and ve of them support DC and ve do not, DC supporters and DC opponents each have a share of 50
percent. If just one person who previously supported DC changes their attitude and no longer supports DC from now on, and one person who previously did not support DC
now supports it, these two people have changed their attitude towards DC. However, the share of 50 percent would remain unchanged.
is the other way around; here respondents are most likely to
assume that the German Federal Government is doing more
than it should (dierence between “target” and “actual” greater
than 5 percentage points in each case).
3.4 How stable are the attitudes towards
development policy and DC?
Even though attitudes towards DC have hardly changed
in Germany in the past, the results of the current Opinion
Monitor suggest that support for DC only has a limited
resilience to multiple crises. For example, the DEL data until
January 2022, which can be seen in the time series in the
previous Opinion Monitor reports, revealed very small or no
changes in the attitudes towards DC, leading to the conclusion
that these are very stable (Schneider and Gleser, 2018; Schneider
et al., 2022). However, the latest analyses in this chapter show
that the respondents’ support for DC has been decreasing
sharply since the beginning of 2022. This may indicate that
the attitudes of the German population towards development
policy are not completely stable and unchanging, especially in
the context of major global challenges and crises. Ultimately,
observations at a population level cannot determine changes
on the individual level as it is possible for shares to remain the
same while individuals change their attitudes.22
One possible explanation for the decline in support for DC
in the context of the current political challenges is that the
attitude towards DC is not particularly stable. For example,
various experimental studies have been able to show that small
changes in the explanation of development policy (framing) or
in the presentation of information can substantially change the
attitude towards DC (see, for example, Bayram and Thomson,
2022; Bayram et al., 2024; Eger et al., 2022; Schneider et al., 2022;
Scotto et al., 2017). With respect to attitudes towards development
policy, the literature also often shows that in spite of a high level
of support for DC and a reported interest in the topic, knowledge
of specic measures in the policy area is low. Therefore, attitudes
towards DC are characterised as “a mile wide and an inch deep”,
i.e. as support that is widespread, but is possibly standing on

27Attitudes towards development policy and development cooperation over the course of time
shaky ground (e.g. Smillie, 1999; see also Riddell, 2007, Chapter 7).
A study by the research institute pollytix that was commissioned
by the BMZ comes to a similar conclusion; it assesses the
German population’s level of knowledge and information with
regard to DC as low and attitudes towards DC as not very stable
(Faltas et al., 2024). These ndings suggest that support for and
interest in DC are rather supercial or might be poorly anchored
in knowledge or an actual belief and therefore are easy to change
in the case of doubt.
The question of the strength of attitudes towards DC – i.e. of
how stable and resistant to change they are – represents a gap
in the research that has a high practical relevance. How strong
or weak an attitude is allows conclusions to be drawn regarding
how changeable it is and how much it inuences our behaviour
(e.g. Krosnick and Petty, 1995; Vogel et al, 2016). Strong attitudes
have a greater impact on behaviour than weak attitudes
(e.g.Conner et al., 2022). At the same time, strong attitudes
are more dicult to change than weak attitudes (e.g. Eagly and
Chaiken, 1995). There are indications of a possible instability of
attitudes towards DC in the literature, but the actual strength
of the attitudes of the German population towards DC has not
been studied to date. The question of which population groups
have very stable or easily changeable attitudes and opinions is
particularly relevant in a polarised discourse in which political
actors want to inuence the climate of opinion. It stands to
reason that the aforementioned actors focus primarily on those
attitudes of the electorate that can be changed.
In the existing research on the strength of attitudes, a large
number of characteristics are studied that make attitudes
resilient to change and have a greater inuence on behaviour.
For example, strong attitudes are characterised by people
a)having more knowledge about the topic, b) rating the topic
as relevant or important on a personal level, c) being certain of
their own attitude, d) being able to retrieve the attitude quickly,
e) having hardly any contradictory thoughts, but also by people
f) having stronger feelings about the topic, g) feeling personally
aected by the topic or by h) attitudes being strongly anchored
in (moral) convictions (for various lists of the characteristics of
strong attitudes, see, for example, Howe and Krosnick, 2017;
Krosnick and Petty, 1995; Vogel et al., 2016).
To investigate the strength of attitudes towards DC in
Germany, a measuring tool has been developed in the context
of this study to measure these characteristics. On the basis
of the literature on attitude strength, 11 characteristics have
been selected as relevant for the attitudes towards DC. These
include perceived knowledge about DC, attitude certainty with
regard to DC, attitude ambivalence towards DC, the subjective
accessibility for the assessment of DC, the personal relevance
of DC, interest in DC, personal involvement in DC, the inuence
of beliefs and moral convictions on the attitude towards DC,
the self-assessed extremity of the attitude towards DC and the
aective intensity (see Figure 11; the precise wording of the
questions relating to the individual characteristics can be seen
in Table 3 in Section 1.1.3 in the online appendix).
Additionally, with the help of this measuring tool, overarching
factors of attitude strength were determined and the
respondents were categorised according to their attitude
strength and support for DC. This allows a nuanced picture to
be drawn of how deeply rooted the attitudes towards DC are and
what factors make the attitudes more stable or more unstable.
3.4.1 Many respondents are condent in their
attitude towards DC and nd the topic relevant;
only a few feel well informed.
An examination of the average values for the individual
characteristics shows that the respondents are relatively
certain in their attitude towards DC, perceive the topic
as personally relevant and are interested in the topic, but
consider themselves to be poorly informed and do not feel
any strong personal involvement. As Figure 11 illustrates, the
characteristics of “attitude certainty”, “personal relevance” and
“interest in the topic” show the highest average score, with
values between 4.9 and 5.2 on a scale from 1 to 7. The sense of
being personally aected (“personal involvement”) by the topic
of DC has a lower score with an average value of 3.9. This also
applies to the aective intensity with regard to DC (3.9) and the
perceived knowledge (4.1).

Attitudes towards development policy and development cooperation over the course of time 28
The 11 characteristics can be summarised in four
superordinate factors of attitude strength towards DC:
(1) perceived knowledge, (2) attitude certainty, (3) the
personal relevance of the topic and (4) the inuence
of personal beliefs on the attitude.23 These factors were
determined using an exploratory factor analysis (see Box 5).
Figure 11 shows which characteristics are assigned to which
factor. Even though there is no consensus in the scientic
literature to date on universal factors of attitude strength, the
four factors described here can also frequently be found in the
literature on attitude strength and reect the key aspects of
attitude strength (Bassili, 2008; Krosnick and Petty, 1995).24
Figure 11 Score assigned to the characteristics of attitude strength
Personal relevance of the attitude
Attitude certainty
Influence of beliefs on the attitude
Feeling informed
1
Low
234567
High
Feeling informed
Affective intensity
Extremity of the attitude
Influence of moral convictions
Influence of beliefs
Subjective accessibility*
Attitude ambivalence*
Attitude certainty
Personal involvement
Interest in the topic
Personal relevance
Average value
Characteristics of attitude strength
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly
assigned to a survey with questions on development policy in general (N = 2,041) or questions on feminist development policy (N =2,009) rst. The gure shows the
mean values with 95% condence intervals. The values only relate to the people who answered the questions about DC rst. The response options vary depending on
the question, but consist of a seven-point scale for all questions, where “1” stands for lower values and “7” for higher values. The value “4” corresponds to the centre
of the scale. With regard to the individual characteristics, higher values are generally an indication of more stable attitudes. In the case of the characteristics that
aremarked with * (“attitude ambivalence” and “subjective accessibility”), lower values are an indicator of more stable attitudes.
23 “Belief” and “attitude” are related, but distinct concepts. Attitudes are appraisals of specic objects, people, ideas or situations, which can be positive, negative or neutral (see,
for example, Maio et al., 2019, p. 4). Beliefs, on the other hand, are deeply rooted assumptions that people have about the world, themselves, other people or particular topics
and are often associated with fundamental values and identities. Attitudes are deemed to be more stable if they are anchored in profound beliefs. They can, however, also be
more supercial and exible and, for example, be based on other inuences that are less stable, for example the current mood (see, for example, Kruglanski and Stroebe, 2005).
24 Details of the results of the exploratory factor analysis can be found in Section 1.4.1 of the online appendix.

29Attitudes towards development policy and development cooperation over the course of time
Box 5 Exploratory factor analysis
The exploratory factor analysis is a statistical method that is used to identify patterns or structures in a data set (Backhaus
et al., 2018, Chapter 7). It is often applied to understand the underlying relationships between variables and to reduce the
complexity of data. Essentially, the exploratory factor analysis attempts to reduce a large number of variables to a smaller
number of factors. These factors are hypothetical constructs that cannot be directly observed but are derived from the
relationships between the variables. They can be understood as umbrella terms that are derived from the data and characterise
the data. As an exploratory factor analysis is a data-driven procedure, the identied factors may dier when it is applied to
dierent data sets.
The analysis of the four factors shows that a large proportion
of the respondents feel poorly to moderately informed
about DC and that the attitudes of the majority are only
slightly to moderately rooted in beliefs. At the same time,
a high proportion of respondents consider the topic of DC
to be very relevant and state that they are very certain
with regard to their attitudes towards DC. The shares of
respondents according to their scores on the factors are shown
in Figure 12. Inline with the scientic literature and previous
insights from the Opinion Monitor, the majority of the
population (83%) reports a low or medium level of perceived
knowledge about the topic area of DC and development
policy , while just 16 percent feel very well informed. Attitudes
towards DC also appear to be less anchored in personal beliefs.
The majority of respondents indicate that the inuence of their
beliefs on their attitude towards DC is only medium (71%)
or low (9%). However, 20 percent of respondents specify that
beliefs have a high inuence on their attitude towards DC.
Even though many respondents feel moderately to poorly
informed about DC and many do not consider their personal
beliefs to have a major inuence over their attitude towards
DC, large sections of the respondents still regard DC as
a highly relevant topic (34%) and are very certain of their
attitude (40%). Only 11 percent of respondents see the topic
as irrelevant, and 2percent are very uncertain of their attitude.
These observations indicate that many people are relatively
condent of their attitude in spite of being poorly informed
about the topic of DC. Inaddition, the topic of DC is denitely
classed as relevant and important by many respondents. In line
with the hypothesis that attitudes towards DC are “a mile wide
and an inch deep”, they are only rooted in beliefs for a relatively
small proportion of the respondents.

Attitudes towards development policy and development cooperation over the course of time 30
Figure 12 Shares of respondents according to their score on the factors of attitude strength
11
55
34
9
71
20
2
58
40
12
71
16
0
25
50
75
100
Feeling
informed
Influence of
beliefs on
the attitude
Personal
relevance
of the topic
Attitude
certainty
Percentage of respondents
High score
Medium score
Low score
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly
assigned to a survey with questions on development policy in general (N = 2,041) or questions on feminist development policy (N =2,009) rst. The values are based on
the share of respondents who answered the questions about DC rst. The gure shows the four factors into which the individual indicators have been conated on the
basis of the exploratory factor analysis, and the shares of the population that have a low (average value < 3), medium (average value 3–5) or high (average value > 5)
score on these factors.
3.4.2 Attitude types: the majority has a moderately
positive, but unstable attitude towards DC
To get a better understanding of how the various
characteristics of attitude strength relate to the support for
DC and of which attitude proles can be found in the German
population, the respondents were segmented. A latent prole
analysis (LPA; see Box 6) was used for this, considering all the
characteristics of attitude strength and the support for DC.
The aim was to determine what proportion of people have
strong or weak attitudes and whether there are dierences in the
attitude strength of advocates and opponents of DC. This sets
the current Opinion Monitor apart from the previous studies on
the topic – for example the Opinion Monitor 2018 (Schneider
and Gleser, 2018, Chapter 7) or the study commissioned by the
BMZ on attitudes towards DC (Faltas etal., 2024). In these
studies, corresponding attitude proles were created, but the
stability of the attitudes was not determined.
Box 6 Latent prole analysis (LPA)
Latent prole analysis (LPA) is a statistical method that is
used to identify groups in a sample that are similar with
regard to characteristics that are not directly observable
– so-called latent characteristics (Spurk et al., 2020;
Vermunt and Magidson, 2002). Essentially, LPA looks for
common patterns of responses, for example to questions
relating to personality characteristics, behaviour patterns
or attitudes. The identied groups or proles can then be
examined further to analyse what factors distinguish them
from one another. As LPA is a data-driven procedure, the
identied proles may vary when it is applied to dierent
data sets.

31Attitudes towards development policy and development cooperation over the course of time
The segmentation by means of an LPA identied four
attitude types: (1) “stable supporters” (19%), (2) “ambivalent
supporters” (5%), (3) “unstable moderate supporters” (58%)
and (4) “stable opponents” (18%).25 The labels of the attitude
types were selected on the basis of the two dimensions that
were relevant for the classication of these attitude types:
(1)the level of support for DC and (2) the attitude strengths.
The proportion of respondents assigned to each of the attitude
types is presented in Figure 13.
Figure 13 Proportion of respondents assigned to the four attitude types (in percent)
19
5
58
18
Stable supporters
Ambivalent supporters
Unstable moderate supporters
Stable opponents
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly
assigned to a survey with questions on DC (N = 2,041) or questions on feminist development policy (N =2,009) rst. The values are based on the share of respondents
who answered the questions about development policy rst. The gure shows the shares for the four attitude proles to which respondents were assigned by a latent
prole analysis based on their responses.
Even though a majority of respondents supports the idea of
DC in principle, only 19percent can be referred to as “stable
supporters”, whereas 58 percent support DC to a moderate
degree, but have a relatively unstable attitude. The group of
“stable supporters” is characterised by very positive attitudes
towards DC, a high attitude certainty and an attitude that is
rooted in beliefs (see Figure 14 and Figure 15). In contrast, the
“moderately unstable supporters” are distinguished by a support
that is on weak ground, as the attitudes are less pronounced
and are relatively poorly rooted in beliefs. In addition, there is
rather less interest in the topic. 5 percent of respondents can
be identied as “ambivalent supporters”. On the one hand,
thisgroup shows the greatest level of support for DC, with this
attitude appearing to be strongly inuenced by beliefs. On the
other hand, it shows a high degree of ambivalence about this
attitude, and the respondents say that they spent a long time
deliberating on their own opinion.
In spite of widespread support for DC, 18 percent of
respondents can be labelled “stable opponents”. This group
is very certain of their attitude, but nds the topic rather
irrelevant, and their attitude is less rooted in personal beliefs
(see Figure 15). This special prole suggests that the attitudes
are stable and will not change easily.
25 The LPA found that a solution with ve groups best describes the data. In a solution with ve groups, two attitude proles were identied for unstable moderate supporters,
which dier only slightly from one another both in terms of their support for DC and in terms of the score given to the various attitude characteristics. For the sake of
simplication, these two attitude proles are conated into a single group in this report. The shares and the attitude proles, taking all ve groups into account, and other
details on the results of the latent prole analysis are documented in Section 1.4.2 of the online appendix.

Attitudes towards development policy and development cooperation over the course of time 32
Figure 14 The proles of the four attitude types
Extremity of the attitude
Attitude ambivalence*
Influence of moral convictions
Influence of beliefs
Feeling informed
Interest in the topic
Personal involvement
Personal relevance
Attitude certainty
Affective intensity
Subjective accessibility*
1
Low
234567
High
Average
value
Stable
supporters:
19 %
Ambivalent
supporters:
5 %
Unstable
moderate
supporters:
58 %
Stable
opponents:
18 %
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 2,041. The gure shows mean values for the
dierent characteristic s of attitude strength, broken down into the four attitude types that were identied by a latent prole analysis. The values are based solely
on the share of respondents who answered the questions about development policy rst. The response options varied depending on the question, but consisted
of aseven-point scale for all questions, where “1” represents lower values and “7” higher values. “4” corresponds to the centre of the scale and a medium score.
Withregard to the individual characteristics, higher values are generally an indication of more stable attitudes. In the case of the characteristics that are marked
with*(“attitudeambivalence” and “subjective accessibility”), lower values are an indicator of more stable attitudes.
The interesting thing is that even the group of opponents
does not completely reject DC in the latest survey. With an
average value of 4.4 on a scale from 0 to 10, this group ranks
near the centre of the scale (see Figure 15). As the survey only
asked to what extent Germany should provide assistance in
countries in the Global South, it is unclear at this juncture what
support for specic DC/ODA expenditure, for example, would
be like in this group. Against the backdrop of multiple crises,
scanty budgetary resources and increased criticism for German
DC, the question is whether support for DC could fall further,
especially amongst opponents and unstable supporters.
Alternatively, the current results could also reect the lowest
level of support for DC. In the context of a stabilising economy
and contingent on a changing environment, support for DC
could also rise again.

33Attitudes towards development policy and development cooperation over the course of time
Figure 15 Average support for development cooperation by attitude type
Stable supporters: 19%
Ambivalent supporters: 5%
Unstable moderate supporters: 58%
Stable opponents: 18%
0
No
support
12345678910
High
support
Average
support
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly
assigned to a survey with questions on development policy in general (N = 2,041) or questions on feminist development policy (N =2,009) rst. The gure shows mean
values with 95% condence intervals for support for DC , broken down into the four attitude types that were identied by a latent prole analysis. The values are based
solely on the share of respondent s who answered the questions about development policy rst.
Stable opponents and unstable moderate supporters identify
more frequently with the AfD and have a lower level of formal
education. With regard to gender and age, opponents are more
often male, but age only plays a role for unstable moderate
supporters. An in-depth analysis examined the extent to which
various sociodemographic characteristics are related to the
assignment to the attitude types. The group of stable supporters
was chosen as a comparison group to ascertain how the other
groups dier from this group in terms of the characteristics of
gender, age, level of formal education and party preference.
This comparison showed that stable opponents are more often
male than stable supporters. There was no gender dierence
between stable and unstable supporters. Unstable supporters
were slightly younger (under 40) than stable supporters; this age
dierence is not evident between stable opponents and stable
supporters. Both unstable supporters and stable opponents are
less likely to have a high level of formal education than stable
supporters. Both groups identify more frequently with the AfD
than the SPD. No signicant dierences can be seen between
the groups with regard to the other parties.26
3.4.3 Attitude strength as an important factor for
understanding uctuations in attitudes towards DC
All in all, the examination of attitude strength has shown that
a large proportion of respondents do have a positive attitude
towards the basic concept of DC, but this attitude is not built
on a strong foundation. Many appear to be poorly informed
about DC, and their attitudes are based less on actual beliefs
than on a supercial feeling that the topic is relevant in principle.
Unstable attitudes drive action less and are easier to change,
and people with unstable attitudes are more receptive to
counterarguments. For the attitude towards DC, this means
that the increased economic bottlenecks combined with
multiple global crises and the ever louder voices of opponents
to DC presumably also have an impact on the opinion of the
German population because many people do not have a
particularly strong attitude towards DC. Hence, these attitudes
can be inuenced by external information. And since the topic
is only really central to a small proportion of the population
26 The results were determined on the basis of a multinomial logistic regression. The characteristics used are categorical variables, i.e. variables that can only take the form of a
limited number of categories. The results of this regression model indicate the probability of belonging to an attitude type compared with a comparison group (see, for example,
Backhaus et al., 2018, Chapter 5). The coecients of the explanatory (categorical) variables provide a simplied indication of how the probability changes if the respondents have
a certain characteristic (for example “male”), specically relative to a reference category (here “female”). Therefore, a hypothetical interpretation could be: “In comparison with
women, men are more likely to be stable opponents than stable supporters.” In the analysis, the reference category “female” was selected for the analysis. Age was divided into
ve groups, with the youngest age group (18-29) as the reference category. The level of formal education was separated into three categories, with the lowest level of education
as the reference category. With regard to party preference, the SPD was selected as the reference category because it is the largest parliamentary group and contains the German
Federal Chancellor after the most recent Bundestag elections (2021). The detailed results of this analysis are documented in Figure 15 in Section 1.4.2 of the online appendix.

Attitudes towards development policy and development cooperation over the course of time 34
and thus does not drive individual behaviour, positive attitudes
towards DC might not be reected in active engagement for the
majority of the population (see also Section 3.7).
Attitude strength can be an important starting point for
future debates about the uctuation of attitudes towards DC;
here, particular attention should be given to the population’s
information environment and beliefs relating to the topic of
DC. In this context, it remains unclear how information about
DC reaches the wider population and what kind of information
can change, but also stabilise their attitude towards it. For
example, the question emerges as to whether strong beliefs on
the topic of DC can arise as a result of external inuences or
ultimately (have to) arise from the people themselves.
3.5 How do the respondents rate their self-
ecacy and how eective do they consider
other development policy actors to be?
Against the backdrop of the 2030 Agenda and its 17 SDGs,
another important indicator concerning engagement with
development policy and sustainable development is the
German population’s sense of being able to achieve something
themselves. This feeling, which is referred to in psychology as
self-ecacy (see Bandura, 1982), has a considerable inuence
on development policy engagement (Schneider and Gleser,
2018; Schneider et al., 2022). Therefore, the DEL surveys ask
both about the respondents’ perceived self-ecacy in the
area of development policy and about the eectiveness that is
attributed to various development policy actors in combating
global poverty.
Self-ecacy in the area of development policy is still low in
the German population and is even on the decline. In the most
recent survey in January 2024, the average was 2.7 (ona scale
from 0 to 10; see Figure 16). This is the lowest observed value
of the whole of the period under investigation; since June 2022,
the point in time with the highest average self-ecacy (3.6), the
self-ecacy rating has been falling with slight uctuations.

35Attitudes towards development policy and development cooperation over the course of time
Figure 16 Eectiveness rating of various actors in development policy
3,3
3,1 3,2
3,2 3,0
6,2
6,1
6,0 6,0
5,6
6,1 5,9
6,0 6,1
5,7
6,8
6,6 6,7 6,7 6,4
6,9
6,7 6,7 6,7
6,3
3,4 3,2 3,3 3,4
2,7
3,0
3,2
3,6 2,8
6,3 6,2
6,0 6,0
5,3
6,2
6,0 6,2
5,2
6,2 6,0 5,9
6,1
5,7
6,0
6,1 6,3
5,2
7,0
6,8 6,8 6,8
6,2
6,5
6,7 6,9
5,9
7,0
6,8 6,7 6,6
6,2
6,6 6,7
6,8
5,7
0
1
2
3
4
5
6
7
8
9
10
9/2019
1/2020
6/2020
9/2020
1/2021
6/2021
9/2021
1/2022
6/2022
10/2022
1/2023
6/2023
10/2023
1/2024
Date of survey
Average effectiveness
International
organisations
Companies
NGOs
German
Federal Government
Me personally
Source: DEval, own visualisation. Source of the data: DEL tracker and DEL panel. The gure shows mean values. The question was: ““ How much of a dierence, if any,
do you think each of the following can make to reducing poverty in poor countries?” Response scale: from 0 = “can‘t make any dierence at all” to 10 = “can make a
great deal of dierence”. “5” represents the centre of the scale.
A much stronger inuence is ascribed to the other actors
that the survey asks about; companies and international
organisations are given the highest score with an average
ascribed eectiveness of 6.2 each on a scale from 0 to 10.
NGOs and the German Federal Government have a slightly
lower average score (5.7 and 5.3 respectively) Since January
2023, a decrease followed by a slight increase in the ascribed
eectiveness has been observed for these actors. However,
the level from before January 2023 has not been reached again.
The decline both in the respondents’ self-ecacy rating and
in the eectiveness ascribed to the various actors could again
be attributed to the political situation during the surveys as
discussed in the introduction.
3.6 How eective do respondents
rate German DC?
The estimated eectiveness of German development
cooperation has also been falling steadily in the survey
waves since January 2022 (after a slight rise in June 2022).27
As Figure 17 shows, the group that rates DC as “(absolutely)
ineective” has grown from around 19 percent in June 2022 to
32 percent in January 2024 (+13 percentage points), whereas
the group that gives the assessment “(very) eective” has fallen
from 27 to 18percent (-9 percentage points).
27 The DEL tracker of January 2022 was the last DEL survey wave that was included in the Opinion Monitor 2022 (see Schneider et al., 2022, Chapter 2).

Attitudes towards development policy and development cooperation over the course of time 36
Figure 17 Assessment of the eectiveness of development cooperation (2013–2024)
6
38
40
15
8
36
38
17
9
36
39
15
11
36
41
12
12
37
39
12
12
37
39
11
11
35
41
12
12
38
38
12
12
37
38
12
13
38
37
11
15
21
43
21
12
20
43
24
13
20
44
23
14
20
44
22
15
19
46
21
13
20
45
22
13
21
44
22
17
21
40
22
19
27
38
17
23
25
39
14
23
24
38
14
24
21
42
13
27
21
41
12
32
18
39
11
0
20
40
60
80
100
12/2013
6/2014
12/2014
6/2015
12/2015
6/2016
12/2016
7/2017
12/2017
7/2018
9/2019
1/2020
6/2020
9/2020
1/2021
6/2021
9/2021
1/2022
6/2022
10/2022
1/2023
6/2023
10/2023
1/2024
Date of survey
Percentage of respondents
(Very) effective
Medium
(Absolutely) ineffective
Don’t know
Source: DEval, own visualisation. Source of the data: AAT panel until 7/2018 and DEL tracker and DEL panel from 2019 to 2024. N AAT panel ≈ 6,000, N DEL tracker
≈ 1,000, N DEL panel ≈ 6,000. The DEL tracker surveys take place in Januar y and June, the panel surveys in September/October. Weighted data. The question was:
“In your opinion, how eective is the government’s nancial support of development cooperation on the whole? Please use a scale from 0 to 10, where 0 = ‘absolutely
ineective’ and 10 = ‘very eective’ ”. The vertical dotted line marks the transition from AAT to DEL.28
The increasing scepticism regarding the eectiveness
of DC since 2022 corresponds to the observed decrease
in support for DC. The global political situation, the
economic development in Germany and the public debate on
development policy might also play a role in this trend. Even
though the topic of “the eectiveness of DC” was already
a controversial issue and relevant to public debate in the
past, criticism of DC in winter 2023/24 also addressed its
eectiveness (e.g. Buchsteiner, 2023).
3.7 How do respondents engage
indevelopmentpolicy?
Alongside public support for development policy, development
policy actors are interested in getting the citizens to engage in
development policy objectives and organisations on a personal
level. Therefore, the BMZ has composed its own strategy for
promoting civic engagement in development policy (BMZ,
2023b). Civil society organisations, too, promote engagement
28 Due to a fundamental change in the structure of the questionnaire, a question-order eect (Oldendick, 2008; Schuman and Presser, 1996) must be assumed. The basic idea is
that preliminary questions can activate assessment criteria, which are then used when answering the subsequent question (see the discussion on question-order eects in
the Opinion Monitor for Development Policy 2022 (Schneider et al., 2022, Chapter 2)). In the present case, the question-order eect probably leads to the eectiveness of DC
being assessed more positively on average in the DEL surveys than in the AAT surveys.

37Attitudes towards development policy and development cooperation over the course of time
and rely on it in their advocacy work, their fund-raising and
frequently also in their project work. Federal states and local
communities also support (decentralised) engagement with
development policy. Accordingly, development policy actors need
a dierentiated picture of how the German population engages
in development policy and how uptake changes over time.
Figure 18 Development engagement over the course of time
20 19 18
9910
16 14 15
434
55 57 57
445
12 11 11
555
445
11 11 11
15 14 13
323
767
None
Contacted a politician
Participated in a demonstration
Voting decision
Purposively consumed
Voluntary work
Voluntary service abroad
Signed a petition
Donated
Boycotted products
Discussed with friends and family
Social media: post
Social media: interacted
1/2023 6/2023 1/2024
1/2023 6/2023 1/2024 1/2023 6/2023 1/2024
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
Date of survey
Percentage of respondents
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2023, July 2023 and January 2024. N for each survey ≈
2,000. The question was: “In the last 12 months, have you taken action to support people in developing countries?” The respondents could choose any number of forms
of engagement.

Attitudes towards development policy and development cooperation over the course of time 38
A new, more dierentiated set of questions was developed for
the Opinion Monitor 2024 to record the German population’s
development engagement. In comparison with the Opinion
Monitor for Development Policy 2022 (Schneider et al., 2022,
Chapter 3), this set of questions allows a more comprehensive
view of development engagement – especially in the area of
non-monetary engagement (e.g. activities on social media
and in development policy organisations). In addition, the
respondents are asked directly about their activities to support
people in countries in the Global South. Figure 18 presents the
results of the three available surveys to date.
The majority of respondents say that they have not engaged in
development policy in the last 12 months. This share is 55 to 57
percent in the three available survey waves. This is consistent with
the analyses in the Opinion Monitor 2022 (Schneider et al., 2022,
Chapter 3.4) and is in line with the nding from Section 3.4 of this
report that a large proportion of the respondents have unstable
attitudes, which are consequently not reected in concrete
behaviour and development engagement.
Discussing topics from the area of development policy,
DC and global poverty with friends and family is the most
frequently mentioned form of engagement (18 to 20% of
respondents). Other common forms of engagement are
donations, consumption-related activities (consumption
and boycotting decisions) and activities on social media.
Consumption decisions were mentioned by 13 to 16 percent
of respondents, donations and activities on social media by 11
to 12 percent. More complicated forms of engagement such
as volunteering at development policy organisations and
volunteer work abroad hardly received any mentions.29 The
minor changes across the three survey waves suggest that there
is no relationship between the political developments of 2023
and the use of individual forms of development engagement.
3.8 Conclusion: German development policy
isat a crossroads
The results of this chapter show that German public opinion
regarding development policy is changing since the beginning
of 2022. This is reected above all in the decline in general
support for DC in the German population and the dwindling
public support for the current development cooperation/ODA
expenditure of the German Federal Government. The growing
doubts about the eectiveness of development cooperation
and a decline in support for a moral obligation towards the
countries of the Global South also suggest that the population
is becoming more reticent with respect to Germany’s
development engagement. Other surveys and studies also
show such a change in public opinion.30 These changes come at
a time when German foreign and development policy is facing
major challenges (e.g. Ukraine, the situation in the Middle
East, global tensions between the West on the one hand and
China and Russia, but also states in the Global South on the
other), while the government is also under pressure because
of domestic policy developments. In this context, development
policy is also receiving greater attention, as the analysis on the
platform Twitter/X shows. Amid these challenges, the German
Federal Government announced that it will commit to a feminist
foreign and development policy in the future.
Against this backdrop, the previous broad consensus in the
German population in favour of development cooperation
appears to be eroding, although it still exists. The data
indicates that the population is increasingly questioning where
tax revenues are going in light of a strained budgetary situation
and nancial burdens for private households as well as public
criticism for specic DC/ODA expenditure – for example the
funding of interventions in Afghanistan or Gaza as well as the
discussion in the media about interventions relating to climate
29 The lower shares on average in comparison with the DEL data (see Schneider et al., 2022, Section 3.1) might be partly because the question used here asked directly about
supporting people in countries in the Global South, whereas the DEL questionnaire talks more generally about poverty and development in the world.
30 For example, the Eurobarometer survey in May/June 2023 for Germany indicated that sections of the population are seeing it less and less as the task of Germany and the
EU to combat poverty in countries in the Global South, and that the EU’s eectiveness in promoting sustainable development in the Global South is perceived as being lower
(Eurobarometer, 2023). The two surveys in the study that the research institute pollytix conducted for the BMZ in July and November 2023 also showed that public support for
the current development cooperation/ODA expenditure is declining, and that the weighting given to national interests is increasing within the population (Faltas et al., 2024).

39Attitudes towards development policy and development cooperation over the course of time
change adaptation in Peru. So far, DC is still supported, even
though attitudes towards it can be classied as unstable in large
parts of the German population and the support is consequently
on shaky ground. Therefore, whether DC takes place is not
(yet) in question, but what, how and where it takes place are
(see Faust, 2024). This is evident, for example, in the dwindling
persuasiveness of motives for DC, which aect the narrative
of a “DC to overcome global challenges”. In conjunction with
the population’s scepticism regarding the achievement of the
objectives of the 2030 Agenda and the possibility of being able
to make a personal contribution to these (see Schneider et al.,
2024), this should be seen as a warning signal for development
policy actors. The implications in Chapter 5 go into more detail
about what dealing with these challenges could look like.

4.
DEVELOPMENT POLICY
ATTITUDES IN THE CONTEXT
OFSECURITY POLICY
CHALLENGES AND FEMINIST
DEVELOPMENT POLICY

41Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
4.1 Overview of the chapter
The complex challenges of our time also make it necessary to
understand the attitudes of the German population towards
development policy in the context of foreign and security
policy and against the backdrop of the guiding principle of a
feminist development policy. Therefore, not only do general
attitudes towards development policy have to be continuously
recorded (see Chapter 3); they also have to be considered in
more detail in the context of (1) public perceptions and attitudes
towards foreign and security policy in general, (2) the Russian
war of aggression against Ukraine and other global crises and
conicts, and (3) the strategic guiding principle of feminist
development policy. Against this backdrop, the subsequent
in-depth section examines the following questions:
What overarching attitudes towards foreign and security policy
does the German population have, and how do these relate to
specic attitudes towards development policy?
Section 4.2 begins by addressing the question of what the
German population’s general attitudes towards foreign and
security policy are, how these attitudes dier according to
political orientation and what the relationship is between these
attitudes and attitudes towards DC. These analyses give a better
understand-ing of what approach the population supports in
international politics, irrespective of current issues, and what role
development policy or DC plays in this. Together with the results
of the chapter on monitoring (see Chapter 3), these analyses serve
as a backdrop for contextualizing the subsequent investigations.
How do attitudes towards development policy behave in the
context of the war against Ukraine and other international
crises and economic developments?
In Section 4.3, the focus is on attitudes towards development
policy in the context of the Russian war of aggression against
Ukraine and general challenges of security policy. These also
inuence development poli-cy. Conversely, development policy
must also be considered in a comprehensive security policy
(seeFederal Foreign Oce, 2023b). In both cases, the attitudes
of the German population play an important role. In a rst step, a
descriptive study to ascertain the German population’s attitude
towards various supporting measures of Ukraine and towards
measures in the context of general security policy challenges
is conducted. In a second step, the impact that the perception
of the economic situation and information about partner
countries have on support for DC are experimentally examined.
Theassumption is that support for DC declines as the population’s
concerns about their own personal economic situation and the
national economic situation increase. Finally, in a third step,
aconjoint experiment is used to analyse the specic preferences
of the citizens with regard to partner countries, taking national
self-interest and geopolitical factors into account.
How is a feminist development policy perceived and supported,
even in the context of global challenges?
Section 4.4 addresses the opinions and attitudes towards feminist
development policy. Since the “trac light coalition” consisting
of the SPD, the Greens and the FDP took oce in December 2021,
German development policy has followed this guiding principle
and aims to strengthen the rights, representation and resources
of women, girls and marginalised population groups in the
countries of the Global South through development cooperation
(BMZ, 2023a). With the term “feminism” and demands for
equal rights and equal treatment, feminist development policy
provides points of attack for (right-wing) populist actors (for the
relationship between populism and feminism, see, for example,
Abi-Hassan, 2017; Kantola and Lombardo, 2019). Therefore,
theattitudes of the German population are an important aspect
here too. Specically, the following questions are studied: what
the population understands by feminist development policy,
its contents and objectives, whether they support these, how
stable the attitudes towards feminist development policy are,
and to what extent the perception of and support for feminist
development policy and its contents are changing in the context
of the current global challenges.
Taken together, the results of this in-depth section are
supposed to help development policy actors to communicate
with the population in an evidence-based way in times of
domestic and foreign policy challenges, with regard to security
policy challenges and the guiding principle of a feminist
development policy, and to critically reect on their own
strategies in light of the attitudes of the German population.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy42
4.2 Attitudes towards development policy in the context of attitudes towards foreign and security policy
At a
glance
Four attitude dimensions with regard to foreign and security policy
Correlations with attitudes towards development policy/development cooperation
4.9
4.5
4.2
3.5
OPINION MONITOR FOR
DEVELOPMENT POLICY 2024
Attitudes towards development policy in the context
of attitudes towards foreign and security policy
01234567
Average support for attitude dimensions
Advocacy of …
cooperative
internationalism support for
increasing DC/ODA expenditure
specic DC measures
measures to support Ukraine
feminist development policy
support for
a rejection of
global justice
isolationism
correlates with:
Average value
Cooperative internationalism
Isolationism
Militant internationalism
Global justice
Collaborating in international institutions,
advocating diplomacy and striving
for a consensus
Keeping out of international aairs
Military strength and deterrent
Solidarity with the Global South
Cooperative internationalism and global justice are, on average,
advocated most strongly by supporters of Alliance 90/The Greens
and least strongly by supporters of the AfD
Militant internationalism is, on average, most popular
with supporters of the CDU/CSU and Alliance 90/The Greens
and least popular with supporters of The Left
An isolationist approach in foreign policy is, on average,
endorsed most strongly by supporters of the AfD
and least strongly by Alliance 90/The Greens
Dierences between supporters of dierent parties

43Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Especially in times of multiple global crises, the various areas
of foreign and security policy – development policy, defence
and security policy, trade policy and diplomacy – must
be organized coherently, and public opinion must also be
examined across the dierent areas. It is not sucient to look
at public opinion regarding development policy in isolation.
On the contrary, a more comprehensive picture of attitudes
towards foreign and security policy, and the positioning of
attitudes towards development policy within this picture are
helpful for development policy actors.
An examination of the general attitudes towards foreign
and security policy allows to draw conclusions about
which specic foreign policy measures are endorsed by the
population or sections of the population and which are not.
As a result, political tensions can be detected and the room
for manoeuvre dened. This particularly applies in view of the
assumption that general attitudes towards foreign policy are
more stable than attitudes towards daily political decisions
(e.g. Mader andSchoen, 2023).
Therefore, the in-depth section begins by focusing on general
attitudes towards German foreign and security policy. The key
questions for the following analyses are:
1. What are the general attitudes of the German population
towards foreign and security policy?
2. What dierences can be found in the German population’s
attitudes towards foreign security policy along the political
spectrum?
3. What relationships exist between attitudes towards
foreignand security policy and attitudes towards
development policy?
To draw a detailed picture of attitudes towards foreign policy,
the model created by Gravelle et al. (2017) was used during
the data collection. The authors of this model condense the
existing academic literature on attitudes towards foreign and
security policy (e.g. Holsti and Rosenau, 1990; Hurwitz and
Peey, 1987; Wittkopf, 1986) and supplement the traditional
areas of “international cooperation” and “military might” with
isolationist attitudes and questions of global justice, solidarity
and redistribution. Specically, the model consists of the four
dimensions “cooperative internationalism”, “isolationism”,
“militant internationalism” and “global justice”. As every dimension
is captured over several items, the model makes it possible to
ascertain attitudes more precisely than would be the case if they
were captured by single items (see, for example, Ansolabehere
et al., 2008).31 In terms of their content, the four dimensions
are characterised as follows (the exact phrasing of the items is
documented in Table 15 in Section 2.1.1 of the online appendix):
1. Cooperative internationalism represents support for
international cooperation (especially in international
organisations such as the United Nations), diplomacy
and striving for consensus and peace (example item:
“Germany should work more through international
organisations like the United Nations”).
2. Isolationism encompasses the desire for one’s own state
tokeep out of international aairs – especially when
thereis a fear that involvement will lead to disadvantages
for the population at home (exampleitem: “Germany
needs to simply mind its own business when it comes
tointernational aairs.”).
3. Militant internationalism describes the premise that
the state should possess or maintain military might and
should take military action, including abroad, in the event
of conicts (example item: “Germany needs a strong
military to be eective in international relations”).
4. Global justice addresses how strong the population’s
support is for the global redistribution of nancial
resources to combat poverty. It has points of
contact with the concept of global solidarity and
corresponds to general support for development
cooperation (example item: “Germany should
spend more money on development aid”).
31 The rst step in calculating the attitude dimensions was to use an exploratory structural equation model (ESEM) to check whether the attitude structure corresponded to the
results found by Gravelle et al. (2017). It did. In order to then calculate the aggregate index, the associated items were then added to a dimension and scaled to a value range
from 1 to 7. Higher values mean stronger support for the respective attitude dimension. The analysis is documented in Section 2.1.2 of the online appendix.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy44
4.2.1 Respondents showed most support forthe
concepts of cooperative internationalism and iso-
lationism and least for the concept of global justice
In the full sample, cooperative internationalism receives
the greatest support from respondents. The average value
of around 4.9 on a scale from 1 to 7 is clearly above the
scale centre of 4. Consequently, international cooperation
is also advocated in times of global crises. This high level
of support corresponds to Germany’s self-image as a civil
power that dedicates itself to a rule-based international
order and multilateralism and is cautious about using military
resources (see, for example, Maull, 2015). Slight dierences
can be seen along party identication lines (Figure 19);
overall, however, sympathisers of all parties as well as people
without a preferred party advocate acooperative approach
in international politics. Only supporters of the AfD deviate
substantially from this pattern and express a neutral attitude.
With an average value of 4.5, isolationism also meets with
ahigh level of support from respondents. This nding points
towards a possible area of tension between public opinion and
outward-looking measures of the German Federal Government –
for example the feminist orientation of foreign and development
policy (Federal Foreign Oce, 2023a; BMZ, 2023a). Supporters
of the AfD most strongly endorse an isolationist course in
foreign and security policy; the lowest level of support can be
found among supporters of Alliance 90/The Greens. Supporters
of the other parties range between those two poles.
Respondents also have a slightly positive attitude towards
militant internationalism, with an average value of 4.2.
This may come as a surprise in view of Germany’s caution
with regard to military measures, but could also be related to
the altered security situation as a result of the Russian war of
aggression against Ukraine.32 Strikingly, besides sympathisers
of the CDU/CSU, it is supporters of the Greens who most
strongly endorse this approach to foreign and security policy.
This is surprising because the party is closely associated
with the peace movement (e.g. Mende, 2011). However, the
dierences between the Green party, the other parties in
the trac light coalition and the CDU/CSU are small. This is
probably because of the consensus between these parties
with respect to supporting Ukraine and the importance of the
capacity for military action that has arisen from the Russian
war of aggression. The lowest level of support is expressed by
sympathisers of The Left, a party that is generally critical of
military action.
Global justice – the dimension that conceptually roughly
corresponds to endorsement of DC – receives the lowest level
of support from the population, with an average value of 3.5.
Here, the average value is clearly below the centre of the scale
(4). This is indicative of a rather negative position across all
respondents. For development policy actors, this nding means
that their policy area has a hard time competing with other
areas of action in foreign and security policy in public opinion.
However, the dierences along the lines of political preferences
that are familiar from analyses of support for DC can also be
seen here (e.g. Schneider and Gleser, 2018; Schneider et al.,
2022): a high level of support among supporters of Alliance 90/
The Greens and The Left, an average level of support among
supporters of the SPD, lower support among those of the CDU/
CSU and FDP and very little support among AfD-sympathisers
(Figure 19, graph below).
32 However, Mader and Schoen (2023) report that general foreign policy orientations were relatively stable before and after the Ukraine war began.

45Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 19 Attitude towards foreign and security policy by party identication
Global justice (M = 3.5)
Militant internationalism (M = 4.2)
Isolationism (M = 4.5)
Cooperative internationalism (M = 4.9)
1
Do not
agree at
all
234567
Completely
agree
AfD
No party
FDP
CDU/CSU
SPD
The Left
Alliance 90/The Greens
Alliance 90/The Greens
SPD
The Left
FDP
CDU/CSU
No party
AfD
The Left
No party
AfD
FDP
SPD
Alliance 90/The Greens
CDU/CSU
AfD
No party
CDU/CSU
FDP
SPD
The Left
Alliance 90/The Greens
Average support
AfD
Alliance 90/The Greens
CDU/CSU
The Left
FDP
No party
SPD
Source: DEval, own visualisation. Source of the data: DEval survey carried out by Respondi/Bilendi in June/July 2023 (N = 2,050). M = average support of the full
sample. The gure shows the mean values with 95% condence inter vals. The vertical dotted line represents the centre point of the calculated approval index and
expresses a neutral attitude towards the respective attitude dimension. For the background to the attitude dimensions, see Gravelle et al. (2017). The items and
analyses used for the dimensions are documented in Table 15 and 16 of Section 2.1 of the online appendix.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy46
4.2.2 Respondents who advocate cooperative
and militant internationalism also support various
development interventions more strongly
General attitudes towards foreign and security policy allow
to make predictions about more specic attitudes towards
development policy/DC. Firstly, there are – as Figure 20 shows
– medium to strong positive correlations between advocacy
of global justice and support for DC and humanitarian aid
measures, support for the funding of economic and social
development and the development of the rule of law in
Ukraine, and support for a feminist development policy. To put
it a dierent way: respondents who advocate global justice
and redistribution also declare themselves in favour of specic
DC measures. The same pattern also applies to cooperative
internationalism – the greater the support for this general
dimension of foreign and security policy, the greater the support
for the various specic development measures. If, however,
respondents advocate an isolationist approach, they also show
less support for the various development measures. There is a
weak positive relationship between militant internationalism
and the attitude towards DC. Therefore, advocating military
might and supporting development interventions are not
mutually exclusive, but are not necessarily linked.33
To sum up, the results show that large sections of the
respondents across almost the entire political spectrum
advocate international cooperation (i.e. cooperative
internationalism) and military might (militant interna-
tionalism); the simultaneously high level of support for an
isolationist foreign policy, the dierences in the support
for the dimension of global justice and the clearly divergent
attitude among AfD sympathisers illustrate current and
possible future areas of tension.
The empirical ndings reect the political positions in more
recent debates about foreign, security and development
policy. These include the discussion about Germany’s support
for Ukraine since February 2022 and the controversial public
discussion of development policy in winter 2023/2024.
Thelatter took place not only between the government and the
opposition, but also within the governing coalition. On the one
hand, for German foreign, security and development policy, the
ndings relating to cooperative and militant internationalism
imply a general level of public support, for instance with regard
to Ukraine policy, the role of the armed forces and cooperation
with international organisations. On the other hand, the
data relating to development policy in general and also with
respect to specic development interventions points towards
a greater spread of attitudes in the German population, and
not just between sympathisers of the governing parties and
those of the opposition parties, but also between supporters
of the three coalition parties. Adding to this is the high level of
support for an isolationist foreign policy across almost the entire
political spectrum and the special position of AfD supporters;
the latter are particularly in favour of an isolationist approach
to foreign and security policy and are also the most reticent in
the dimensions of “cooperative internationalism” and “global
justice”. As international conicts are unlikely to calm down in
the foreseeable future and foreign, security and development
policy topics require interdepartmental handling and are likely
to continue to experience a high level of public attention, such
tensions could be exacerbated in the future. This particularly
applies in view of the numerous elections that are coming up in
2024 and 2025 and the associated requirement for the parties to
convey their political positions to the electorate.34
33 Further correlation analyses show that a high level of support for cooperative internationalism goes hand in hand with a high level of support for global justice (Pearson’s r = 0.48) and
militant internationalism (Pearson’s r = 0.32). This illustrates the close relationship between areas of action in foreign policy in the eyes of the population. Only isolationism deviates
from this pattern. Respondents who advocate Germany keeping out of international aairs and only watching out for its own interests tend to reject global justice (r = -0.46). However,
there is no correlation with cooperative internationalism and militant internationalism (r = 0.08 or 0.11). This illustrates the special position occupied by isolationist attitudes.
34 At this point, it is important to bear in mind that the data was collected both before the attack on Israel by Hamas in October 2023 and before the Federal Constitutional Court
ruling on the federal budget and the farmers’ protests in January 2024. This might have had an eect on various attitudes (see also the changes in attitudes in Chapter 3).

47Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 20 Relationship between attitudes towards foreign and security policy
and attitudes towards development cooperation
Source: DEval, own visualisation. Source of the data: DEval survey carried out by Respondi/Bilendi in June/July 2023. N = 2,050 . The gure shows Pearson’s correlation
coecient (r). A value of 1 means a perfect positive relationship; a value of -1 a perfect negative relationship. The value 0 means that there is no relationship.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy48
4.3 Attitudes towards development cooperation in the context of security policy and economic challenges
Concerns about the national
economic situation
Concerns about the
respondent’s personal
economic situation
less support
for DC
Information about
economic bottlenecks
At a
glance
Economic concerns as an important factor for decreasing DC-support
Support for engagement
in emergency situations
Relationship between the economic situation
and support for development cooperation
Support for development cooperation in the face of military conicts
Preferences for partner countries of German development cooperation
greater sense of being aected
by the situation in Ukraine
perceived security threat as a result
of the situation in Ukraine
greater perceived similarity
with the people in Ukraine
Support for engagement
in emergency situations
is high
Country
characteristics
Geopolitical factors
Close partner
of China
Support for war
of aggression
against Ukraine
Energy supplier
Needs-oriented properties
Vulnerability to
the consequences
of climate change
Level of poverty
… war in
Ukraine (near)
Information
about …
… war in Yemen
(far away)
lower support
for DC
greater support
for DC
Stronger for Ukraine
than for other
countries
Ukraine: 14 percentage
points more than in 2020
Moldova: 12 percentage
points more than in 2020
 
 
 
 
no Data
Greatest support for countries in sub-Saharan Africa
Lowest support for China and countries
on the Arabian peninsula
OPINION MONITOR FOR
DEVELOPMENT POLICY 2024
Attitudes towards development cooperation in
the context of security policy and economic challenges

49Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Germany and the world are facing a large number of foreign
and security policy crises, and this poses major challenges
for DC. Around the world, more than 200 armed conicts and
wars are in progress, the most prominent examples of which
are Russia’s war of aggression against Ukraine and the conict
in Israel and Gaza (HIIK, 2023). There are also numerous
geopolitical tensions, for example between the USA and China.
Other regional powers with divergent interests, such as Saudi
Arabia, Qatar, Iran or Turkey, as well as other countries in the
Global South, have also gradually played a more condent
role in international politics. In addition, the trend towards
autocratisation, which has been observed for years now, is
continuing. For example, more than 70 percent of the world’s
population lived in autocracies in 2023 – that is 21 percentage
points more than in 2013 (Nord et al., 2024). All these tensions
and conicts make it more dicult to overcome global
challenges such as combating climate change, containing
global pandemics or achieving the SDGs (United Nations,
2023a). German DC and its partners are facing the challenge of
navigating and functioning in the context of international crises
and the associated costs (Blumenau, 2022; Schulze, 2023).
The Russian attack on Ukraine in February 2022 was a watershed
moment for Germany, which is reected both in politics and in
public opinion. For the rst time in the recent past, a country
in Europe was attacked by another country. The German Federal
Government responded, for example, with a special fund of 100
billion euros for the Bundeswehr, which means that Germany is
spending at least 2 percent of its gross domestic product (GDP)
on defence for the rst time since the end of the Cold War
(SIPRI, 2024). In this context, Federal Chancellor Olaf Scholz
coined the term “Zeitenwende” (watershed era) (German Federal
Government, 2022). This watershed is also noticeable in German
public perception. Therefore, increased defence expenditure and
deliveries of arms are receiving more popular support than they
have for a long time (Graf, 2022, 2024). The German population
regards the war against Ukraine as by far the greatest foreign
policy challenge facing Germany (Körber Foundation, 2023), and
wars, conicts and terrorism are (among) the biggest concern for
the German population, even when compared to domestic policy
challenges (Morini, 2023a, 2023b, 2023c).
Development policy cannot be looked at and conceived
separately from foreign and security policy. The challenges
described above are so multifaceted and complex that an
integrated policy approach is needed to overcome them. For
example, the understanding of the concept of security has long
been more dierentiated than previously, and development
cooperation is regarded as an integrated part of security policy,
which is reected in the National Security Strategy published
by the German Federal Government in 2023, among other
things (Federal Foreign Oce, 2023b; Leininger and Hornidge,
2024; Schulze et al., 2023, 2024).
DC is at risk of restrictions caused by the current economic
and nancial situation. The repercussions of the Covid-19
pandemic, the interruption of supply chains caused by the war
against Ukraine and the sanctions policy resulting from the
war have been a persistent burden on the global economy in
recent years. Like other export-oriented countries, Germany is
particularly aected by this (ifo, 2022), and this is noticeable
in the strain on the government budget and in nancial losses
for many citizens. At the beginning of 2024, in the wake of
the budget freeze and protests by large numbers of farmers,
it became clear that DC is coming under pressure and being
called into question as a result of conicts over the distribution
of nancial resources.
Against this backdrop, the following four questions are
examined in this section:
1. What kind of German engagement in international
crises and emergency situations does the population
support in general and specically for Ukraine?
2. Is there a relationship between the assessment of
the economic situation and support for DC?
3. Are attitudes towards DC resilient in the face of economic
challenges and nancial burdens and the consequences
of the war against Ukraine and other current wars?
4. What kinds of countries does the German population
prefer as partner countries for DC, and are strategic
and geopolitical factors important here?

4.3.1 How should Germany engage in
international crises and emergency situations?
Germany is involved in many countries and a lot of dierent
emergency situations around the world. As part of an
integrated security concept, DC is an important element
of this. As one of the biggest OECD donor states, Germany
is active in 65 countries as a DC partner (BMZ, 2020; OECD,
2023). Furthermore, Germany provides humanitarian aid to
numerous countries facing acute emergency situations such
as earthquakes, tsunamis or oods. The country’s military is
also active on the world stage and is currently participating in
18 UN, NATO and EU operations abroad (Bundeswehr, 2024).
Where and in what form an engagement is possible must be
carefully weighed up against the backdrop of the numerous
and complex emergency situations.
Respondents support Germany’s international engagement
rst and foremost in the case of natural disasters,
health crises and humanitarian crises. In the event of
economic crises and armed conicts, terrorism and wars, an
engagement is endorsed and rejected by equal shares of the
population. When asked about dierent kinds of emergency
situations in which Germany should get involved on an
international level, the vast majority of respondents (82%)
advocates an engagement in the case of natural disasters
such as earthquakes, oods or hurricanes (see Figure 21).
An engagement in health crises (73%), humanitarian crises
(71%) and environmental crises (63%) is also endorsed by the
majority. Only in the category of “armed conicts, terrorism
and wars” is German engagement rejected on average more
than it is endorsed (40% versus 36%). This pattern suggests
that the respondents support engagement in human-made
crises such as armed conicts and economic crises less than in
emergency situations that are apparently inuenced by force
majeure (forthe underlying mechanisms, see, for example,
Zageaetal., 2011).35
Figure 21 Support for Germany’s engagement in dierent crisis and emergency situations
8882 2
402036 4
131273 2
332242 4
191663 2
141271 3
Armed conflicts,
terrorism and wars
Economic crises (e.g. financial crises,
inflation or unemployment)
Environmental crises (e.g. climate
change or pollution)
Humanitarian crises (e.g. famines
or mass migration)
Health crises (e.g. pandemics
or epidemics)
Natural disasters (e.g. earthquakes,
floods or hurricanes)
0 20 40 60 80 100
Percentage of respondents
Endorse Neither endorse nor don’t endorse Don't endorse Don’t know
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in July 2023. N = 2,050. The dependent variable is the question
“In each case, please indicate to what extent you endorse Germany providing aid in the following crisis and emergency situations .” The responses “rather endorse”,
“endorse” and “completely endorse” have been conated into the category “endorse” and the responses “rather don’t endorse”, “don’t endorse” and “don’t endorse
atall” into the categor y “don’t endorse”. The full wording of the question can be found in Table 15 in Section 2.1.1 of the online appendix.
35 This should not ignore the fact that the number and intensity of particular natural disasters and other emergency situations that apparently stem from force majeure may be
a direct consequence of human-made climate change.

51Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Support for Ukraine and other countries
facingacute crises
General support for Ukraine is high and, particularly
concerning the topics “refugees” and “arms deliveries”, it
is higher than for other crisis countries. When asked about
dierent types of support for Ukraine, respondents were
particularly in favour of humanitarian aid (with an average value
of 5.6 on a scale from 1 to 7; see Figure 22 and also the results in
Section 3.1.4). Humanitarian aid also receives the highest level
of support in the question about acute crises in other countries
(average value 5.5). On average, respondents endorse nancial
assistance with reconstruction and the funding of economic and
social development and the development of the rule of law –
both DC-related topics – to the same extent for Ukraine as for
countries in the Global South that are experiencing acute crises.
A dierence can be seen in the reception of refugees: here the
respondents’ support is signicantly higher, with an average
value of 4.8, for refugees from Ukraine than for refugees from
other countries faced with acute crises (average value 4.3).
The respondents’ support for assistance for refugees who have
ed to countries other than Germany is also slightly higher
for people from Ukraine than for people from other countries
(see,for example, de Coninck, 2023).
Respondents endorse arms deliveries to and military
assistance for Ukraine more than for other partner countries
in acute crises. For Ukraine, the average value given by
respondents is right in the centre of the scale (4). Therefore, this
kind of support is, on average, neither endorsed nor rejected.
Overall, however, attitudes are spread more widely here than
is the case with other kinds of support, which also reects the
polarised debate on this issue. For other countries in the Global
South experiencing acute crises, the average support for arms
deliveries and military assistance has a value of 3.2. This means
that respondents, on average, tend to reject arms deliveries to
and military assistance for such countries.
Respondents are cautious about admitting Ukraine into
the EU. With an average value of 4.2, the assessment is only
slightly positive. The respondents show more support for
eorts to integrate other partner countries of German DC into
international organisations, with an average value of 4.6.
Greater support for development cooperation goes hand in
hand with greater support for Ukraine and other countries
in crisis. The results of multiple regression analyses show that
all seven recorded kinds of support have a positive correlation
with general support for development cooperation – both for
Ukraine and for other partner countries of German DC faced
with acute crises. This means that respondents who support
DC are also more likely, on average, to endorse support for
countries in crisis.36 This relationship is obvious for types of
support that are based in the area of DC or humanitarian
aid. However, it is also evident with regard to the reception
of refugees as well as arms deliveries and military support.
To this extent, these ndings are consistent with the German
population’s attitudes towards foreign and security policy
(see Section 4.2) inasmuch as a strong link between the
attitudes towards international cooperation, global justice
and military might can also be observed there. A simultaneous
endorsement of DC, diplomacy and multilateralism, but also
military might, is therefore not necessarily mutually exclusive.
36 See Table 23 and Table 24 in Section 2.2.2 of the online appendix for detailed results.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy52
Figure 22 Approval of support for Ukraine and other countries experiencing acute crises
Efforts for admission and integration into the EU/international organisations
Arms deliveries and military assistance
Funding of economic and social development and development of the rule of law
Financial assistance with reconstruction
Assistance for refugees who have fled to countries other than Germany
Reception of refugees
Humanitarian aid (short-term aid in emergency situations)
1
Do not
agree at
all
234567
Completely
agree
Crises in general
Ukraine
Crises in general
Ukraine
Crises in general
Ukraine
Crises in general
Ukraine
Crises in general
Ukraine
Crises in general
Ukraine
Crises in general
Ukraine
Average support
Crises in general Ukraine
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in July 2023. N = 2,050. The gure shows the mean values with 95%
condence intervals. The variables shown are the questions “In connection with the war against Ukraine, Germany is providing various kinds of support. Below, we will
present some of the possible kinds of support for Ukraine. In each case, please indicate the extent to which you agree with these” (Ukraine) and “In the context of its
international engagement, Germany is providing various kinds of suppor t not only in Ukraine, but also in developing countries experiencing acute crises. Below, we will
present some of the possible kinds of support for developing countries experiencing acute crises. In each case, please indicate the extent to which you agree with these.”
(Crises in general). The full wording of the question can be found in Table 15 in Section 2.1.1 of the online appendix. The value “4” corresponds to the centre of the scale.

53Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
4.3.2 Economic crises and the German
population’s support for DC
The Russian war of aggression against Ukraine is having
anegative impact both on the German state budget and on
the nancial situation of the citizens. While the state spends
money on military, humanitarian and development measures,
the citizens are aected by rising prices, for example for food
and energy (Destatis, 2024).
This strained nancial situation could lead to the population
scrutinising government spending more closely and thus also
to a decline in support for expenditure on DC. This is associated
with real or perceived distribution conicts, which may arise
or be exacerbated in such crisis situations. From a (socio-)
psychological perspective, thinking in terms of ingroups and
outgroups is important in this context. This is a way of thinking
that, in the present case, views the population of Germany
(ingroup) in contrast with people in other countries (outgroup)
(Mullen et al., 1992; Tajfel and Turner, 1979). The ingroup is
preferred to the outgroup, especially in conict situations or
when resources are scarce – in other words: an aid situation
becomes a competitive situation (see the so-called “intergroup
threat theory”; Stephan et al., 2015). In line with this assumption,
research on the European economic and nancial crisis from
2010 onwards has shown, for example, that the population of
the EU tends to advocate supporting the population of their own
country in times of crisis, followed by the population of other EU
states (Gerhards et al., 2020, p. 150-151). It is conceivable that,
in similar crisis situations, the focus is placed more strongly on
the interests of the ingroup, whereas the needs of people who
are further away geographically – such as people in the Global
South – receive less attention. DC as a policy area that is rather
distant from everyday life could be particularly aected in this
context, especially because the population does not recognise
the economic benets of DC (on support of DC in economic
crises see Heinrich et al., 2016).
The analyses in Subsection 3.1 already provide empirical
evidence that the German population’s support for DC has
fallen since the war against Ukraine began in February 2022.
Against this backdrop, the following section explores the
question of the extent to which the respondents’ perception of
their own personal nancial situation and the national nancial
situation correlates with the endorsement of DC.
Assessment of one’s own personal economic
situation and the national economic situation
In order to determine whether a changed economic situation
has an impact on support for DC over the course of time, it is
necessary to examine whether there is a relationship between
the assessment of the economic situation and support for DC.
An initial analysis, using cross-sectional data collected by the
DEL in October 2022, has already revealed that support for DC
is lower, the more negative the assessment of the development
of the respondent’s own personal economic situation and
the national economic situation (Zille et al., 2023). Such a
data basis makes it possible to determine the relationship
between perception of the economic situation and support
for DC. However, it does not show the impact that changes
in this perception on the part of citizens have on support for
DC. Therefore, the ve survey waves of the DEL panel from
2019 to 2023 are consulted and the hypothesis is examined
that a more negative assessment of the future development
of the respondent’s own personal economic situation and the
national economic situation relates to less support for DC.37
This ties in with the question of whether such a relationship
has strengthened since the beginning of the Russian war of
aggression.
37 The DEL questionnaire asks about the “general economic situation in your country” and the “nancial situation in your household”. For reasons of clarity, “economic situation”
is always used in the text.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy54
Figure 23 Assessment of the economic situation in Germany and of the respondent’s own household (2019–2023)
8
40
41
11
16
27
48
9
24
33
33
10
8
18
66
8
12
30
50
8
15
51
25
9
13
52
27
8
14
49
29
7
10
26
57
7
13
40
40
7
Assessment of the economic situation in own private household
Assessment of the economic situation in Germany
9/2019 9/2020 9/2021 10/2022 10/2023
0
20
40
60
80
100
0
20
40
60
80
100
Date of survey
Percentage of respondents
Get better
Stay the same
Get worse
Don’t know
Source: DEval, own visualisation. Source of the data: DEL panel wave 1 to 5; N per wave ≈ 6,000, weighted data. Wording of the questions: “How do you think the
general economic situation will develop in your country in the next 12 months?” and “To what extent do you think that the nancial situation in your household will
change within the next 12 months?” The response categories were: “get much better”, “get slightly better”, “stay the same”, “get slightly worse”, “get much worse”,
“don’t know”. The response categories have been conated to “get better” and “get worse” for the graph.

55Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
The respondents have a much more negative view of their
future personal economic situation and the future national
economic situation since the Russian war of aggression began
than they did before the war. This is illustrated by the analyses
of the two panel waves in 2022 and 2023 (see Figure 23).
In October 2023, 50 percent of respondents assumed that
the economic situation in Germany would get worse. With
regard to the private economic situation, this was 40 percent.
Only12 or 13 percent of respondents believed that their personal
economic situation or the national economic situation would
get better. However, these assessments are more positive than
in 2022. Interestingly, the years of the coronavirus pandemic
do not stand out in the data, even though this crisis was also
associated with nancial burdens.38
This illustrates that the challenges associated with the war
against Ukraine are reected in the public perception of the
nancial situation – an initial indication that a pessimistic
nancial outlook could also have an eect on the German
population’s support for DC. At this point, it is not possible
to denitively ascertain whether the war against Ukraine is
the sole reason for this trend in how the economic situation is
perceived or whether this results from an accumulation of crises,
starting with the Covid-19 pandemic. Furthermore, the period
under consideration covers neither the ruling of the Federal
Constitutional Court on the federal budget in November 2023
nor the farmers’ protests in January 2024, which might also have
had an impact on the perception of the economic situation.
Perception of the economic situation
and support for DC
To understand the relationship between the perception of the
economic situation and support for DC, both individual changes
and general dierences between the respondents must be
considered. The following section starts by dealing with how
the respondents’ assessment of the economic situation relates
to support for DC over time (within eect). To put it another way:
if respondents have an increasingly negative perception of the
economic situation over the course of the survey waves, does
their support for DC also fall? It also looks at the relationship
between the average assessment of the economic situation and
the respondents’ support for DC (between eect). In other words:
do the respondents who, on average, have a more negative view
of their own (or Germany’s) economic situation across surveys
also support DC less, on average, than those who have a positive
view of the economic situation?39
A negative assessment of the respondent’s own personal
economic situation and the national economic situation
correlates with less support for DC over the course of time.
The longitudinal analysis on an individual level yields two key
results (see Figure 24): (1) respondents whose perception of
the economic situation gets worse over the course of time also
support DC less on average (within eect). (2) Respondents
who generally give a poorer assessment of the economic
situation, measured on the basis of their average perception
across all survey waves, support DC less than those who rate
the economic situation more positively (between eect).
This pattern is repeated if public support for the current
development cooperation/ODA expenditure is analysed
instead of general support for DC.40 This means that the
indicator that more strongly reects support for current
development policy activities and for which a clear decline has
been observed since January 2022, as described in Chapter 3,
isalso negatively related to the respondents’ perception of their
own personal economic situation and the national economic
situation. All things considered, the ndings can be regarded,
both for general support for DC and for public support for
the current development cooperation/ODA expenditure, as
an initial indication that the more negative perception of the
economic situation is causing public backing for DC to dwindle
(see also the results in Chapter 3).
38 The monthly data from the political polling survey Politbarometer has shown an identical pattern since 2009. Here too, a much more negative assessment of the respondent’s
own personal economic situation and the national economic situation can be seen since the Russian war of aggression against Ukraine began, even though the economic
situation was already rated more negatively by some people during the coronavirus pandemic (Forschungsgruppe Wahlen, 2024).
39 All the analyses were carried out with all the available observations (unbalanced panel). This means that the respondents did not need to have taken part in all ve surveys.
Alternatively, all the models were therefore computed for the respondents who took part in all ve waves. The results are not substantially dierent.
40 Furthermore, it is apparent that the majority of the spread in support for DC can be attributed to dierences between people and not to individual changes over the course of time
(see also Section 3.4 on attitude strength). The intraclass correlation coecient is 68 percent for general support for DC and 65 percent for support for the DC budget. This key gure
indicates that around 70 percent of the spread in support for DC can be attributed to dierences between people and around 30 percent to changes over time (see, for example,
Snijders and Bosker, 2012, p. 17-23). By analogy with the results relating to attitude strength, this suggests that the respondents’ support is stable, but not entirely invariable.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy56
Figure 24 Inuence of the assessment of the economic situation on support for DC
-0.12 ***
-0.08 ***
-0.67 ***
-0.24 ***
Own personal economic situation
(between)
National economic situation
(between)
Own personal economic situation
(within)
National economic situation
(within)
-0.8 -0.4 0.0 0.4 0.8
Unstandardised coefficients
Source: DEval, own visualisation. Source of the data: DEL panel wave 1 to 5; N per wave ≈ 6,000. Multi-level model for longitudinal data (random intercepts). The
following variables were also controlled for: education, gender, age, political orientation and survey wave. The coefficients of the control variables can be found in Table
25 in Section 2.2.3 of the online appendix. The full wording of the question for all the variables used can be found in Table 18 in Section 2.2.1 of the online appendix.
The relationship between the perception of the economic
situation and general support for DC has only changed slightly
since the war in Ukraine began. Now the question is whether the
assessment of the economic situation has a greater inuence on
support for DC in the wake of the Russian attack on Ukraine and
the resulting economic impact on Germany than in the previous
survey waves. This could be a further indication to explain the
declining support for DC in Germany (see Chapter 3). However,
the analysis shows that the relationship between the assessment
of the economic situation and general support for DC only varies
slightly with the survey waves (see Figure 47 in the Annex).41
Overall, therefore, the results do not suggest that the negative
relationship has strengthened as a result of the war.42
The relationship between the perception of the economic
situation and public support for the current development
cooperation/DC expenditure changes relatively little over
the course of time, albeit somewhat more substantially
than the relationship with the general endorsement of DC.
AsFigure 48 in the Annex shows, a more negative assessment
of the national situation in the survey waves after the beginning
of the war against Ukraine in February 2022 goes hand in hand
with less support for the current development cooperation/
ODA expenditure.43 This nding can be seen as an indication
that the perception of the economic situation is having a
greater negative impact on support for DC in the course of the
war against Ukraine. However, this is not a causal relationship.44
41 For the analysis, interaction terms were added to the models between the within indicators relating to the perception of the economy, and to the dummy variables for
recording the survey waves. The interaction models can be found in Table 25 in Section 2.2.3 of the online appendix.
42 Detailed comparisons of the eect of the assessment on the national economic situation yield no statistically signicant dierences between the two survey waves of the DEL
panel after the Ukraine war began in February 2022 and the three preceding waves. When it comes to the respondents’ assessment of their own personal economic situation,
the eect only diers between wave 5 and wave 3 (p < 0.05). With regard to statistical signicance, see also Box 4.
43 The comparison of the eect in wave 5 and wave 4 is signicant at the 10 percent level, whereas the comparison with wave 3 is signicant at the 5 percent level. This means
that the dierence between wave 4 and 5 is aicted with greater statistical uncertainty. For more detailed results, see Table 26 in Section 2.2.3 of the online appendix.
44 This is because the analysis relates to Germany and thus does not contain a control group that is not aected by the consequences of the Ukraine war.

57Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
4.3.3 How stable is support for DC in the face
ofmilitary conicts and economic worries?
Following on from the results of the previous sections, the
question is whether fear of military conicts and economic
worries could be causing the decline in support for DC. It has
already been shown that relationships exist between support
for DC and the assessment of the economic situation as well
as between support for DC and attitudes towards support for
Ukraine and other countries in emergency situations.
Therefore, a survey experiment has been conducted,
investigating, on the one hand, whether information about the
wars in Ukraine and Yemen has an inuence on attitudes towards
DC.45 These wars were used in the experiment because they dier
in terms of a characteristic that could inuence support for DC:
psychological distance from the German population. “Construal
level theory” describes how psychological distance – spatial,
temporal or social – inuences how people interpret and respond
to events (Trope and Liberman, 2010). Ukraine, as a European
country with a predominantly white, Christian population, could
be perceived by the German population as psychologically closer
than the predominantly Muslim Yemen, which is located on the
Arabian peninsula. In addition, the war against Ukraine has a
greater media presence, and the economic consequences are
more noticeable. The smaller distance from the war in Ukraine
could lead to the mental “image” of this war being more concrete
and evoking more solidarity than the potentially more abstract
image of the more distant war in Yemen (Kogutetal., 2018).
On the other hand, the experiment tests whether
information on the tense situation surrounding the federal
budget has an impact on attitudes towards DC. To induce
economic worries, respondents read a text about the freeze
on additional expenditure for the federal budget that was
decided upon at the end of 2023. At the time of the survey,
this was a very present topic in the media, encompassed
a lot of policy areas and therefore had the potential to be
clearly felt by many citizens. As mentioned at the beginning
of Section 4.3, this could lead to respondents increasingly
prioritising the population of Germany (ingroup) over the
population of developing countries (outgroup) when it comes
to the use of public funds (see also Heinrich et al., 2016).
It is also conceivable that economic worries have dierent
eects on DC-related attitudes depending on whether
respondents have read the text about Ukraine or the text
about Yemen. As people in Ukraine are possibly perceived to
be closer and more similar, an increased ingroup eect could
occur here, leading to greater support (Kossowska et al., 2023).
However, the (nancial) consequences of this war are much
more noticeable and thus more concrete, which in turn could
lead to a higher level of rejection of DC/ODA expenditure.
On the other hand, the strained nances could lead to a
devaluation of the outgroup, in this case Yemen, as a country
that is geographically and socially more distant, and thus to less
support for DC. However, it could also be the case that Yemen
is perceived more as a “developing country” and thus evokes
greater solidarity and therefore support for DC.46
45 For the full structure of the experiment and the associated information texts, see Table 18 in Section 2.2.1 of the online appendix. The experiment was also pre-registered on
the platform OSF before the data was collected (see Chapter 2) and can be viewed there at https://osf.io/khvxp.
46 Corresponding variables were also recorded for this purpose and evaluated in an exploratory analysis. Information about this and about the hypotheses formulated above can
be viewed in the pre-registration document.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy58
Figure 25 Eect of information on the budget freeze and on the wars in Yemen/Ukraine
on support for development cooperation
5.1
4.7 5.3
5.1 5.1
4.8
No support at all 0
1
2
3
4
5
6
7
8
9
A great deal of support 10
Budget freeze No budget freeze (control group)
Average
support
War in Yemen War against Ukraine No war (control group)
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in January 2024. N = 2,046. The dependent variable is support for
German DC. The gure shows the mean values in the respective experimental groups with 95% condence intervals. The dotted line represent s the centre of the scale,
the value “5”. In the “no war” control group, the respondents were shown a text about a UNESCO World Heritage Site in Denmark. In the “no budget freeze” control
group, the respondent s received a text about the Bundestag resolution on telephone sick notes. For the full treatments used in the experiment, see Table 18 in Section
2.2.1 of the online appendix.
The results of the experiment show that economic worries
lead to less support for DC. Respondents support DC less
if they were made aware of economic worries in advance
through a reference to the budget freeze than if these worries
were not present. This was evident irrespective of whether
the people were also informed about a military conict. This
eect is small, with the support for DC having decreased by
0.3 scale points; however, the intervention only consisted of
a small text. Even the brief reference to the budget freeze
can therefore lead to a small, but statistically signicant
decline in support for DC. This eect can also be seen across
various demographic and socio-economic characteristics of
the respondents.47 And it even relates to respondents who
tend to be more in favour of DC. These are, for example,
people who place themselves more on the left of the
political spectrum or have a high level of formal education.
The results also show that respondents support DC more
strongly if they have been confronted with information about
the war against Ukraine than if they received information on
the war in Yemen. This dierence is also 0.3 scale points on
average, is statistically signicant and can be seen irrespective
of whether the respondents had read the text about the
budget freeze. Therefore, the war against Ukraine, which is
psychologically near, appears to trigger more support for DC
than the distant war in Yemen.
47 Providing information about the budget freeze has a signicant eect on support for DC when the experiment controls for income, gender, education and political orientation
(left-right scale). Providing information about the budget freeze shows no signicant interaction eects with these variables (with the exception of a high level of formal
education), which means that the strength of this eect hardly diers across these characteristics (see Table 27 in Section 2.2.4 of the online appendix).

59Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 26 Eect of information about the wars in Ukraine/Yemen on support for development cooperation
Support for DC
Security threat
Sense of
being affected
Similarity
Geographical
proximity
Financial burden
Developing country
Ukraine
(v. Yemen)
Eects
posive
indirect
negave
indirect
Relaonship
posive
negave
none
Source: DEval, own visualisation. The gure maps out possible mechanisms through which the information on the wars in Ukraine and Yemen can have dierent
eects on support for DC. These have been examined in six separately calculated mediation analyses. Orange arrows represent a signicant positive correlation in the
mediation analysis, dark blue arrows a signicant negative correlation, and grey arrows no signicant correlation. Orange boxes mean that the “war against Ukraine”
has a signicant positive indirect eect on support for DC, which is mediated, i.e. conveyed, by the respective variable. Dark blue boxes mean that there is a signicant
negative indirect eect, grey boxes that there is no signicant indirect eect. The DEval survey conducted by Respondi/Bilendi in Januar y 2024 with a subsample size
of N = 1,384 serves as the source of the data. Respondents who were in the “no war” control group have been eliminated from the subsample, as they did not receive
the questions about Ukraine and Yemen. The detailed results can be found in Tables 29–34 in Section 2.2.4 of the online appendix.
The greater support for DC from those who were informed
about the war in Ukraine can be explained as follows:
(1) people in Ukraine are perceived by the respondents as
being similar to themselves, (2) the respondents feel aected
by the situation in Ukraine, and (3) the war in Ukraine is
perceived as a threat to Germany’s security. In a mediation
analysis (see Bruder et al., 2020; MacKinnon et al., 2012),
various mechanisms were studied that can potentially explain
the dierence in support for DC between respondents who
were informed about the war in Ukraine and those who were
informed about the war in Yemen (Figure 26 illustrates these
mechanisms).48 The perceived similarity, the sense of being
aected and the perceived security threat are factors that are
more pronounced with regard to the situation in Ukraine than
with regard to the situation in Yemen and are also associated
with greater support for DC. At the same time, thegeographical
proximity of Ukraine in comparison with Yemen is not a factor
that is connected to greater support for DC. The fact that the
nancial burden caused by the war against Ukraine is perceived
as higher and that Ukraine is regarded less as a developing
country than Yemen is associated with a lower level of support
for DC among respondents. Nevertheless, support for DC
among the respondents who read the text about the war in
Ukraine is greater overall than among the respondents who read
the text about the war in Yemen (see Figure 25). Overall, these
results suggest that the perceived psychological distance from
individual countries (i.e. sense of being aected by the situation
and perceived similarity), as well as security considerations,
play a role in the attitudes towards DC, especially with regard
to support for countries in acute war situations.
48 For the precise results of the mediation analyses, see Tables 29-34 in Section 2.2.4 of the online appendix.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy60
4.3.4 Which countries are preferred as
partnercountries for German DC in the context
ofgeopolitical tensions?
Scarce nancial resources and public criticism for DC raise
the question, in public and political debate, of whether
Germany should reduce the number of its partner countries
or prioritise between them. As nancial resources for DC are
in short supply, a key challenge for development policy actors,
above all the BMZ, is to determine both the thematic focal
points and the partner countries. In the BMZ 2030 reform
strategy (BMZ, 2020), the BMZ denes the partnership
categories of “bilateral partners”, “transformation partners”,
“global partners” and “nexus and peace partners”. The aim is
to use DC resources more eectively and eciently thanks to
a targeted selection of partners.
The selection of partner states is also a subject of political
discourse. For example, the AfD parliamentary group in the
Bundestag submitted an interpellation about the basic criteria
for the selection of partner countries in June 2020 (German
Bundestag, 2020). The collaboration with specic countries is
also regularly scrutinised – for example in minor interpellations
submitted by various parliamentary groups in the Bundestag
(German Bundestag, 2019, 2022a, 2023). Most recently, DC
with China, for example for gender training, has been publicly
criticised and called into question (Focus online, 2024). Against
this backdrop, German DC is subject to increasing pressure to
justify itself, especially in terms of the selection of projects and
partner countries.
The selection of the partner countries and the attitudes of
the German population towards this also play a vital role
for civil society. For example, NGOs are primarily nanced
by donations and voluntary engagement in addition to
government funds (Dreher et al., 2012; Verbrugge and Huyse,
2020). The extent to which cooperation with particular
countries is supported by the German population is likely to
be relevant for the willingness to donate.
The prioritisation of partner countries and the possible
withdrawal from partner countries are accompanied by the
risk that actors such as China or Russia will ll the resulting
gap and thus also pose major challenges for German
development policy. The inuence of emerging countries such
as China, Russia and India as development policy actors and as
donor states has been growing continuously for several years
(see, for example, Dreher et al., 2022; Fuchs and Vadlamannati,
2013; Gray, 2015). If, as in the case of some countries in Africa,
cooperation with traditional donor states – here France – is
terminated because of their colonial past, countries with rival
systems are ready to ll this gap. The Western world’s inuence
in the Global South is threatening to dwindle. This was
demonstrated particularly impressively after Russia invaded
Ukraine in February 2022. Many countries in the Global South
abstained in the subsequent resolutions of the UN General
Assembly, which led to great surprise and resentment in some
of the Western states (Plagemann, 2022).
In politics and society, the question arises of whether
geopolitical factors should play a role in the selection of
partner countries. For example, the question of whether the
German Federal Government should get involved in states that
do not share Germany’s social, democratic and regulatory values
is the subject of public discussion (ntv, 2024). At the same time,
DC is facing the challenge that more than 70 percent of the
people in the world live in autocracies (Nord etal., 2024) and
a considerable portion of DC is implemented in fragile states
(Faust et al., 2023; Wencker and Verspohl, 2019).
In order to gure out whether geopolitical factors have an
inuence on which countries the population prefers as partner
countries for German DC, a so called choice-based partial
prole conjoint experiment was conducted on the basis of
the experiment in the Opinion Monitor 2021 (Schneider et al.,
2021a, Chapter 6) (see Box 7). This experiment made it possible
to determine what characteristics of potential partner countries
are likely to lead to respondents preferring one country over
another as a partner country. In this study, various strategic and
geopolitical characteristics have been added to the experiment
in comparison with the Opinion Monitor 2021.

61Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Specically, respondents were asked about the following eleven
attributes (for a detailed overview of all the attributes and their
levels, see Table 20 in Section 2.2.1 of the online appendix): 49
1. Thematic focal point of the cooperation
2. Religion of the population
3. Corruption in the country
4. Poverty in the country
5. Performance capacity of the state
6. State authority
7. Impact of climate change
8. Number of refugees coming from the country to Germany
9. Role as an energy supply for Germany
10. Political partnership with China or the West
11. Attitude to the war against Ukraine
In a second choice-based conjoint experiment, the preferences
for specic countries as partner countries for German DC
were studied.
Box 7 Structure of the conjoint experiment
In choice-based conjoint experiments, respondents are presented with two options and must choose between them (see,
forexample, Hainmueller et al., 2014; Raghavarao et al., 2010). The two options can, for example, be proles of dierent
products. Thereby, a series of dierent attributes is presented for each pair. The levels of the attributes that are shown are
determined on a random basis. In a partial prole conjoint experiment, the attributes shown are selected at random from a
larger total number of attributes for each pair. The participants must then make a decision based on these characteristics,
while the options “don’t know” or “both equally” are deliberately omitted. Based on these decisions, it is possible to determine
which levels of the attributes statistically lead to the choice in favour of a product more frequently or less frequently
For the Opinion Monitor 2024, such a choice-based partial prole conjoint experiment was used to investigate which
characteristics respondents consider as criteria for selecting recipient countries of German DC (see also Doherty et al.,
2020). Five pairs were presented to the respondents for selection, each with two country proles, each of which contained six
randomly selected attributes (concrete examples of the pairs can be found in Figure 16 in Section 2.2.5 of the online appendix).
The respective levels of the presented attributes were also selected on a random basis. The presented country proles were
not based on actual countries, but were made up of a random selection of attributes and levels. The respondents then had
to select a country from each pair that they would prefer the German Federal Government to support with development
cooperation.
Following the selection, regression analyses were used to calculate the average marginal component eects (AMCEs). These
indicate how the probability of a country prole being selected changes when an attribute changes from a reference level to
another level. This study design makes it possible to draw causal conclusions about which levels are more likely to lead to a
country prole being selected.
49 With the exception of the attributes “thematic focal point of the cooperation”, “religion of the population” and “poverty”, the respective levels of the attributes allow for
subjective perception. For example, the levels “high level of corruption” and “close political partner of China” are not quantied more precisely. Therefore, precisely what they
mean is a matter for the subjective perception of the respondents.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy62
Preferences for country characteristics
Self-interest and geopolitical trade-os can play an important
role in the prioritisation of DC partner countries. Especially in
the context of current events such as the war against Ukraine
or geopolitical tensions with China, greater emphasis could be
placed on such trade-os. In order to examine this role more
closely, three additional attributes were used in the (hypothetical)
country proles that were not used in the 2020 survey: whether
the country supplies Germany with energy, whether a country
is a close partner of China or the West, and what a country’s
position is on Russia’s war of aggression against Ukraine.
(Hypothetical) countries with the attribute “energy supplier
for Germany” are clearly favoured as DC partner countries.
In the course of the Russian war against Ukraine, energy
security has become a focus of social and political debate. After
decades of depending on fossil fuels from Russia and the lack
of gas supplies since the war began, the questions of which
countries Germany should obtain energy from and what role
the political conditions in the respective countries play are
becoming increasingly important. In this context, possible
energy partnerships and their compatibility with the principles
of a feminist foreign and development policy were also
discussed. The results in Figure 27 show that energy security
plays an important role in the respondents’ attitudes towards
DC. The probability that a country is prioritised as a DC partner
is 10 percentage points higher if the country is presented as an
energy supplier for Germany.
The geopolitical alignment of a potential partner country
is also relevant to the respondents: if a country is deemed
to be a close partner of China, it is likely to be rejected as a
potential partner country for German DC. In this case, the
probability of the country being selected by the respondents as
a DC partner country decreased by 8 percentage points. On the
other hand, no statistically signicant dierence was evident if
a hypothetical country was presented as a close partner of the
West or has not committed itself either to the West or to China
as a partner.
Countries that are presented as supporters of Russia’s attack
on Ukraine experience by far the highest level of rejection as
potential partners of DC. As mentioned above, many countries
in the Global South abstained in the UN General Assembly
resolutions condemning the Russian war against Ukraine. In
comparison with these countries, countries that condemn this
war are 3 percentage points more likely to be prioritised as a
partner country. Only very few states demonstrate open support
for Russia’s attack on Ukraine. For example, in February 2023,
only six other countries aside from Russia itself voted against
Resolution A/ES-11/L.7 of the UN General Assembly and thus
against condemning the Russian attack (United Nations, 2023b,
2023c). Such a position led to a high level of rejection among
the respondents: countries that support Russia’s attack are 28
percentage points less likely to be favoured. This represents the
strongest eect in the results.

63Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 27 Eect of information on the choice of potential partner countries for DC
War against Ukraine
Political partnership
Number of refugees
Role as an energy supplier
State authority
Capacity of the state
Climate change
Poverty
Corruption in the country
Religion of the population
Thematic focal point
-0.3 -0.2 -0.1 0.0 0.1
Education
Health care
Peace and security
Economic promotion
Food security and agriculture
Climate change
Support for women and gender equality
Predominantly Christian
Predominantly Muslim
Medium level of corruption
High level of corruption
Low level of corruption
Poor
Very poor
Extremely poor
Country is hardly affected by the consequences of climate change
Country is heavily affected by the consequences of climate change
State can reliably meet the population’s basic needs
State can only meet the population’s basic needs to a limited extent
State can largely ensure law and order
State can only ensure law and order to a limited extent
Is not an energy supplier for Germany
Is an energy supplier for Germany
No refugees
Many refugees
Few refugees
Country is not committed to China or the West as a partner
Country is a close partner of the West
Country is a close partner of China
Is neutral towards Russia’s attack on Ukraine
Condemns Russia’s attack on Ukraine
Supports Russia’s attack on Ukraine
Variation in the probability that a
profile will be selected
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in October 2023. N = 3,355. The dependent variable is whether one
country is selected over another country. The gure shows the average marginal component eects (AMCEs) of the various explanatory variables. The AMCEs were
calculated using a linear probability model. The reference category of the respective variable can be seen on the zero line. The rest of the dots represent the respective
AMCE coecient s. A positive coecient means that this characteristic has a positive eect on the dependent variable. Conversely, a negative coecient means a
negative eect on the dependent variable. The horizontal bars through the dots represent the 95% condence intervals. If these intersect the zero line, this variable has
no statistically signicant eect. Otherwise, the eect is statistically signicant .

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy64
Attributes geared towards the needs of the countries, such as
poverty, limited performance capacity of the state and a high
level of vulnerability to climate change make the respondents
more likely to favour the country as a partner country. For
example, a country prole is 5 percentage points more likely to
be selected if the country is heavily aected by the consequences
of climate change. If the majority of the population of a country
is described as being very poor or extremely poor, the country is
3 or 6 percentage points respectively more likely to be preferred
as a partner country than a country that is only described as
poor. States that can only meet the basic needs of the population
to a limited extent are 3 percentage points more likely to be
preferred. The state authority, i.e. the indication of whether
a state can largely ensure law and order or can only do this to
a limited extent, does not play a statistically signicant role
when prioritising partner countries over one another. The same
applies to the number of refugees that come to Germany from
a country. This is interesting in light of the fact that “combating
the causes of ight” is frequently put forward as an argument to
DC-sceptics (see, for example, BMZ, 2024; Federal Government
Commission on the Root Causes of Displacement, 2021).
A high level of corruption leads to less support as a partner
country. Countries with a high level of corruption are
5percentage points less likely to be prioritised than countries
with a medium level of corruption. In contrast, countries for
which a low level of corruption is specied are 3 percentage
points more likely to be preferred. One explanation could be
that corruption can decrease the eectiveness of the DC funds
used, leading to reservations on the part of the respondents
(see also Bauhr et al., 2013).
Countries with a predominantly Muslim population are much
less likely to be supported as a partner country than countries
with a predominantly Christian population. The dierence
here is 12 percentage points, which is the second largest eect
in these results. Thus, the dierence is greater than,
for example, for the level of poverty or corruption. This anti-
Muslim bias is also relevant for the acceptance of refugees
(Bansak etal., 2016), as demonstrated by the debate about the
reception of Ukrainian refugees (Esposito, 2022). It can also be
found in other studies relating to DC preferences (Blackman,
2018). Prejudices against people of the Muslim faith could be
the decisive factor here (see, for example, Strabac and Listhaug,
2008; Sides and Gross, 2013).
Preferences for partner countries
The analysis in the second part of the experiment, in which
people had to choose between real countries, shows that the
respondents particularly support cooperation with countries
in sub-Saharan Africa. Figure 28 presents the results of the
second choice-based conjoint experiment. For each of the
89 countries included in the experiment, the map shows the
percentage of cases in which it was chosen as a partner country
over another randomly selected country. The most frequently
selected countries were Namibia, Ethiopia, Kenya and Togo,
all of which are located in sub-Saharan Africa. Three of these
countries (Namibia, Tanzania, Togo) are also former German
colonies. Among the 15 countries with the greatest support,
there are only two that are not in sub-Saharan Africa. Therefore,
this appears to be where the German population sees the
regional focus of German DC. This picture is consistent with
the results from the Opinion Monitor for Development Policy
2021 (Schneider et al., 2021a, Chapter 6).
China and the Arab region are least likely to be preferred as
partner countries by the German population. At around 29
percent, China is the country that was selected by far the least
frequently (see Figure 28). Iraq, Saudi Arabia and the United
Arab Emirates also have a much lower selection rate than the
other countries, at 34 to 35 percent.50 In total, eight of the ten
least frequently selected countries are in North Africa or on the
Arabian peninsula.
50 The United Arab Emirates and Saudi Arabia are not partner countries of German DC. These countries were included in this experiment to ensure that the data could be
compared with the data from 2020.

65Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 28 Endorsement of possible partner countries for DC
Selected as
a partner country
50-60 %
40-50 %
30-40 %
20-30 %
No data
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in October 2023. N = 3,355. The gure shows the frequency with
which a country would be chosen over another randomly selected countr y. 50 percent represents the average and thus the expected value. Orange countries were
selected with a frequency of more than 50 percent, blue countries with a frequency of less than 50 percent. All the countries that are on the List of ODA Recipients
ofthe Development Assistance Committee of the Organisation for Economic Co-operation and Development’s (OECD-DAC) were included in the survey.
Countries in Central Asia and Eastern Europe, especially
Ukraine and the Republic of Moldova, have gained
support in comparison with 2020. Whereas Ukraine was
still one of the least frequently selected partner countries
in 202051 with 37 percent support, support for the country
is now 14 percentage points higher, at around 51 percent.
Therefore, it is clear that DC with Ukraine has been endorsed
by considerably more people since the Russian war of
aggression began than previously – even though Ukraine is not
perceived as a “developing country” by the majority of people
(see Section 4.3.3). In this context, it is also interesting that
the Republic of Moldova was selected more frequently than in
2020 (increase by almost 12 percentage points). At 56 percent,
this makes it the sixth most frequently selected country.
Thistoo should be understood in the context of the war against
Ukraine. The Republic of Moldova is an immediate neighbour
of Ukraine, received (measured by the number of inhabitants
of the country) a large number of refugees from Ukraine
and has a secessionist region, Transnistria, that receives
military assistance from Russia. Therefore, countries that are
potentially in Russia’s political and geographical sphere of
inuence have gained support as partner countries for German
DC since the Russian war of aggression began. One possible
explanation could be that geopolitical and security incidents
have an inuence on the respondents’ attitudes towards DC.
51 The data used for the Opinion Monitor for Development Policy 2021 was collected in a conjoint experiment in August and September 2020.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy66
4.3.5 Conclusion: support for international
engagement is high, but could wane as a result
ofongoing economic challenges
Overall, respondents show a high level of support for
Germany’s engagement in international crises and emergency
situations. Humanitarian and development engagement
receive more support than military engagement.
Even two years after the Russian attack on Ukraine, support
for the country is strongly endorsed. The high level of
support sends an important message to politicians, especially
with regard to the vast funds required for the reconstruction
of Ukraine.
However, the strained economic and nancial situation
potentially has a negative eect on support for DC. Across
dierent demographic and socio-economic groups, economic
worries lead to a decline in support for DC, while at the same
time a large proportion of respondents rate the current and
future economic situation as poor. Considering the challenges,
which are not getting any smaller, and the associated costs
– both for the citizens and for the state – the previous
broad consensus in the population regarding development
policy could disappear in the long term (see also Chapter 3
of this report).
Strategic and geopolitical factors play a role in the selection
of potential partner countries. Factors that represent the
needs of the local people, such as poverty or being aected by
the consequences of climate change, are still important to the
respondents when choosing their preferred partner countries.
However, the country’s role as an energy supplier for Germany
and its position regarding Russia and China are also taken into
account in the decision. In addition, several Eastern European
and Central Asian countries have gained support as partner
countries, which must undoubtedly also be seen in the context of
Russia’s attack on Ukraine. The latest incidents relating to foreign
and security policy play an important role in shaping attitudes
towards development policy. Designing and communicating
a development policy that shows consideration for these
preferences could therefore lead to a higher degree of acceptance.
DC only appears to be perceived by respondents as an
eective crisis tool to a limited extent. This appears to depend
on the respective context. On the whole, respondents endorse
Germany’s engagement in countries experiencing acute crises
as well as DC interventions and humanitarian aid for Ukraine.
However, the results also show that countries plagued by crises
such as Iraq, Afghanistan or South Sudan have lost support as
partner countries in comparison with 2020. Here, the largely
negative outcome of German engagement in Afghanistan
(Hartmann et al., 2023), which became clearly apparent once
again after the withdrawal of the troops in 2021 and was also
visible in the media, could have diminished the population’s
condence in DC as an eective crisis tool.

67Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
4.4 The attitude of the German population towards feminist development policy
have heard of it,
but do not know
what it is
have heard of it
Familiarity
with the term
is increasing
know what it is
Feminist development policy
is rated more negatively than
a “peace-promoting” or
“human rights-based”
development policy
general support
Unstable moderate
supporters
no support
At a
glance
Familiarity Support for feminist development policy
Support in the context of global crises
Which objectives are supported?
Attitude strength
Polarising eect of the term
25 %
52 %
27 %
19 %
37 %
28 %
16 %
52%
20%
7 percentage points
less than in 2023
5 percentage points
more than in 2023
15 percentage points
more than in 2022
18 percentage points
more than in 2022
9%
In comparison:
the Sustainable
Development Goals
are known by
Instrumental justication
Normative justication
Equal rights are
a human right
Equal rights in order to achieve
other DC objectives Greater support
&
greater
perception
of eectiveness
Less support for
transformative
objectives or the
concrete provision of
nancial resources
In the context of global crises,
feminist development policy that is presented
without justication receives less support
Discrepancy between
the supporters of dierent
parties is greater when
the label “feminist” is used
Strengthening the 3 Rs is also
rather supported by opponents
Stable opponents
Stable supporters
Ambivalent
supporters
Strong support for the 3 Rs:
Rights
Resources
Representation
18 percentage points
fewer than feminist
development policy
OPINION MONITOR FOR
DEVELOPMENT POLICY 2024
The attitude of the German population
towards feminist development policy

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy68
In its coalition agreement of 2021, the German Federal
Government explicitly announced a feminist foreign policy
for the rst time, and the BMZ presented a strategy for a
feminist development policy in March 2023 (Federal Foreign
Oce, 2023a; BMZ, 2023a; BMZ, 2023c; SPD, Alliance 90/The
Greens and FDP, 2021). A feminist development policy that ties
in with the gender equality goal (SDG 5) of the 2030 Agenda
adopted by the United Nations in 2015 continues the tradition of
taking gender equality into account in foreign and development
policy and in DC. It was 40 years ago that the parties to the
United Nations Convention on the Elimination of all Forms of
Discrimination Against Women (CEDAW, in force since 1981 and
ratied by Germany in 1985) undertook to promote women’s
rights around the world. 15 years later, the Platform for Action
for gender equality in all areas of society was adopted at the
1995 World Conference on Women in Beijing. Here, there
were already clear references to DC, such as the reduction of
poverty among women or the consideration of gender-specic
dierences in educational opportunities (e.g. BMFSFJ, 2021;
UN Women, 2020). In addition, former Federal Ministers for
Economic Cooperation and Development advocated gender
equality as part of their mandate and promoted important
aspects of a feminist development policy, even if these were
not explicitly referred to as such (Bohnet, 2019).
Current feminist development policy diers from previous eorts
to promote gender equality, at least in terms of its aspirations.
For example, current feminist development policy is supposed
to have a gender-transformative eect in particular and thus to
focus on systemic change and breaking up existing patriarchal
power structures. The strategy also goes beyond gender equality;
it has an intersectional aspiration, in which not only women and
girls, but also other marginalised groups are supposed to be taken
into account (BMZ, 2023a, 2023c). The key objectives of feminist
foreign and development policy are to strengthen the rights,
representation and resources of women and other marginalised
groups and to promote social diversity (“3R+D” formula; Federal
Foreign Oce, 2023a; BMZ, 2023a, 2023c; SPD, Alliance 90/The
Greens and FDP, 2021). The principles, contents and objectives of
feminist foreign policy overlap considerably with those of feminist
development policy, even though the elds of action are distinct
(Federal Foreign Oce, 2023a; BMZ, 2023a, 2023c).
With the feminist guiding principle for foreign and
development policy, Germany is following the example of
Sweden, where a feminist foreign policy was established as
early as 2014 (Zilla, 2022). Other countries have subsequently
announced a feminist foreign and/or development policy
(Khillare, 2023). However, the concept does not enjoy universal
support; there are often reservations in some sections of
politics and among the general public (Hudson et al., 2023;
Sassenhagen et al., 2023; Schneider et al., 2024b). For example,
it is denied that feminist foreign and development policy does
justice to the acute global security crises or is the appropriate
guiding principle for these.52 For instance, after the change
of government in 2022, Sweden at least abandoned the term
“feminist foreign policy”.
On the one hand, the feminist guiding principle for German
development policy has the potential of attracting attention
to existing inequalities, injustices and structural power
disparities. On the other hand, it is associated with the risk of
the basic consensus on development policy (Schneider et al.,
2022) that has hitherto existed within the population being
called into question in an increasingly polarised political
debate. From this perspective, it is particularly important to
capture public discourse and public opinion regarding feminist
development policy. While there is already comprehensive
academic literature on feminist foreign policy (e.g. Aggestam
et al., 2019; Jezierska, 2022; Scheyer and Kumskova, 2019;
Sundström et al., 2021; Thompson and Clement, 2019; Thomson,
2020; Towns et al., 2023; Zhukova, 2023), this does not apply
specically to feminist development policy. In particular,
the public attitude towards both feminist foreign policy and
feminist development policy has hitherto remained largely
ignored, apart from a few exceptions (but see Sassenhagen
et al., 2023; Schneider et al., 2024b on feminist development
policy). However, the public attitude is an important indicator
both for democratic legitimacy and as information for political
actors and their decision-making processes.
52 This is how the chairman of the CDU/CSU parliamentary group in the Bundestag, Friedrich Merz, expressed it in his speech in the Bundestag shortly before the Russian attack
on Ukraine began (German Bundestag, 2022b).

69Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Therefore, the aim of this chapter is to get a clearer picture
of what the German population’s understanding of feminist
development policy and how broad the support is for the
label and its contents, especially in the context of global
crises. As such, this chapter intends to give development policy
actors feedback and orientational knowledge regarding public
opinion of feminist development policy and thus information
for further policy decisions. For example, potential for political
communications and for the substantive implementation of
feminist development policy can be determined.
Against this backdrop, this section deals with the following
key questions:
1. What does the German population understand by feminist
development policy?
2. To what extent does the population support a feminist
development policy and its contents?
3. How stable are the attitudes towards feminist
development policy?
4. How do the perception of and support for feminist
development policy change in the context of acute global
challenges and crises?
4.4.1 What does the German population
understand by feminist development policy?
Even though the term “feminist development policy” was
already more familiar to the participants than the SDGs in
a 2022 survey, it was still unknown to the majority (59%)
(Sassenhagen et al., 2023).53 In March 2023, both the Federal
Foreign Oce’s strategy for a feminist foreign policy and the
BMZ’s strategy for a feminist development policy were published.
They received a lot of media attention. Both the public interest
and knowledge of the topic could have increased as a result.
Therefore, a key question is how much the German population
currently knows about feminist development policy and what
they understand by the term.
Feminist development policy on Twitter/X
As an indicator of the German population’s information
environment on the topic of feminist foreign and
development policy, the short message service Twitter/X
has been analysed.54 Records were made of how frequently
the terms “feminist foreign policy” and “feminist development
policy” were mentioned.
The attention given to feminist foreign and development
policy on Twitter/X remains largely unchanged since 2022.
Apart from a few spikes, only a few posts per day were counted
(see Figure 29). This number has largely remained stable and
shows no average increase over time. This suggests that even
the presentation of the policy strategies in March 2023 did not
initiate increased discussions about the topic in the long term.
During this period, there was just a brief rise in the number of
relevant posts.
A comparison of the posts published per day about both
policy areas show that most posts are concerned with feminist
foreign policy, while feminist development policy receives
hardly any attention. This could be because, in comparison
with foreign policy, development policy is only of interest to a
smaller section of the population and the political elite. On the
other hand, it could be the case that people barely dierentiate
between feminist foreign and development policy. For example,
the strategies of both ministries were presented together,
and the ministers were seen together in many of their public
appearances (e.g. Tagesschau, 2023).
53 The Körber Foundation’s study “The Berlin Pulse” came to a similar conclusion with respect to the term “feminist foreign policy” (Körber Foundation, 2022).
54 The same limitations apply to these analyses as to the analyses of general mentions of development policy on Twitter/X. These limitations are explained in Section 2.3.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy70
Figure 29 Feminist development policy/foreign policy on Twitter/X (2021–2024)
SPD/Green/
FDP/
coalition agreement
Friedrich Merz’s
Bundestag speech
Start of the
war in
Ukraine
Discussion
of German foreign policy
Ukraine war
Protests
in Iran
Presentation
of feminist foreign and
development policy
strategies
Parody Twitter
account
of Annalena Baerbock
Hamas
terror attack
on Israel/
War in Gaza
Delivery of
Eurofighters to
Saudi Arabia
0
500
1.000
1.500
2.000
2.500
08/21
09/21
10/21
11/21
12/21
01/22
02/22
03/22
04/22
05/22
06/22
07/22
08/22
09/22
10/22
11/22
12/22
01/23
02/23
03/23
04/23
05/23
06/23
07/23
08/23
09/23
10/23
11/23
12/23
01/24
02/24
03/24
04/24
Date
Number of tweets per day
Development policy Foreign policy
Source: DEval, own visualisation. Source of the data: around 69,000 posts that were downloaded between 1 May 2019 and 29 February 2024 using the analysis tool
Meltwater, with the help of a search term list . The search terms were „feminist foreign policy“ and „feminist development policy“.
Familiarity with feminist development policy
In comparison with the results obtained by the DEL in 2022,
the term “feminist development policy” is much better
known in more recent surveys. As can be seen in Figure 30,
only 43 percent of the respondents in January 2024 indicated
that they had never heard of or read about this term. For
the rst time, a small majority (52%) even said that they had
already heard of or read about feminist development policy. The
proportion of respondents who are familiar with the term and
know what it means has risen by 15 percentage points between
June 2022 (DEL survey) and January 2024 (DEval survey) – to
27 percent. This could indicate that the strategy published in
March 2023 and the associated formulation of the contents of
feminist development policy helped to give at least some of the
population a clearer idea of a feminist development policy.55
55 In this survey, it was not possible to clarify whether respondents can actually distinguish between development policy and foreign policy, as the respondents were not asked
about a distinction between the two terms and the questions only used the term “feminist development policy”. Therefore, it is conceivable that respondents say that they
know what the term means even if they only know about “feminist foreign policy” and do not dierentiate between foreign and development policy.

71Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 30 Familiarity with feminist development policy over the course of time
12
22
59
7
13
20
61
6
23
22
50
5
25
24
47
4
27
25
43
5
0
25
50
75
100
6/2022
(DEL)
10/2022
(DEL)
1/2023
(DEval)
6/2023
(DEval)
1/2024
(DEval)
Date of survey
Percentage of respondents
Yes, and I know what it is
Yes, but I don’t really know what it is
No
Don’t know
Source: DEval, own visualisation. Source of the data: 6/2022: DEL sandbox (N = 2,059), 10/2022: DEL panel (N = 6,008). DEL data is weighted data. 1/2023, 6/2023,
1/2024: DEval tracking (N for each survey ≈ 2,000). The question was: “Have you ever heard or read anything about ‘feminist development policy?”
The respondents who reported being more familiar with
feminist development policy are: respondents with a higher
level of formal education, men, people interested in politics,
people interested in and informed about DC, and people who
place themselves on the fringes of the political spectrum (left
or right). Attitudes towards feminism or populist attitudes,
on the other hand, have no relationship with familiarity with
feminist development policy.56
Public perception of feminist development policy
When the respondents are asked to express, in their own
words, what they believe feminist development policy is
about, it is evident that most people think that it is primarily
about women, women’s rights and equality. These three
terms were mentioned most frequently.57 “No idea” was also
among the 15 most common responses. Positive words such as
“promote” or “empower” were also frequently included in the
responses. Interestingly, some respondents also believe that the
term relates to topics such as “quotas of women”, “more women
in politics and business” or “women in positions of leadership”.
Therefore, both women’s rights and the representation of
women appear to be present in the responses – and both are
key aspects of the BMZ strategy (BMZ, 2023a). On the other
hand, the aspects of “resources” and “diversity” largely went
unmentioned.58
Reservations about a feminist development policy include
doubts as to whether the objectives are achievable, the
concern that men could reject this policy or be excluded from
it, and questions regarding the extent to which a feminist
development policy clashes with the values of other cultures.
For example, respondents express the fear that “nothing is
really changing and it is just a lot of talk” or that it could “treat
women in a one-sided manner and discriminate against men”.
It was also feared that “dierent cultures […] will reject […] this
development policy”. Respondents also indicate that a feminist
development policy reects Western values in particular,
which would be imposed on the countries of the Global South
(“We are trying to inuence prevailing structures, to impose
56 The results of the in-depth regression analysis, which takes a closer look at the relationships between demographic variables and attitude variables and familiarity with
feminist development policy is documented in Section 2.3.2 of the online appendix.
57 An overview of the 15 most frequently mentioned words in response to the question of what people think a feminist development policy is about can be seen in Figure 18 in
Section 2.3.3 of the online appendix.
58 Open-ended questions were analysed with the help of various computer-assisted text analysis methods. Details of the analyses of the open-ended questions in the survey can
be found in Section 2.3.3 of the online appendix.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy72
our standards on other countries […]”. However, these are
presumably the reservations of informed citizens, as a large
proportion of the population claims to have no reservations
or no knowledge about this. “No idea”, “don’t know” and “no
reservations” are amongst the most common phrases used in
the response to this question.59
Target groups of feminist development policy
according to the German population
On average, respondents believe that the target group of a
feminist development policy is women and girls in particular,
with other groups left far behind. There is a discrepancy
here from the target groups that should be reached by
development policy in the eyes of the respondents. As
can be seen in Figure 31, respondents indicate that women
and girls are and should be one of the most important
target groups of development policy. Regarding almost all
other groups, the priorities of the respondents deviate from
their idea of which groups are actually reached by a feminist
development policy. It is evident that the intersectional
aspiration of feminist development policy is not present for
the respondents. The label “feminist” could sound to many
laypeople as if development policy will only be about women
and girls in the future. The fact that the target group “men and
boys” shows the greatest discrepancy, i.e. is least frequently
regarded as the actual target group of feminist development
policy while being seen as a key desired target group of
development policy, demonstrates the concern that men and
boys could be neglected as a result of the focus on women,
girls and marginalised groups. This concern is consistent with
the reservations regarding feminist development policy that
have already been mentioned, in which “discrimination against
men” was mentioned as one aspect.
Figure 31 Target groups of feminist development policy according to the German population
Men and boys
Refugees
Religious/ethnic minorities
Indigenous population
LGBTQIA+
Entrepreneurs in developing countries
People with disabilities
People living in poverty
Women and girls
1
Not
reached
2 3 4 5 6 7
Reached
Average assessment
Feminist development policy
would like to reach
this target group
Development policy
measures should reach
this target group
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 2,000. The gure shows the mean values with
95% condence intervals. The mean values correspond to the average assessment of the respondents as to which target groups a feminist development policy would
like to reach (“Please use the following scale to assess the extent to which you believe that feminist development cooperation would like to reach the specied group of
people”) or which target groups development interventions in general should take into account (“Thinking of development interventions in general, to what extent do
you think that the specied group of people should be taken into account by development interventions.”) The scale ranged from 1 = “not reached at all”/“should not be
reached at all” to 7 = “denitely reached”/“should denitely be reached ”. The value “4” corresponds to the centre of the scale.
59 An overview of the 15 most frequently mentioned words in response to the question of what reservations people have about a feminist development policy is presented in
Figure 20 in Section 2.3.3 of the online appendix.

73Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
4.4.2 Does the population support feminist
development policy and its content?
The majority of respondents support a feminist development
policy, even though support is declining over the course of
time. As Figure 32 shows, around 52 percent of respondents in
January 2024 stated that they support a feminist orientation
as a focus of development policy, while just under 20 percent
stated that they do not support such a development policy.
Around 20 percent selected the option “neither support nor
don‘t support” and around another 9 percent “don’t know”.
Corresponding to support for DC in general, a decline can be
seen here in comparison with January 2023. The proportion of
respondents supporting a feminist development policy was
around 7 percentage points higher in January 2023. The group that
doesn’t support it was also smaller (around 5 percentage points).
Figure 32 Support for a feminist development policy over the course of time
59
15
18
8
56
18
18
8
52
20
20
9
0
25
50
75
100
1/2023 6/2023 1/2024
Date of survey
Percentage of respondents
Support
Neither support nor don't support
Don't support
Don't know
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2023, July 2023 and January 2024. N for each survey
≈ 2,000. The seven-point response scale was condensed for the visualisation. The question was: “One focus of the current German Federal Government is feminist
development policy. This is about increasing the rights, representation, resources and equal opportunities of women, girls and disadvantaged groups in developing
countries. These groups include people who are disadvantaged because of their religion, sexual orientation, gender identity or a disability, for example. What is your
opinion of this focus?”
Feminist development policy receives less support than other
possible focuses of development policy. Sassenhagen et al.
(2023) have already demonstrated, with DEL data from June
2022, that the focus on feminist development policy comes
o substantially worse than a “human rights-based” or “peace-
promoting” development policy, for example. A more recent
survey from August 2023 reaches a similar conclusion (see
Figure 33). Feminist development policy is rated slightly more
positively here than in June 2022, but still more negatively than
a “peace-promoting” or “human rights-based” development
policy. Only a development policy “guided by German
interests” is assessed even more negatively.60
60 The results are based on data collected from two dierent surveys (a DEL survey and a DEval survey conducted by Respondi/Bilendi), which dier both in terms of the contents
of the overall survey and in terms of the order of the questions. Even if the wording of the question remained the same, it cannot be ruled out that dierences may have arisen
due to the dierent survey modes.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy74
Figure 33 Support for dierent focuses of development policy
Human rights-based
Feminist
German interests
Promoting peace
1
Very
negative
234567
Very
positive
Average assessment
6/2022
(DEL)
10/2022
(DEL)
3/2023
(DEval)
8/2023
(DEval)
Source: DEval, own visualisation. Source of the data: DEL sandbox in June 2022 (N = 2 ,059), DEL panel in September 202 2 (N = 6,008); DEL data is weighted data.
DEval surveys in March 2023 (N = 2,048) and August 2023 (N = 2,000). The gure shows the mean values with 95 % condence intervals, broken down according
to the date of the survey and the focus of development policy. Respondent s were supposed to provide, on a scale from 1 (“very negative”) to 7 (“very positive”), their
assessment of dierent types of development policy. The value “4” corresponds to the centre of the scale and thus to an assessment that is neither negative nor positive.
Respondents were asked about six kinds of development policy in total. To simplify the visualisation, only four kinds of development policy are shown here.
Support for the objectives
of feminist development policy
The BMZ strategy describes objectives that are to be pursued
with the guiding principle of a feminist development
policy. Respondents primarily support the overarching
objectives of strengthening the “rights”, “resources” and
“representation”, whereas the provision of funds for projects
promoting equal rights receives less support. As can be seen
in Figure 34, rights, representation and resources (“the 3 Rs”)
meet with the highest level of support from both advocates
and opponents of feminist development policy. The specic
increase in funds for projects promoting equal rights receives
least support in all groups. This appears to contradict the fact
that the improved access to resources for women, girls and
disadvantaged groups receives the strongest endorsement
even by the opponents of feminist development policy. It is
possible that the provision of funds is rejected, while improved
access to other resources, for example by awarding land rights
or loans to women is supported.
Nevertheless, there are great discrepancies between advocates
and opponents of feminist development policy with regard to
the level of support for all the objectives. As expected, Figure 34
shows that opponents of feminist development policy support all
the objectives less, whereas supporters of such a policy show a
high level of support for all the objectives. The topics of “rights”,
“resources” and “representation” are only just below or on the
centre of the scale for opponents of feminist development policy,
but this support is still 2 scale-points lower than that shown
by advocates of feminist development policy. The discrepancy
is particularly great when it comes to the support shown by
opponents and advocates for the topics “institutional change
within German DC”, “formation of international alliances” and
“provision of funds for equal rights”. Here, the discrepancy is
almost 3 points on the scale; these objectives are clearly rejected
by opponents of the policy with a mean value of 3, whereas
advocates still strongly support them with a mean value of just
under 6. At this point, it should once again be noted that, as
illustrated in Figure 32, the group of supporters was considerably
bigger than the group of opponents in this survey too.

75Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 34 Support for the objectives by support for feminist development policy
Financial resources: increasing funds
for promo�ng equal rights
Transformation: dismantling power
structures and dependencies
Alliances: strengthening feminist
development policy
in interna�onal alliances
Institutional change: ins�tu�onal
changes in German DC
Data: using and analysing data
on the everyday reality of
women/girls/disadvantaged groups
Representation: strengthening the
representa�on of women/
disadvantaged groups
Rights: tackling disadvantages
and strengthening rights for
women/girls/disadvantaged groups
Resources: improved access to resources
and control over resources for
women/girls/disadvantaged groups
1
Low
234
567
High
Average support
Against feminist
development policy
Neither for
nor against
For feminist
development policy
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 2,000. The gure shows mean values with
95% condence intervals, broken down according to support for feminist development policy. For the breakdown, the seven-point response scale has been conated to
three categories: “against feminist development policy” (values 1-3), “neither for nor against” (value 4) and “ for feminist development policy” (values 5-7). Respondents
were supposed to indicate, on a scale from 1 (“don‘t support at all”) to 7 (“completely support”) the extent to which they support various objectives of a feminist
development policy. The value “4” corresponds to the centre of the scale (“neither support nor don’t support”).

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy76
Who supports feminist development policy?
Feminist development policy is particularly supported by
women, people with a high level of formal education and
respondents who place themselves on the left of the political
spectrum. Respondents who have a more positive attitude
towards feminism overall, are more interested in DC-related
topics and feel well-informed about DC-related topics are also
more likely to support feminist development policy. Respondents
from East Germany tend to support feminist development policy
less than those from West Germany, although this dierence
is not signicant.61 Weaker correlations are also found for age
and interest in politics. Older people tend to support feminist
development policy slightly less than younger people.62
Figure 35 Factors that predict support for feminist development policy
0.43 ***
-0.01 *
0.07
0.24 *
-0.21 ***
-0.20
Place of residence
(East Germany v.
West Germany)
Political ideology
(1 = left,
10 = right)
Formal education
(high v. low)
Formal education
(medium v. low)
Age
Gender
(female v.
male)
-1 0 1
Unstandardised
regression coefficients
Demographic factors
0.91 ***
-0.05 *
0.29 ***
0.06
-0.04
Populist attitudes
(higher values =
more populist attitudes)
Feeling informed
about DC
Interest in DC
General
interest in politics
Attitude towards feminism
(higher values =
more positive attitude)
-1 0 1
Unstandardised
regression coefficients
Attitudes and interests
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 2,000. Unstandardised regression coecients
of a multiple regression are shown. The dependent variable is support for feminist development policy. Two multiple regressions were calculated, in which suppor t
for feminist development policy was predicted rstly by demographic variables (on the left) and secondly by psychological factors (on the right). Positive coecients
(yellow) mean that the factor has a positive correlation with support for feminist development policy. Therefore, people with a high score in this characteristic are
more likely to support feminist development policy. Negative coecients (blue) mean that the factor has a negative correlation with support for feminist development
policy. Correspondingly, people with a high score in this characteristic show less suppor t for feminist development policy. For categorical predictors, the category
towhich the coecient applies is in brackets in each case. Coecient s marked with an asterisk are statistically signicant. *** p < 0.001, ** < 0.01, * < 0.05 .
61 In the analysis presented here, Berlin is encoded as “East Germany”. If the same analysis is conducted with Berlin encoded as West Germany, the relationship is signicant.
Peoplefrom East German states (not including Berlin) support feminist development policy signicantly less frequently than people from West German states (includingBerlin).
The results of this additional analysis are documented in Figure 42 in Section 2.3.4 of the online appendix.
62 Interestingly, when a number of predictors are included (see Figure 35), a negative correlation can be seen in the regression analysis between a general interest in politics and
support for feminist development policy. In the rst place, this suggests that respondents who are more interested in politics support feminist development policy slightly
less. When we look solely at the correlation between an interest in politics and support for feminist development policy, without taking other predictors into consideration,
this correlation is positive (Pearson’s r = 0.13; p < 0.001). This means that respondents who are more interested in politics in general are also more likely to support feminist
development policy. As there is a high correlation between the two forms of interest (Pearson’s r = 0.52; p < 0.001), the multiple regression analysis can accurately determine
the inuence of the individual variables. Therefore, it must be assumed that both forms of interest are associated with greater support for feminist development policy.

77Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Inuence of the feminist label and feminist
contents on support for development cooperation
The “feminist” label could have a negative eect on support
for development cooperation in the population. So far, results
have shown that, on the whole, a majority of respondents
supports a feminist development policy. However, a feminist
development policy is rated more negatively than other possible
focuses such as a “human rights-based” or “peace-promoting”
development policy. This could suggest that the label “feminist”
leads to negative reactions and that using it could reduce
support for development cooperation overall.
At the same time, certain contents of feminist development
policy could also adversely aect support for development
cooperation in parts of the population. In the question relating
to support for feminist development policy, the contents
“rights”, “representation” and “resources” are explained, and
these appear, on the whole, to receive support from a large
proportion of the population (see Figure 32 and Figure 34).
At the same time, a large discrepancy is evident in support
for the objectives of feminist development policy between
opponents and advocates of such a policy, especially with
regard to transformational objectives or nancial resources.
Therefore, specic feminist contents could also lead to a decline
in support for development cooperation in general when a
feminist development policy is pursued.
To test this hypothesis, an experiment was conducted in
which respondents were randomly presented with one of
four descriptions of a development policy, which either
carried a feminist label or described feminist contents.
Theywere then supposed to state how much they supported
this development policy. The four descriptions either contained
the term “development policy” (without an additional label) or
the term “feminist development policy” and a description that
either outlined development policy in general (without feminist
contents) or also specied feminist contents. When formulating
the contents, transformative objectives were also explicitly
included in the description of feminist development policy.63
This experiment can show whether the label “feminist” – in
comparison with no label – increases or decreases support for
development cooperation and whether feminist contents in DC
lead to more or less support than traditional DC contents.64
Contrary to expectations, no signicant dierences were
found between the dierent groups with regard to general
support for development policy. The average support
was lowest for “development policy” without feminist
contents (mean value = 4.4) and highest for “development
policy” without a feminist label but with feminist contents
(meanvalue = 4.6). Overall, dierences between the average
values are minimal and there are no statistically signicant
dierences.65
Regarding the question about expenditure on DC, greater
support can be seen if a development policy was previously
presented just with a feminist label (without feminist
contents) or just with feminist contents (without a feminist
label). If a “traditional” development policy (without
a feminist label or contents) or a development policy with
both a feminist label and feminist contents is presented,
the support for expenditure is lower. Respondents want to
cut funds for DC less if they previously read about a “feminist
development policy” without specic feminist contents or
about a “development policy” without this label but with
feminist contents than if they read about a “development
policy” without feminist contents or a “feminist development
policy” with feminist contents (see Figure 36). Concerning
these results, it is important to emphasise that this eect is
statistically signicant, but very small. The dierence between
the groups is just 0.1-0.2 points on the scale. Furthermore,
itis evident in accordance with the insights from Section 3.1.2
that the respondents want to reduce rather than increase the
funds for DC across all experimental groups.
63 The precise wording of the experimental texts is documented in Table 36 in Section 2.3.1 of the online appendix.
64 We recorded our hypotheses in advance in a pre-registration, which can be accessed at https://osf.io/x7h6y (for the background to pre-registrations, see Chapter 2).
65 The experiment was analysed with a variance analysis. The detailed results of these analyses are documented in Section 2.3.4 of the online appendix.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy78
This result could suggest that the feminist label and feminist
contents in DC are supported and are also rated more
positively than “traditional” DC, but that there could also be
“too much” or “too little” feminism. Overemphasising feminist
topics (as in the condition with a feminist label and feminist
contents) could potentially be perceived as “too much” and
thus be rated more negatively. At the same time, the absence
of a feminist perspective could be perceived as “too little”
and rated more negatively, as in the condition without
a feminist label and without feminist contents, for example.
Figure 36 Support for Development cooperation/ODA expenditure by experimental group
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in November 2023. N = 1,628 . The gure shows mean values with
95% condence intervals, broken down by experimental condition. The question was: “In 202 2, Germany spent approx. 33.3 billion euros on development cooperation.
Thiscorresponds to approximately 0 .8 percent of the gross national income (GNI). What should Germany do about this amount in the future? ” The response categories were:
“increase a great deal” ( 5), “increase somewhat” (4), “don’ t change” (3), “reduce somewhat ” (2) and “reduce a great deal” (1). The value “3” corresponds to the centre of the scale.
An interesting nding by Sassenhagen et al. (2023) was that
supporters of dierent parties showed greater dierences
in their attitudes towards feminist development policy than
in their attitudes towards other focuses of development
policy. This was interpreted to mean that the use of the label
“feminist” for the new focus of development policy could call
the previous broad public consensus regarding development
policy into question and lead to a more polarised debate (see
also Schneider et al., 2024b).
In the current experiment too, it is evident that the feminist
label can be a challenge for cross-party acceptance, whereas
feminist contents in development policy are more capable
of achieving a consensus across party lines. As can be seen
in Figure 37, the discrepancy in support for DC between
supporters of dierent parties is greater when the “feminist”
label is used than when this label is not used. If development
policy is labelled “feminist”, the dierence in support between
the highest (Alliance 90/The Greens) and the lowest (AfD)
values is 2.5 points on the scale on average. The greatest
consensus is shown for a development policy without the
“feminist” label but with feminist contents. Here, there are
just 1.5 points on the scale between the highest (Alliance 90/
The Greens) and the lowest (AfD) values. Even people who
identify with the AfD show most support for a development
policy with feminist contents (mean value = 3.8), but only if
it is not referred to as “feminist”. However, the average value
for AfD sympathisers is below the centre of the scale in all
experimental conditions, whereas the average values for all
those who identify with parties other than the AfD are above
the centre of the scale. It is also evident that the feminist label
leads to a greater rejection of the described development
policy particularly among people who identify with the FDP
or the CDU/CSU. On the other hand, it is primarily among
sympathisers of the Greens that the label leads to greater
support for the described development policy.
Increase 5
preference
Average
Reduce 1
2
Feminist content
No feminist content
Feminist label
No label
2,4 2,6 2,5 2,4
3
4

79Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 37 Support by experimental group and party identication
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in November 2023. N = 1,628 . The gure shows mean values
with 95% condence intervals, broken down by the experimental condition and the party identication. The question was: “To what ex tent do you support the
development policy described above?”. The scale ranged from 1 (“don’t support at all”) to 7 (“completely support”). The value “4” corresponds to the centre of the
scale(“neithersupport nor don’t support”).
No label -
No feminist content
No label -
Feminist content
“Feminist” label
No feminist content
“Feminist” label
Feminist content
1
Low
2 3 4 5 6 7
High
Average support
Average
AfD
CDU/CSU
FDP
SPD
Alliance 90/The Greens
The Le�
No party

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy80
4.4.3 How stable are the attitudes
towardsfeminist development policy?
Also with regard to attitudes towards feminist development
policy, the question arises of how stable these attitudes are.
On the one hand, an explicitly feminist development policy has
been formulated in Germany for the rst time. The previous
analyses suggest that the contents of this guiding principle are
unknown to a large proportion of the population. This could
indicate that attitudes towards feminist development policy
are still fragile and unstable. On the other hand, the analyses
showed that the feminist label led, to some extent, to polarised
attitudes, which could mean that attitudes towards feminist
development policy are very pronounced and therefore stable,
as they are possibly strongly rooted in beliefs.
To investigate how stable attitudes towards feminist
development policy are and how they dier from attitudes
towards DC in general, a survey was carried out with ten
characteristics of attitude strength. The measuring tool
explained in Section 3.4, which was used to determine the
strength of attitudes towards development policy, was
also used to measure attitude strength towards feminist
development policy. The ten66 characteristics of attitude
strength included attitude certainty, attitude ambivalence,
subjective accessibility for the assessment of the attitude,
personal relevance, interest, personal involvement, the
inuence of beliefs and moral convictions on the attitude, the
self-assessed extremity of the attitude and aective intensity
(see Figure 38; the precise wording of the questions relating to
the individual characteristics can be seen in Table 3 of Section
1.1.3 of the online appendix).67
When it comes to the topic of feminist development policy,
respondents are, on average, slightly more condent, have
less ambivalence and form their attitude more quickly than
when it comes to development policy in general, but feel
less aected by it. In Figure 38, it can be seen that the average
value for the characteristic of attitude certainty is higher (by
0.4 points on the scale) for feminist development policy than
for development policy without this label, but the average
values for ambivalence and for subjective accessibility are
lower (by0.4 and 0.2 points on the scale respectively). Overall,
therefore, the respondents appear to be slightly more condent
in their attitude towards feminist development policy than in
their attitude towards development policy in general. This is
also evident when examining the factors of attitude strength
(see also Section 3.4.1). As shown in Figure 39, 51 percent of
respondents feel very condent in their attitude towards
feminist development policy. This is 11 percentage points
more than for development policy without the feminist label.
This could be because the topic of feminism triggers stronger
reactions in the population than the topic of development
policy, and it is possible that more stable attitudes exist
towards feminism than towards development policy in general.
This greater attitude certainty leads to the conclusion that
attitudes towards this type of development policy may be less
susceptible to change by external inuences and that people
may persist more strongly in their opinion on this topic.
66 The characteristic “perceived knowledge” was not included for the analysis of attitude strength towards feminist development policy, in contrast to the analyses in Section
3.4, as only respondents who had stated that they had already heard of feminist development policy were asked about this characteristic. As a large proportion of respondents
therefore show a lack of data points for this characteristic, the characteristic was not included in the analysis.
67 Every person went through both the survey on attitude strength regarding development policy in general and the survey on attitude strength regarding feminist development
policy. To avoid question-order eects, the block of questions that was to be answered rst was randomly determined (see Oldendick, 2008; Schuman and Presser, 1996). The
results shown in this chapter are based exclusively on the half of the sample that completed the block on feminist development policy rst.

81Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 38 Characteristics and factors of attitude strength by type of development policy
Personal relevance of the attitude
Attitude certainty
Influence of beliefs on the attitude
1
Low
234567
High
Affective intensity
Influence of moral convictions
Extremity of the attitude
Influence of beliefs
Subjective accessibility*
Attitude ambivalence*
Attitude certainty
Personal involvement
Personal relevance
Interest in the topic
Average value
Characteristics of attitude strength
Development policy Feminist development policy
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly assigned
to a survey with questions on development policy in general (N = 2,041) or questions on feminist development policy (N =2,009) rst. The gure shows mean values with
95% condence intervals, broken down by type of development policy, for the people who answered the questions on the respective development policy rst. The response
options varied depending on the question, but consisted of a seven-point scale for all questions, where “1” represents lower values and “7” higher values. “4” corresponds to
the centre of the scale and a medium score. With regard to the individual characteristics, higher values are generally an indication of more stable attitudes. In the case of
the characteristics that are marked with * (“attitude ambivalence” and “subjective accessibility”), lower values are an indicator of more stable attitudes.
The fact that the respondents , on average, feel less affected
by feminist development policy than by development
policy without this label is partly because men in particular
feel that a feminist development policy addresses them
less than women. For example, an in-depth analysis
has shown that the personal involvement of men and women
does not differ when it comes to the topic of development
policy, butmen feel less personally affected when the same
development policy is referred to as “feminist”.68
68 The details of this analysis are documented in Table 48 in Section 2.3.5 of the online appendix.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy82
Figure 39 Factors of attitude strength by type of development policy (share of respondents)
11
55
34
9
71
20
2
58
40
11
68
20
13
56
31
1
47
51
Development policy
Feminist development policy
Influence of
beliefs
on the
attitude
Personal
relevance
of the topic
Attitude
certainty
Influence of
beliefs
on the
attitude
Personal
relevance
of the topic
Attitude
certainty
0
25
50
75
100
Percentage of respondents
High score
Medium score
Low score
Source: DEval, own visualisation. Source of the data: DEval survey by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly assigned to a survey
with questions on development policy in general (N = 2,041) or questions on feminist development policy (N =2,009) rst. In each case, the values are based solely on
the share of the respondents who answered the questions on the respective topic rst. The gure shows the three factors into which the individual characteristics have
been conated on the basis of the exploratory factor analysis, and the shares of the population that give a low (average value < 3), medium (average value 3– 5) or high
(average value > 5) score for these factors.
In order to understand how the dierent characteristics of
attitude strength relate to support for feminist development
policy, the respondents were segmented. As in Section 3.4.2,
a latent prole analysis (LPA; see Box 6) was used, which
included ten characteristics of attitude strength as well as
support for a feminist development policy. All the variables
included in the LPA are shown in Figure 41.
During the segmentation, four attitude types were identied:
(1)“stable supporters” (28%), (2) “ambivalent supporters” (16%),
(3) “unstable moderate supporters” (37%) and (4) “stable
opponents” (19%). The proportion of respondents assigned to
each of the attitude types is presented in Figure 40.69 The names of
the attitude types were chosen on the basis of the two dimensions
“support for feminist development policy” and “attitude strength”.
69 The LPA found that a solution with ve groups best describes the data. In a solution with ve groups, two attitude proles were identied for unstable moderate supporters,
which dier only slightly from one another both in their support for feminist development policy and in the score given to the various attitude characteristics. For the sake of
simplication, these two attitude proles have been conated into a single group in this report. The shares and the attitude proles, taking all ve groups into account, and
other details of the results of the latent prole analysis are documented in Section 2.3.5 of the online appendix.

83Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
The four attitude types with regard to feminist development
policy coincide with those that were determined for attitudes
towards development policy (see Section 3.4.2). This could
indicate that the attitude types found represent a general
pattern that emerges for attitude strength in relation to
attitudes towards DC. Even though the attitude types are based
on similar variables, the analyses were calculated independently
of one another. Therefore, the shares of the attitude types are
only comparable to a limited extent.
More than a quarter of respondents could be identied
as stable supporters of a feminist development policy;
more than half as unstable or ambivalent supporters.
Thus, there is a  core of stable supporters whose attitude is
unlikely to change as a result of external inuences and who
stand rmly behind a feminist development policy. At the same
time, a majority is undecided with regard to their attitudes
towards feminist development policy. In this group, support is
still unstable and can be changed more easily.
Figure 40 Share of respondents assigned to the four attitude types for feminist development policy (in percent)
28
16 37
19
Stable supporters
Ambivalent supporters
Unstable moderate supporters
Stable opponents
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly
assigned to a survey with questions on development policy in general (N = 2,041) or questions on feminist development policy (N =2,009) rst. The values are based on
the share of respondents who answered the questions about feminist development policy (N = 2,009) rst. The gure shows the shares for the four attitude types to
which the respondents were assigned by an LPA on the basis of their responses.
A fth of respondents reject a feminist development policy in
general and thus fall into the category of opponents. As can
be seen in Figure 41, the average value given by this group for
support for feminist development policy is clearly below the
centre of the scale. This contrasts with the analysis of attitude
types regarding development policy in general, in which even
the stable opponents had average support values near the
centre of the scale. Therefore, this nding too could suggest
that the label “feminist development policy” can lead to more
polarised attitudes.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy84
Figure 41 The proles of the four attitude types regarding feminist development policy
Extremity of the attitude
Attitude ambivalence*
Influence of moral convictions
Influence of beliefs
Feeling informed
Interest in feminist
development policy
Personal involvement
Personal relevance
Attitude certainty
Affective intensity
Subjective accessibility*
Support for feminist
development policy
1
Low
234567
High
Average
value
Stable
supporters:
28 %
Ambivalent
supporters:
16 %
Instable
moderate
supporters:
37 %
Stable
opponents:
19 %
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in August 2023. N = 4,050. The respondents were randomly
assigned to a survey with questions on development policy in general (N = 2,041) or questions on feminist development policy (N =2,009) rst. The gure shows mean
values for the dierent characteristics of attitude strength, broken down into the four attitude types that were identied by a latent prole analysis. The values are
based solely on the share of respondents who answered the questions about feminist development policy rst. The response options varied depending on the question,
but consisted of a seven-point scale for all questions, where “1” represents lower values and “7” higher values. “4” corresponds to the centre of the scale and a medium
score. With regard to the individual characteristics, higher values are generally an indication of more stable attitudes . In the case of the characteristics that are
marked with * (“attitude ambivalence” and “subjective accessibility”), lower values are an indicator of more stable attitudes.

85Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Opponents and unstable supporters of a feminist
development policy deviate, as expected, from stable
supporters in terms of their party preferences; other
dierences can be seen with regard to the sociodemographic
variables of gender, age and level of formal education.
An in-depth analysis was carried out to investigate
the relationship between group membership and the
characteristics of gender, age, formal level of education
and party preference. The comparison group is the group of
stable supporters. In this analysis, it was evident that stable
opponents tend to be male and between 30 and 59 years old.
Unstable supporters, on the other hand, do not dier from
stable supporters in terms of their age and gender. Both
opponents and unstable supporters tend to have a lower
level of formal education. With regard to party preferences,
the picture is as expected: both opponents and unstable
supporters identify less frequently with the Greens and more
frequently with the CDU, the AfD, other unspecied parties or
no party (incomparison with the SPD).70
4.4.4 Feminist development policy
intimesofglobal crises
The previous analyses show that support for a feminist
development policy depends on various personal factors.
Which label is selected and whether and how contents of the
policy are communicated is also important. Furthermore, the
analysis of attitude strength shows that, in addition to stable
supporters and stable opponents, there is also a large group of
people who are not certain of their attitude. Thus, the attitude
towards feminist development policy is probably still malleable
in a large section of the population. Whether this section of the
population supports or rejects a feminist development policy
in the long term will probably depend on how the contents and
objectives of this policy orientation are formulated.
One previously unheeded factor is that a feminist foreign
and development policy’s ability to deliver results is
increasingly called into question in times of acute crises
and that its implementation into political decisions and
measures is regarded with scepticism (Ganter and Stamm,
2022). For example, in a speech in March 2022, leader of
70 For the rationale behind these analyses, see footnote 26. The detailed results of this analysis are documented in Figure 30 in Section 2.3.5 of the online appendix.
the CDU Friedrich Merz indicated that a feminist foreign
and development policy conicts with expenditure on
military objectives in the context of the war against Ukraine
(GermanBundestag, 2022b). In addition, there are doubts as to
whether a feminist development policy really changes anything
and whether it is translated into concrete interventions
(e.g.OECD, 2023, p.97–99).
Another aspect that could have a negative eect on support
for a feminist development policy in times of global crises
is that people may weigh up dierent development policy
objectives against one another. In Section 3.3.3, for example,
it was demonstrated that the general public is more likely to
endorse a traditional approach to DC and is more likely to
support objectives from traditional DC sectors (education,
food, health care) than more progressive objectives such as
promoting women’s rights, equality or protecting minorities.
The BMZ’s feminist strategy is pursuing a holistic approach and
shows how close the links are between various elds of action
(for example gender equality and poverty) of DC. Within the
population, however, these relationships could be less visible or
could be viewed more critically, potentially leading to a trade-
o between the various elds of action.
A survey experiment was conducted to investigate the extent
to which feminist development policy is supported in times
of global crises and how dierent narratives can change this
support. First, the respondents received a text about current
global crises. Then they were randomly given a text in which
feminist development policy was justied on either normative
or instrumental grounds. All respondents received the same
text to begin with, which drew their attention to current global
challenges and crises such as the war of aggression against
Ukraine, the war in the Middle East, the Covid-19 pandemic,
climate change, and economic cut-backs and increased social
inequality in the world and in Germany. At the end of the text,
itsays “All these global crises also present challenges for German
development policy and development cooperation. The current
German Federal Government has shifted its focus regarding
development policy in recent years and is now pursuing afeminist
development policy.” After this introduction, the respondents
were randomly assigned to four experimental groups. The rst
group received additional information about the objectives and

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy86
contents of feminist development policy. This explained that
women’s rights and equality are key contents and objectives
of development policy, as they are ahuman right that has so
far not been achieved in any country in the world (normative
justication). The second group also received information about
feminist development policy. Women’s rights and equality were
described as key objectives and contents of development policy,
with the justication that the participation of women, girls and
marginalised groups is seen as a prerequisite for better and more
sustainable development results (instrumental justication).
Athird group received both sets of information about feminist
development policy. The control group received no additional
information beyond the introductory sentence. Then the
respondents had to indicate their general level of support for
feminist development policy.71
With this approach, the eect that dierent narratives
have on support for feminist development policy can be
investigated. It can also be determined whether a normative
and an instrumental justication of feminist development
policy have dierent eects, espe cially in the context of global
crises. For example, an instrumental justication could be more
likely to lead to people judging a feminist development policy to
be eective and capable of delivering results, even in times of
crises. In addition, the control group of the experimental design
can be used as an indicator for determining whether a feminist
development policy is assessed dierently in the context of
global crises and when these crises are not directly present.
Both a normative and an instrumental justication increased
support for feminist development policy in the context
of global crises in comparison with the control group.72
As can be seen in Figure 42, the average support for feminist
development policy is below the centre of the scale (“4”).
This means that this policy is rather rejected. Both the
presentation of an individual justication and the presentation
of both justications increased support for feminist
development policy by 0.7 points on the scale; with an average
value of 4.5, this was clearly above the centre of the scale. No
dierence in eect is evident for the dierent justications.
Even when both justications were presented, the support
did not increase further. This result suggests that additional
information about the objectives and motivations of feminist
development policy can increase support that has fallen in the
context of global challenges. Both instrumental and normative
justications can contribute to this; in the setting studied,
a combination is not more eective than a normative or an
instrumental justication alone.
The same eect can be seen for the perception of the
eectiveness of feminist development policy: if respondents
are given a more detailed explanation of feminist
development policy, they not only support it more, but
also class it as more eective. On a scale from 1 to 10, the
mean value for the assessment of the eectiveness of feminist
development policy without further justication is 4.5.
Withanormative justication or a combination of a normative
and an instrumental justication, this value increases by
0.7points on the scale to 5.2. An interesting observation is that
the eectiveness of feminist development policy is rated most
highly when only an instrumental justication is provided.
Here, the mean value is 5.4 (see Figure 31 in Section 2.3.6
of the online appendix).
71 The experiment was previously recorded in a pre-registration (see Chapter 2), which can be accessed at https://osf.io/37na9.
72 The experiment was analysed with a variance analysis. The detailed results of these analyses are documented in Section 2.3.6 of the online appendix.

87Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
Figure 42 Support for feminist development policy depending on the justication
3.7
4.5
4.5 4.5
No support 1
2
3
4
5
6
High support 7
No normative
justification
Normative
justification
Average
support
No instrumental
justification
Instrumental
justification
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in December 2023. N = 2,146 . The gure shows mean values with
95%condence intervals, broken down by experimental condition. The left-hand bar represents the control group and thus a feminist development policy without further
justication. The right-hand bar represent s feminist development policy with both a normative and an instrumental justication. The question was: “What is your opinion
about the current focus of the German Federal Government’s development policy, the so-called feminist development policy?” The response was provided on a seven-point
scale from 1 (“don’t support at all”) to 7 (“completely suppor t“). The value “4” corresponds to the centre of the scale (“neither support nor don’t support”).
The dierent justications of a feminist development policy
in this study had no inuence on the respondents’ opinions as
to whether the budget for DC should be increased or reduced
in general. This could lead to the conclusion that information
about feminist development policy has an eect primarily on
direct support for such a policy, but not on development policy
in general. Alternatively, this nding could also mean that the
presentation of a feminist development policy does have an
inuence on general support for development policy in the
context of global crises, but that even detailed justications
cannot change an individual attitude when it comes to the
distribution of scarce nancial resources.
Gender dierences
Even though women generally support feminist development
policy more than men in this experiment, everyone,
regardless of gender, is more likely to support a feminist
development policy if it is also justied.73 As can be seen in
Figure 43, the average support for feminist development policy
is 0.3 to 1.0 points on the scale higher for women than for men.
While both women and men in the control group show a level
of support below the centre of the scale, this support is on or
above the centre of the scale for all respondents, irrespective
of gender, in the conditions in which feminist development
policy is justied.
3.7
4.5
4.5 4.5
No support 1
2
3
4
5
6
High support 7
No normative
justification
Normative
justification
Average
support
No instrumental
justification
Instrumental
justification
73 Only people who identify either as “female” or male” were included in the analysis, as the group of people identifying as “non-binary” was too small for a meaningful analysis.
For a detailed explanation of how gender information was collected in the surveys behind this report, see Box 2.

Development policy attitudes in the context ofsecurity policy challenges and feminist development policy88
Figure 43 Support for feminist development policy by justication and gender
4.3
4.0
4.3
3.5
4.7
5.0
4.8
3.8
No support 1
2
3
4
5
6
High support 7
Normative and
instrumental
justification
Only
instrumental
justification
Only
normative
justification
No
justification
(control group)
Average
support
Male
Female
Source: DEval, own visualisation. Source of the data: DEval survey, carried out by Respondi/Bilendi in December 2023. N = 2,146 . The gure shows mean values
with 95% condence intervals, broken down by justication and gender. The question was: “What is your opinion about the current focus of the German Federal
Government’s development policy, the so-called feminist development policy?” The response was provided on a seven-point scale from 1 (“don’t support at all”) to 7
(“completely support”). The value “4” corresponds to the centre of the scale (“neither support nor don’t support”).
The dierence in support for feminist development policy
between women and men is particularly great if feminist
development policy is justied with instrumental objectives.
This is where the highest level of support from women and,
apart from the control condition, the lowest level of support
from men can be seen, leading to a dierence of a whole point
on the scale (see Figure 43).
As the focus of the instrumental justication for feminist
development policy was on the positive consequences of
women, girls and marginalised groups being decision-makers,
such wording could cause men to believe that the status
of their own group is threatened. This could, in turn, bring
about a greater rejection of feminist development policy. In
accordance with this hypothesis, experimental psychological
research shows, for example, that men support feminist
movements less if they feel that their own status is under threat
(Rivera-Rodriguez et al., 2022).
4.4.5 Conclusion: a challenging label with
contents that lend themselves to a consensus
The guiding principle of feminist development policy could
attract attention and bring development policy more
strongly into public debate. However, it is doubtful whether
this means that development policy will lose its status as a
niche topic. The term “feminist development policy” is gaining
prominence and is already better known than the term “SDGs”.
Thus, it has the potential to make development policy topics
more visible – even among people who may not previously
have engaged with these topics. At the same time, the term
appears to have failed to stimulate discourse on a wider scale in
the general public. For example, the information environment
on Twitter/X only changed selectively in the wake of the
presentation of the strategy by the BMZ. The topics of “gender
equality” and “representation of women” are mentioned by
respondents as contents of feminist development policy, but
specic contents of the strategy are less well known.

89Development policy attitudes in the context ofsecurity policy challenges and feminist development policy
The “feminist” label could contribute to a decline in support
for development policy, especially against the backdrop of
global crises, and the previous consensus on development
policy could dwindle as a result of a more polarised political
debate. For instance, a feminist development policy receives
less support than other guiding principles such as a “human
rights-based” or “peace-promoting” development policy. In
addition, the dierences between supporters of dierent
political parties in terms of their support for DC are greater
when the label “feminist” is chosen for development policy.
Furthermore, respondents tend to be more condent in their
attitudes towards feminist development policy than in their
attitudes towards development policy in general. This could
foster further political polarisation.
The contents of a feminist development policy appear to
lend themselves to consensus more than the corresponding
label and thus provide opportunities for development policy
communications. For example, there is more agreement
between supporters of dierent parties when it comes to
their support for a development policy with feminist contents
but without a feminist label. The objective of strengthening
the rights, representation and resources of women, girls and
marginalised groups is also largely endorsed by opponents
and supporters of feminist development policy. However,
discrepancies can be seen with regard to transformational
objectives or the redistribution of nancial resources. This also
makes it clear that support varies according to the contents.
Support for a feminist development policy will ultimately
depend on how it is implemented and communicated and on
what contents and narratives are brought to the fore.

5.
IMPLICATIONS

91Implications
Implication 1: Support for development cooperation is waning. In this situation, development policy actors should look into
possible changes to their strategies and design policy communication and education proactively to meet future challenges.
Support for DC has been declining substantially across the
entire political spectrum in Germany since the beginning of
2022; the consensus regarding development policy is eroding,
although it has not yet been broken. This development falls
in a period that is characterised by nancial burdens for the
government and the individual and, at least sporadically, by
high media attention for development policy. At the same time,
public discussion is increasingly polarised. This polarisation
does not only relate to the fringes of the political spectrum.
There are also clear divergences between supporters of the
government and the opposition and even within the current
governing parties – especially between supporters of the FDP
and those of the other coalition parties – when it comes to
attitudes towards development, foreign and security policy.
This poses challenges for development policy, both with regard
to its implementation and with respect to communications
about the policy area.
In relation to the design of development policy, the declining
support can be seen as an opportunity for development policy
actors to reect on their own actions and thus to consider
which objectives German DC would like to pursue, which tools
should be used and how to better monitor the achievement
of objectives and the eectiveness of the policy. Evaluations
and studies repeatedly point to challenges in development
policy. Key concepts include a geographical and sectoral
fragmentation of the portfolio, coordination and coherence
problems within DC in the narrower sense (in particular relating
to the coordination of technical and nancial cooperation) and
with other departments and international donors; diculties
associated with reviewing the achievement of objectives and
the eectiveness; and challenges when it comes to learning
from evidence (see, for example, OECD, 2021). Against the
backdrop of the persistently strained budgetary situation
and the associated debate on government expenditure, the
declining support should provide grounds for self-critically
questioning how DC funds can be used more eectively and
more eciently and to what extent more use could be made of
scientic evidence, in the spirit of a culture of learning, to push
forward a global sustainable development.
The situation can be used to prepare development
communications for future crises. In the past, development
policy communications faced the challenge that the attention
threshold for a public discussion was not generally reached.
Examples of this are the low level of familiarity with the SDGs
as well as the low level of knowledge about development policy
topics in the German population. In the future, however, it is
at least possible that negative news about DC in particular will
produce attention peaks – the discussions in winter 2023/2024
about cycle paths in Peru, training in positive masculinity in
Rwanda and DC with China and India provide a vivid illustration
of this. These challenges for communications can be anticipated
by development policy actors and accordingly translated into
forward-looking action. The what, how and why of DC and its
impact should be transparent and comprehensible to broader
parts of the population and be based on coherence between
dierent actors (BMZ, DC implementing organisations) in the
public sector. Last but not least, a selectively raised awareness
of the topics can also provide a chance to reach people who
do not otherwise deal with DC (for the limitations of these
communications, see Box 8).

Implications92
Implication 2: A large proportion of the population has moderately positive, but unstable attitudes towards development
policy. It makes sense for development policy actors to actively seek communication in order to maintain the existing
consensus regarding development policy.
The analyses of attitude strength indicate that a large section
of the population can potentially be reached by development
communications. While approx. 19percent can be characterised
as stable supporters and 18percent as stable opponents of DC,
the majority of respondents (58%) have a moderately positive,
but relatively unstable attitude towards development policy.
Ifthese people come into contact with information on the topic,
this could lead to them changing their attitudes. Depending on
the specic content of this information, this change may be
positive or negative. For development policy actors, thequestion
arises as to how best to operate in this environment?
One option is to wait until the economic and budgetary
situation improves again, the public attention to DC declines
and support for DC among the German population increases
again or the fragile basic consensus about DC is re-established
on the basis of unstable, but positive attitudes. The possibility
of this scenario occurring cannot be ruled out, but is not very
likely: the war against Ukraine, the situation in the Middle East,
other global challenges and trouble spots, and the ongoing
challenges for the German economy and the German federal
budget are likely to shape the political agenda in the short and
medium term. There are also the upcoming elections in 2024
and 2025 and the corresponding positions of the political actors.
In view of this situation, a return to a broad consensus based on
rather unstable attitudes is unlikely in the foreseeable future.
A second option for development policy actors is to actively
shape the public debate on the topic. This can be done with
more or less strongly polarising labels, but should build
upon contents that largely lend themselves to a consensus
at the same time. The results presented in Chapter 4.4 show
that the contents of a feminist development policy meet with
relatively broad support. The label “feminist development
policy” is associated with much more negative support values
among people who place themselves (more) on the right of
the political spectrum than is an alternative development
policy label with related contents (for example, “human
rights-based” or “peace-promoting”). At the same time, the
label “feminist development policy” is comparatively well
known, and therefore appears to generate a certain level of
attention. In communications about development policy,
appropriate consideration should be given to how attention
can be attracted to the policy area – on the premise that the
communications are also as broadly connective as possible.
This could take place via the contents of a feminist development
policy that are supported by the German population, for
example the “3 Rs” (resources, rights and representation) or
via objectives of DC that receive aparticularly high level of
support, such as improvements to the conditions in the areas
of WASH, education, health, or food and agriculture. In this
context, it should also be borne in mind that greater gender
equality can also have a positive eect on other areas (and
vice versa). An example of this is the signicance of women for
peace processes and generally for the establishment of more
peaceful societies (Hudson et al., 2020, UN Women, 2015). For
instance, studies suggest that peace treaties are longer lasting
if women are involved in negotiating the treaty (e.g. Adjei,
2019). This relationship also exists for the degree of equality of
women in the societies in question.

93Implications
Box 8 Limitations of development policy communications
Various psychological mechanisms can inuence the impact of information. These include the nding that people react
more strongly to negative information than to positive information (negativity bias; e.g. Soroka et al., 2019) and take a more
critical stance towards information that contradicts their own views or comes from people and organisations that they do
not like (motivated reasoning; Taber and Lodge, 2006). For development policy actors, this means that reaching the sections
of the German population that have a more critical attitude towards the Federal Government is particularly challenging in
anews landscape that is shaped by negative information about the Global South (Maurer and Reinemann, 2006, Chapter 4.3).
Againstthis backdrop, non-governmental DC actors and social media play an important role in communication.
Implication 3: The increasing interconnectedness between DC and geopolitical and security policy aspects is leading to new
areas of tension in development policy; here it is important to nd a new balance of interests and to communicate this
appropriately.
Experimental ndings have made it clear that strategic
or geopolitical and security policy factors are playing an
increasingly important role in which partner countries are
preferred for DC. This is evident with respect to (1) support
for Ukraine (and other countries in Eastern Europe and Central
Asia), (2) reservations about countries that have ties with Russia
and China and (3) the signicance of partner countries as energy
suppliers for Germany. The perception of threat is relevant to
development policy attitudes in this context. If the security policy
benet of DC is not suciently clear to the German population,
this could lead to a loss of support for DC. Particularly because
the policy areas cannot be viewed separately, several areas of
tension are arising for development policy actors:
1. Normative tensions: from a normative perspective,
geopolitical and security policy objectives can clash with
human rights obligations and objectives. A hypothetical
example is the case of a partner country that is important
to Germany on a geopolitical level systematically
violating the rights of women and LGBTIQ+ people.
2. Impact-related tensions: with regard to a conict between
the impacts of development policy and geopolitical
or security policy objectives, one area of tension
may lie in the fact that security policy considerations
(e.g.“Thesecurity of Germany is also being defended
in the Hindu Kush”; statement by former Defence
Minister Peter Struck on the deployment of the German
army in Afghanistan) contribute towards longer-term
development policy objectives being neglected and thus,
in the worst case scenario, failing in both target areas
(Hartmann etal., 2023). This may in particular be the
case if geopolitical and security policy aspects outweigh
considerations of eectiveness when DC resources are
allocated and the funds are not used where they can
achieve most. Analogously, a focus on geopolitical and
security policy objectives can also contribute towards
problems of corruption that hamper development being
exacerbated asa result of a more prominent role of
the military (e.g.Gupta et al., 2001). At the same time,
it is necessary to bear in mind that military operations
in fragile countries that are marked by conicts make
DC interventions possible in the rst place by creating
an acceptable level of security (for DC in fragile states,
seeFaust et al., 2023; Wencker andVerspohl, 2019).
3. Thematic and spatial tensions: the high value assigned
to geopolitical and security policy interests, could have
a negative eect both on DC sectors and on cooperation
with partners and regions with less direct relevance
togeopolitics and security policy. Support for Ukraine,
towhich a growing share of German DC resources are
being committed, can be cited to illustrate this. In the
medium term, this can also contribute to new geopolitical
and security policy challenges in other regions – especially
in view of the increasing activities of China (e.g. Dreher
et al., 2022), India (e.g. Fuchs and Vadlamannati, 2013)
and Russia (e.g. Gray, 2015) in the Global South.

Implications94
Development policy actors must nd a balanced approach in
these multiple areas of tension and communicate this to the
public. On the one hand, this relates to a convincing conceptual
delimitation of the policy areas that names interfaces
transparently, without also abandoning independent objectives
and normative obligations (Faust, 2024; Leininger and
Hornidge, 2024). On the other hand, it will be crucial to link
a comprehensive security concept that covers various aspects
of security (defence, preventing pandemics, energy security,
etc.) to development policy objectives, as denoted by the term
“integrated security” in the new National Security Strategy
(Federal Foreign Oce, 2023b).
Implication 4: The polarising eect of the term “feminist development policy” requires careful consideration of whether and how
the label is used; contents that are capable of achieving a consensus should be implemented and communicated to the public
The polarising eect of the term “feminist development policy”
presents challenges for development policy communications
and the existing consensus on development policy in the
German population. Feminist development policy receives
less support on average than other, related development
policy orientations; it is more polarising than development
policy without this label, and in the context of global crises it
is perceived as a less eective approach to development policy
and is less likely to be supported. However, the label is relatively
well known. It could attract public attention to development
policy and, at the very least, encourage intermittent attention
peaks. However, there is no evidence that the label helps to tap
into new target groups. Therefore, it could have contributed to
the declining support for development policy, especially in the
context of global crises, and towards the debate surrounding
development policy becoming polarised.
If the term “feminist development policy” is to be used, itis
important to weigh up what objectives and target groups
it is aiming to reach and whether the benets balance out
the costs. In certain groups of people, the feminist label can
stabilise support for DC, whereas it could lead other groups
of people that were previously more likely to endorse DC to
rethink this attitude and to reject at least the concrete form
of feminist development policy. In some groups of society,
a rejection could be determined less by the contents than by
the theoretical points of reference of a feminist development
policy. Examples of this are intersectional, post-colonial and
queer feminist approaches and approaches that are critical of
capitalism from the humanities and social sciences, which meet
with criticism in some sections of academia and the media and
in dierent political camps (see, for example, Chibber, 2013;
Elbe et al., 2022; Feddersen and Gessler, 2021; Mounk, 2023).
Arejection could, on the other hand, also occur because of the
overriding associations that the term “feminism” triggers for
some people. Increasingly powerful networks of political actors,
think tanks, media platforms and campaign organisations, which
pick up on and foster such a sentiment, are potentially relevant
here. However, there is still the question of how advocates of
the current feminist development policy would react to the
label being abolished and whether this could lead to a sense
of disappointment in people who are traditionally strong
supporters of DC (Aspington and Shekh Mohamed, 2024).
Contents of feminist development policy that are capable of
achieving consensus should reveal their impact when they are
actually put into practice. Development policy actors should
emphasise the contents and impact of feminist development
policy in their communications and clearly explain the focus
on feminist topics. The overriding objective of strengthening
the rights, resources and representation of women, girls and
marginalised groups is supported by both opponents and
advocates of feminist development policy. And even if women’s
rights and equality were not the top priority as objectives of
German DC, they did not come far behind WASH, education,
health care, and food and agriculture. Therefore, they were
more likely to be endorsed than protecting the climate and
the environment, promoting the economy and promoting
democracy and the rule of law, for example. This shows that the
contents of feminist development policy are supported by many
respondents and are regarded as an important element of DC.
In particular, the fact that supporters of dierent parties agree

95Implications
that a feminist development policy without afeminist label but
with feminist contents is worth supporting demonstrates that
a social consensus has been reached regarding key contents,
such as gender equality. Furthermore, in the context of global
crises, respondents were more likely to support a feminist
development policy if this was justied on normative or
instrumental grounds than if it was presented without further
explanation. From this perspective, it makes sense for the
contents of a feminist development policy to be put into practice
and to take eect. Their implementation and impact should
also be communicated to the general public in an evidence-
based manner. This could help to make the relevance of more
consensual contents even more visible in practice (Oce of the
Auditor General of Canada, 2023; Towns et al., 2023).

6.
LITERATURE

97Summary
Abi-Hassan, S. (2017), “Populism and Gender”, in Kaltwasser,
C. R. et al. (ed.), The Oxford Handbook of Populism,
OxfordUniversity Press, Oxford, p. 426–444.
ACLED (2024), ACLED Conict Index: Januar 2024,
https://acleddata.com/conict-index/#downloads
(accessed on 13 May 2024).
Adjei, M. (2019), “Women’s participation in peace processes:
areview of literature.”, Journal of Peace Education, Vol. 16,
Nr. 2, p. 133–154.
Aggestam, K. et al. (2019), “Theorising Feminist Foreign
Policy”, International Relations, vol. 33, No. 1, p. 23–39.
Andress, H. J. et al. (2013), Applied Panel Data Analysis for
Economic and Social Surveys, Springer, Berlin/Heidelberg.
Ansolabehere, S. et al. (2008), “The Strength of Issues:
Using Multiple Measures to Gauge Preference Stability,
Ideological Constraint, and Issue Voting”, American Political
Science Review, vol. 102, No. 2, p. 215–232.
Aspington, C. and I. Shekh Mohamed (2024), Responses to the
Abolition of the Swedish Feminist Foreign Policy: Insights from
document analysis and interviews with Swedish civil society
organisations (dissertation), https://urn.kb.se/resolve?
urn=urn:nbn:se:mau:diva-68579 (accessed on 13 May 2024).
Backhaus, K. et al. (2018), Multivariate Analysemethoden:
Eineanwendungsorientierte Einführung, 15th edition,
Springer, Berlin/Dordrecht/London/New York.
Baker, R. et al. (2010), “Research Synthesis: AAPOR Report
onOnline Panels”, Public Opinion Quarterly, vol. 74, No. 4,
p. 711–781.
Bandura, A. (1982), “Self-Ecacy Mechanism in Human
Agency”, American Psychologist, vol. 37, No. 2, p. 122.
Bansak, K. et al. (2016), “How Economic, Humanitarian,
andReligious Concerns Shape European Attitudes toward
Asylum Seekers”, Science, vol. 354, No. 6309, p. 217–222.
Bassili, J. N. (2008), “Attitude Strength”, in Crano,
W.D.andR.Prislin (ed.), Attitudes and attitude change,
Psychology Press, New York, p. 237–260.
Bauer, P. C. et al. (2017), “Is the Left-Right Scale a Valid Measure
of Ideology?”, Political Behavior, vol. 39, No. 3, p. 553–583.
Bauhr, M. et al. (2013), “Does Corruption Cause Aid Fatigue?
Public Opinion and the Aid-Corruption Paradox”,
International Studies Quarterly, vol. 57, No. 3, p. 568–579.
Bayram, A. B. et al. (2024), “Manipulating the masses?
Right-wing populist messages, political ideology,
andpublic support for multilateral development aid in
Germany, the United States, and the United Kingdom”,
JournalofInternational Development, 1–20.
https://doi.org/10.1002/jid.3927
Bayram, A. B. and C. P. Thomson (2022), “Ignoring the
Messenger? Limits of Populist Rhetoric on Public Support
for Foreign Development Aid”, International Studies
Quarterly, vol. 66, No. 1, sqab041.
Bryman, A. (2016), Social Research Methods,
OxfordUniversity Press, Oxford.
BDI (2024), Grundsatzpapier. Zeit für eine entwicklungs-
politische Zeitenwende: Strategisch – Nachhaltig – Innovativ,
Federation of German Industries, Berlin, https://bdi.eu/
publikation/news/zeit-fuer-eine-entwicklungspolitische-
zeitenwende (accessed on 13 May 2024).
Bertelsmann Stiftung (2024), Transformation Index BTI 2024:
Governance in International Comparison, Gütersloh.
Blumenau, B. (2022), “Breaking with Convention? Zeitenwende
and the Traditional Pillars of German Foreign Policy”,
International Aairs, vol. 98, No. 6, p. 1895–1913.
Blackman, A. D. (2018), “Religion and Foreign Aid”,
Politics and Religion, vol. 11, No. 3, p. 522–552.

Literature 98
BMFSFJ (2021), PEKING +25: Bericht der Bundesregierung
der Bundesrepublik Deutschland zur Umsetzung der
Pekinger Erklärung und Aktionsplattform: 25. Jahrestag der
vierten Weltfrauenkonferenz und der Verabschiedung der
Pekinger Erklärung und Aktionsplattform (1995), Federal
Ministry for Family Aairs, Senior Citizens, Women and
Youth (BMFSFJ), Berlin, https://www.bmfs.de/resource/
blob/160870/4d2e3cc99214a5708b338d785387e7d9/
peking-25-bericht-data.pdf (accessed on 13 May 2024).
BMZ (2020), Länder, BMZ and Development (BMZ), Bonn/Berlin,
https://www.bmz.de/de/laender (accessed on 13 May 2024).
BMZ (2022), Das erste Jahr Entwicklungspolitik in der
20. Legislaturperiode, Federal Ministry for Economic
Cooperation and Development (BMZ), Bonn/Berlin,
https://www.bmz.de/resource/blob/132878/bmz-jahresbilanz-
2022-nal.pdf (accessed on 15 February 2023).
BMZ (2023a), Feministische Entwicklungspolitik: Für gerechte
und starke Gesellschaften weltweit, Federal Ministry
forEconomic Cooperation and Development (BMZ),
Bonn/Berlin, https://www.bmz.de/resource/blob/146200/
strategie-feministische-entwicklungspolitik.pdf
(accessed on 15 February 2023).
BMZ (2023b), Engagement weltweit: Strategie zur Förderung
bürgerschaftlichen Engagements in der Entwicklungs-
zusammenarbeit in und aus Deutschland, Federal Ministry
for Economic Cooperation and Development (BMZ),
Bonn/Berlin, https://www.bmz.de/resource/blob/154810/
engagement-weltweit.pdf (accessed on 13 May 2024).
BMZ (2023c), Dritter entwicklungspolitischer Aktionsplan zur
Gleichstellung der Geschlechter (2023–27), FederalMinistry
for Economic Cooperation andDevelopment (BMZ), Bonn/
Berlin, https://www.bmz.de/resource/blob/196130/dritter-
entwicklungspolitischer-aktionsplan-zur-gleichstellung-der-
geschlechter.pdf (accessed on 13 May 2024).
BMZ (2024), Deutschland braucht den Sahel: Warum Entwick-
lungspolitik auch eine Frage der Sicherheit ist, 5 March 2024,
Federal Ministry for Economic Cooperation and Development
(BMZ), Bonn/Berlin, https://www.bmz.de/de/aktuelles/reden/
ministerin-svenja-schulze/gastbeitrag-schulze-sahel-frankfurter-
rundschau-205566 (accessed on 13 May 2024).
Bodenstein, T. and J. Faust (2017), “Who Cares? European
Public Opinion on Foreign Aid and Political Conditionality”,
Journalof Common Market Studies, vol. 55, No. 5, p. 955–973.
Bohnet, M. (2019), Geschichte der deutschen Entwicklungs-
politik: Strategien, Innenansichten, Zeitzeugen,
Herausforderungen, 2nd revised and enlarged edition,
utbGmbH, Stuttgart.
Bosker, R. and T. A. Snijders (2012), Multilevel Analysis:
AnIntroduction to Basic and Advanced Multilevel Modeling,
2nd edition, Sage Publications Ltd., Los Angeles.
Bruder, M. et al. (2020), “Using Mediation Analysis to
Uncover Psychological Mechanisms of Attitude Change
ina Development Volunteer Program”, New Directions
forEvaluation, vol. 2020, No. 167, p. 131–143.
Buchsteiner, J. (2023), “Die Fragwürdigkeit des
deutschenGeldregens”, FAZ.net, 22 December 2023,
https://www.faz.net/aktuell/politik/inland/deutsche-helfer-
in-not-19403517.html (accessed on 28 February 2024).
Bundeswehr (2024), Weltweit gefordert: Die Bundeswehr
im Einsatz, https://www.bundeswehr.de/de/einsaetze-
bundeswehr (accessed on 28 February 2024).
Carr, D. F. (2023), One Year Into Musk’s Ownership, X (Twitter)
Down By Every Measure, similarweb blog, 17 October 2023,
https://www.similarweb.com/blog/insights/social-media-
news/x-twitter-musk/ (accessed on 12 June 2024).
Chibber, V. (2013), Postcolonial Theory and the Specter of Capital,
Verso, New York.
Conner, M. et al. (2022), “Testing Predictors of Attitude Strength
as Determinants of Attitude Stability and Attitude-Behaviour
Relationships: A Multi-Behaviour Study”, European Journal
ofSocial Psychology, vol. 52, No.4, p. 656–668.

99Literature
Cornesse, C. et al. (2020), “A Review of Conceptual
Approaches and Empirical Evidence on Probability and
Nonprobability Sample Survey Research”, Journal of
SurveyStatistics and Methodology, vol. 8, No. 1, p. 4–36.
Cunningham, H. et al. (2017), “Bilateral Foreign Aid:
HowImportant is Aid Eectiveness to People for
ChoosingCountries to Support?”, Applied Economics Letters,
vol. 24, No. 5, p. 306–310.
De Coninck, D. (2023), “The Refugee Paradox During Wartime
in Europe: How Ukrainian and Afghan Refugees are (not)
Alike”, International Migration Review, 57(2), p. 578–586,
https://doi.org/10.1177/01979183221116874.
Destatis (2024), Inationsrate im Jahr 2023 bei +5,9 %,
PressRelease No. 020 dated 16 January 2024,
https://www.destatis.de/DE/Presse/Pressemitteilungen/
2024/01/PD24_020_611.html (accessed on 25 April 2024).
Doherty, D. et al. (2020), “The Public’s Foreign Aid Priorities:
Evidence from a Conjoint Experiment”, American Politics
Research, vol. 48, No. 5, p. 635–648.
Dreher, A. et al. (2012), “Financial Dependence and Aid
Allocation by Swiss NGOs: A Panel Tobit Analysis”, Economic
Development and Cultural Change, vol. 60, No. 4, p. 829–867.
Dreher, A., et al. (2022), Banking on Beijing: The Aims and
Impacts of China’s Overseas Development Program.
Cambridge University Press.
Eagly, A. H. and S. Chaiken (1995), “Attitude Strength,
AttitudeStructure, and Resistance to Change”, in PettyR.E.
andJ.A. Krosnick (ed.), Attitude Strength: Antecedents
andConsequences. Psychology Press, New York, vol. 4.,
No.2, p.413–432.
Elbe, I. et al. (ed.) (2022), Probleme des Antirassismus.
Postkoloniale Studien, Critical Whiteness und Intersek-
tionalitätsforschung in der Kritik, Edition Tiamat, Berlin.
Erikson, R. S. and K. L. Tedin (2011), American Public Opinion,
Longman, Boston, p. 72–73.
Esposito, A. (2022), “The Limitations of Humanity: Dierential
Refugee Treatment in the EU”, 14 September 2022, Harvard
International Review, https://hir.harvard.edu/the-limitations-
of-humanity-dierential-refugee-treatment-in-the-eu/
(accessed on 13 May 2024).
Eurobarometer (2023), Special Eurobarometer 537 onEU citizens
and development cooperation, EuropeanCommission,
Brussels, https://europa.eu/eurobarometer/api/deliverable/
download/le?deliverableId=88756 (accessed on 13 May 2024).
Eger, J. et al. (2023), “Does Evidence Matter? The Impact
of Evidence Regarding Aid Eectiveness on Attitudes
TowardsAid”, European Journal of Development Research,
vol.35, No. 5, p.1149–1172.
Evans, G. et al. (1996), “Measuring left-right and libertarian-
authoritarian values in the British electorate”,
BritishJournal of Sociology, vol. 47, No. 1, p. 93–112.
Faltas, C. et al. (2024), Perspektiven auf Entwicklungspolitik:
Qualitative und quantitative Forschung zur Verankerung
und Kommunikation entwicklungspolitischer Themen und
Herausforderungen, Pollytix, Berlin, https://pollytix.de/
wp-content/uploads/2024/03/pollytix_Perspektiven-auf-
Entwicklungspolitik.pdf (accessed on 13 May 2024).
Faust, J. (2024), “Nicht ob, sondern wie”, 26 February 2024,
Internationale Politik, p. 100–104, https://internationalepolitik.
de/de/nicht-ob-sondern-wie (accessed on 13 May 2024).
Faust, J. et al. (ed.) (2023), Under challanging Conditions.
Development Coorporation in fragile States, FocusReport,
German Institute for Development Evaluation (DEval), Bonn.
Feddersen,J. and P. Gessler (2021), Kampf der Identitäten: Für
eine Rückbesinnung auf linke Ideale, Ch. Links Verlag, Berlin.
Federal Constitutional Court (2023), Zweites Nachtragshaus-
haltsgesetz 2021 ist nichtig, Press Release No. 101/2023 of
15November 2023, Federal Constitutional Court, Karlsruhe,
https://www.bundesverfassungsgericht.de/SharedDocs/
Pressemitteilungen/EN/2023/bvg23-101.html
(accessed on 13 May 2024).

Literature 100
Federal Foreign Oce (2023a), Feministische Außenpolitik
gestalten: Leitlinien des Auswärtigen Amts, Federal
Foreign Oce, Berlin, https://www.auswaertiges-amt.de/
blob/2585008/d444590d5a7741acc6e37a142959170e/
ll-p-data.pdf (accessed on 13 May 2024).
Federal Foreign Oce (2023b), Wehrhaft. Resilient.
Nachhaltig. Integrierte Sicherheit für Deutschland:
Nationale Sicherheitsstrategie, Federal Foreign Oce,
Berlin, https://www.nationalesicherheitsstrategie.de/
Sicherheitsstrategie-DE.pdf (accessed on 13 May 2024).
Federal Government Commission on the Root Causes of
Displacement (2021), Krisen vorbeugen, Perspektiven
schaen, Menschen schützen: Bericht der Fachkommission
Fluchtursachen der Bundesregierung, Commission on the
Root Causes of Displacement, Berlin, https://www.bmz.de/
resource/blob/75434/bericht-fachkommission-uchtursachen.
pdf (accessed on 13 May 2024).
Feeny, S. et al. (2019), “Donor Motives, Public Preferences
and the Allocation of UK Foreign Aid: A Discrete Choice
Experiment Approach”, Review of World Economics, vol.155,
No. 3, p. 511–537.
Focus online (2024), Die Milliarden-Liste: Lesen Sie, in welche
Winkel der Welt deutsches Steuergeld ießt, 17February
2024, https://www.focus.de/nanzen/steuern/nanzhilfen-
fuer-indien-peru-co-die-milliarden-liste-lesen-sie-in-welche-
winkel-der-welt-steuergeld-iesst_id_259654766.html
(accessedon21 February 2024).
Forschungsgruppe Wahlen (2024), Politbarometer: Langzeit-
entwicklung wichtiger Trends aus dem Politbarometer
zuwirtschaftlichen Themen. As at 15 August 2024,
https://www.forschungsgruppe.de/Umfragen/Politbarometer/
Langzeitentwicklung_-_Themen_im_Ueberblick/Wirtschaft_
und_Soziales/ (accessed on 22 August 2024)
Fransman, J. and H.-B. Solignac Lecomte (2004),
MobilisingPublic Opinion Against Global Poverty.
OECDDevelopment Centre Policy Insights No. 2, Paris.
Freedom House (2023), Freedom in the World 2023: Marking50
Years in the Struggle for Democracy, FreedomHouse,
Washington, D.C.
Fuchs, A. and Vadlamannati, K. C. (2013), “The NeedyDonor:
An Empirical Analysis of India’s Aid Motives”,
WorldDevelopment, vol. 44, p. 110–128.
Gaines, B. J. et al. (2007), “The Logic of the Survey Experiment
Reexamined”, Political Analysis, vol. 15, No. 1, p. 1–20.
Ganter, J. and L. Stamm (2022), “We Need to Talk:
Communicating Feminist Foreign Policy”, 31 October 2022,
Internationale Politik Quarterly, https://ip-quarterly.com/
en/we-need-talk-communicating-feminist-foreign-policy
(accessed on 13 May 2024).
German Bundestag (2019), Kleine Anfrage der Abgeordneten
Dr. Christoph Homann, Alexander Graf Lambsdor,
TillMansmann, Olaf in der Beek […] und der Fraktion der
FDP: Die deutsche Entwicklungszusammenarbeit und
Brasiliens Waldächenbilanz, 11 April 2019, printed paper
19/9315, https://dserver.bundestag.de/btd/19/093/1909315.pdf
(accessed on 21 February 2024).
German Bundestag (2020), Kleine Anfrage der
Abgeordneten Markus Frohnmaier, Dietmar Friedho
und der Fraktion der AfD: Partnerstaaten der deutschen
Entwicklungszusammenarbeit im Jahr 2020, Printed
Paper 19/20423, https://dserver.bundestag.de/
btd/19/204/1920423.pdf (accessed on 21 February 2024).
German Bundestag (2022a), Kleine Anfrage der Abgeordneten
Markus Frohnmaier, Dietmar Friedho, Edgar Naujok,
Stefan Keuter, Dr. Malte Kaufmann und der Fraktion der
AfD: Deutsche Entwicklungszusammenarbeit mit und in der
Ukraine, Printed Paper 20/2042, https://dserver.bundestag.
de/btd/20/020/2002042.pdf (accessed on 21 February 2024).
German Bundestag (2022b), Plenarprotokoll 20/24,
23.März2022, https://dserver.bundestag.de/
btp/20/20024.pdf (accessed on 21 February 2024).

101Literature
German Bundestag (2023), Kleine Anfrage der Abgeordneten
Cornelia Möhring, Anke Domscheit-Berg, Andrej Hunko,
Kathrin Vogler und der Fraktion DIE LINKE: Vertreibung der
Massai in Tansania und die Konsequenzen für die deutsche
Entwicklungszusammenarbeit, Printed Paper 20/6661,
https://dserver.bundestag.de/btd/20/066/2006661.pdf
(accessed on 21 February 2024).
German Federal Government (2022), Reden zur Zeiten-
wende: Bundeskanzler Olaf Scholz, enlarged 2nd edition,
October2022, Press and Information Oce of the
GermanFederal Government, Berlin, https://www.
bundesregierung.de/resource/blob/975292/2138164/
1dd57df3a594489f4c23bdbb7ad3efdc/bundeskanzler-olaf-
scholz-reden-zur-zeitenwende-2-au-download-bpa-data.
pdf?download=1 (accessed on 13 May 2024).
Goodman, S. (2008), “A Dirty Dozen: Twelve P-Value Miscon-
ceptions”, Seminars in Hematology, vol. 45, No. 3, p. 135–140.
Graf, T. (2022), Zeitenwende im sicherheits- und verteidigungs-
politischen Meinungsbild: Ergebnisse der ZMSBw-Bevölkerungs-
befragung 2022, Research Report 133, Bundeswehr Centre
ofMilitary History and Social Sciences, Potsdam.
Graf, T. (2024), Was bleibt von der Zeitenwende in den Köpfen?
Sicherheits- und verteidigungspolitisches Meinungsbild in
der Bundesrepublik Deutschland 2023, Research Report 136,
Bundeswehr Centre of Military History and Social Sciences,
Potsdam.
Gravelle, T. B. et al. (2017), “The Structure of Foreign Policy
Attitudes in Transatlantic Perspective: Comparing the United
States, United Kingdom, France and Germany”, European
Journal of Political Research, vol. 56, No. 4, p. 757–776.
Gray, P. (2015), “Russia as a Recruited Development Donor”,
European Journal of Development Research, vol. 27, No. 2,
p.273–288.
Grävingholt, J. et al. (2023), Wiederauau in der Ukraine:
Wasdie internationale Gemeinschaft jetzt beachtenmuss,
IDOS Policy Brief 2/2023, German Institute of
Development and Sustainability (IDOS), Bonn.
Grépin, K. A. et al. (2018), “Allocating External Financing
forHealth: A Discrete Choice Experiment of Stakeholder
Preferences”, Health Policy and Planning, vol. 33, p. i24–i30.
Gupta, S. et al. (2001), “Corruption and Military Spending”,
European Journal of Political Economy, vol. 17, No. 4, p. 749–777.
Hainmueller, J. et al. (2014), “Causal Inference in Conjoint
Analysis: Understanding Multidimensional Choices
viaStated Preference Experiments”, Political Analysis,
vol.22, No. 1, p. 1–30.
Hansen, P. et al. (2014), “Developing Countries in Need:
Which Characteristics Appeal Most to People When
Donating Money?”, Journal of Development Economics,
vol.50, No. 11, p. 1494–1509.
Hartmann, C. et al. (2023), Ressortgemeinsame strategische
Evaluierung des zivilen Engagements der Bundesregierung in
Afghanistan. Ressortspezischer Bericht zum Engagement des
BMZ in Afghanistan, German Institute forDevelopment
Evaluation (DEval), Bonn.
Heath, A. et al. (1994), “The Measurement of Core Beliefs
andValues: The Development of Balanced Socialist/
LaissezFaire and Libertarian/Authoritarian Scales”,
BritishJournal of Political Science, vol. 24, No. 1, p. 115–132.
Heinrich, T. et al. (2016), “Public Opinion and Foreign Aid Cuts
in Economic Crises”, World Development, vol. 77, p. 66–79.
Heinzel, M. et al. (2024), “Transparency and Citizen Support for
Public Agencies: The Case of Foreign Aid”, Governance, p. 1–22.
Herold, M. et al. (2023), Polarisierung in Deutschland und Europa.
Eine Studie zu gesellschaftlichen Spaltungs tendenzen in zehn
europäischen Ländern, MIDEM study 2023–2, Mercator
Forum for Migration and Democracy (MIDEM), Dresden.
HIIK (2023), Conict Barometer 2022: Disputes, Non-violent
Crises, Violent Crises, Limited Wars, Wars, Heidelberg
Institute for International Conict Research (HIIK),
Heidelberg.

Literature 102
Holbrook, A. L. and J. A. Krosnick (2010), “Social Desirability
Bias in Voter Turnout Reports Tests Using the Item Count
Technique”, Public Opinion Quarterly, vol. 74, p. 37–67.
Holsti, O. R. and J. N. Rosenau (1990), “The Structure of
Foreign Policy Attitudes among American Leaders”,
TheJournal of Politics, vol. 52, No. 1, p. 94–125.
Howe, L. C. and J. A. Krosnick (2017), “Attitude Strength”,
Annual Review of Psychology,vol. 68, No. 1, p. 327–351.
Hudson, V. et al. (2020), The First Political Order: How Sex
Shapes Governance and National Security Worldwide,
Columbia University Press, New York.
Hudson, J. et al. (2023), Understanding AttitudesTowards
Feminist Foreign Policy & Gender Inequality,
DevelopmentEngagement Lab, London.
Hurwitz, J. and M. Peey (1987), “How are Foreign Policy
Attitudes Structured? A Hierarchical Model”, American
Political Science Review, vol. 81, No. 4, p. 1099–1120.
ifo (2022), Geopolitische Herausforderungen und ihre Folgen für
das deutsche Wirtschaftsmodell, study report, as at August
2022, ifo Institute, Munich, https://www.ifo.de/DocDL/
Studie_Geopolitische_Herausforderungen_Folgen_deutsches_
Wirtschaftsmodell.pdf (accessed on 13 May 2024).
IPBES (2019), “The Global Assessment Report on Biodiversity
and Ecosystem Services of the Intergovernmental Science-
Policy Platform on Biodiversity and Ecosystem Services”,
Bonn, https://doi.org/10.5281/zenodo.3831673.
IPCC (2023), “Climate Change 2023”, in Lee, H. and Romero,
J.(ed.), Synthesis Report. Contribution of Working
Groups I, IIand III to the Sixth Assessment Report of the
Intergovernmental Panel on Climate Change, Geneva,
Switzerland, p. 1–184, doi: 10.59327/IPCC/AR6-9789291691647.
Ipsos (2021), Weltfrauentag: Jeder dritte Mann glaubt, dass
Feminismus mehr schadet als nützt und die traditionelle
Männlichkeit bedroht ist, 5 March 2021, Ipsos, Hamburg,
https://www.ipsos.com/de-de/weltfrauentag-jeder-dritte-
mann-glaubt-dass-feminismus-mehr-schadet-als-nutzt-und-
die-traditionelle (accessed on 13 May 2024).
Jankowski, M. et al. (2019), “To adapt or to defend? Comparing
position shifts among Bundestag candidates between 2013
and 2017”, West European Politics, vol. 42, No. 4, p. 895–913.
Jankowski, M. et al. (2023), “How Stable are ‘Left’ and ‘Right’?
AMorphological Analysis using Open-Ended Survey
Responses of Parliamentary Candidates“, Party Politics,
vol.29, No. 1, p.26–39.
Jerit, J. and J. Barabas (2023), “Are Nonprobability Surveys Fit for
Purpose?”, Public Opinion Quarterly, vol. 87, No. 3, p.816–840.
Jezierska, K. (2022), “Incredibly Loud and Extremely Silent:
Feminist Foreign Policy on Twitter”, Cooperation and
Conict, vol. 57, No. 1, p. 84–107.
Jungherr, A. (2019), “Kommunikation auf sozialen
Netzwerkplattformen”, in Faas, T. et al. (ed.), Einstellungs-
und Verhaltensforschung: Handbuch für Wissenschaft und
Studium, Nomos, Baden-Baden, p. 184–206.
Kantola, J. and E. Lombardo (2019), “Populism and Feminist
Politics: The Cases of Finland and Spain”, European
Journalof Political Research, vol. 58, No. 4, p. 1108-1128.
Khillare, T. (2023), The Feminist Foreign Policy Agenda: Whatis
It and Why Should India Engage with It?, HeinrichBöll
Foundation, Regional Oce New Delhi, India.
Kobayashi, Y. et al. (2021), “Public Support for Development
Aid during the COVID-19 Pandemic”, World Development,
vol.138, p. 1–11.
Koch, W. (2023), “Ergebnisse der ARD/ZDF-Onlinestudie2023.
Soziale Medien werden 30 Minuten am Tag genutzt –
Instagram ist die Plattform Nummer eins”,
MediaPerspektiven, vol. 2023, No. 26, p. 1–8.

103Literature
Körber Foundation (2022), The Berlin Pulse: Rethinking Security
for Germany and Europe, Körber Foundation, Berlin.
Körber Foundation (2023), The Berlin Pulse: Of Paradigms
andPower Shifts, Körber Foundation, Berlin.
Kogut, T. et al. (2018), “How Far is the Suering? The Role
of Psychological Distance and Victims’ Identiability in
Donation Decisions”, Judgment and Decision Making, vol.13,
No. 5, p. 458–466.
Kohler, U. and J. C. Post (2023), “Welcher Zweck heiligt die
Mittel? Bemerkungen zur Repräsentativitätsdebatte in der
Meinungsforschung”, Zeitschrift für Soziologie, vol. 52, No.1,
p. 67–88.
Kossowska, M. et al. (2023), “The Role of Fear, Closeness,
and Norms in Shaping Help towards War Refugees”,
ScienticReports, vol. 13, No. 1465, p. 1–9.
Kreuter, F. et al. (2008), “Social Desirability Bias in CATI, IVR, and
Web Surveys: The Eects of Mode and Question Sensitivity”,
Public Opinion Quarterly, vol. 72, No. 5, p. 847–865.
Krosnick, J. A. and R. E. Petty (1995), “Attitude Strength:
AnOverview”, in Petty, R. E. and J. A. Krosnick (ed.), Attitude
Strength: Antecedents and Consequences, Psychology Press,
New York, p. 1–24.
Kruglanski, A. W. and W. Stroebe (2005), “The Inuence of
Beliefs and Goals on Attitudes: Issues of Structure, Function,
and Dynamics”, The Handbook of Attitudes, 1, 323–368.
Leininger, J. and A. Hornidge (2024), Sicherheitspolitik istnicht
Entwicklungspolitik, The Current Column, 19February 2024,
German Institute of Development andSustainability
(IDOS), Bonn.
MacKinnon, D. P. et al. (2012), “Statistical Mediation Analysis”,
inCooper, H. et al. (ed.), APA Handbook of Research Methods
in Psychology, vol. 2: Research Designs: Quantitative, Qualitative,
Neuropsychological, and Biological, AmericanPsychological
Association, p.313–331. https://doi.org/10.1037/13620-018.
Mader, M. and Schoen, H. (2023), “No Zeitenwende (yet):
Early Assessment of German Public Opinion Toward
Foreign and Defense Policy After Russia’s Invasion of
Ukraine”, Politische Vierteljahresschrift 64, p. 525–547.
Maio, G. R. et al. (2019), The Psychology of Attitudes
andAttitude Change. SAGE Publications Ltd.
Martini, F. et al. (2021), “Bot, or Not? Comparing Three
Methods for Detecting Social Bots in Five Political
Discourses”, Big Data & Society, vol. 8, No. 2, https://
journals.sagepub.com/doi/10.1177/20539517211033566.
Maull, H. W. (2015), “Deutsche Außenpolitik nach der
»Review2014«: Zivilmacht 2.0?”, Zeitschrift für Politik,
vol.62, No. 3, p. 323–341.
Maurer, M. and C. Reinemann (2006), Medieninhalte: Eine
Einführung, VS Verlag für Sozialwissenschaften, Wiesbaden.
Mende, S. (2011), ‘Nicht rechts, nicht links, sondern vorn’: Eine
Geschichte der Gründungsgrünen, Oldenbourg-Verlag, Munich.
Milner, H. V. and D. Tingley (2013), “Public Opinion and
Foreign Aid: A Review Essay”, International Interactions,
vol.39, No. 3, p. 389–401.
Morini, P. (2023a), DEL Dashboard – Germany January 2023.
Development Engagement Lab, London.
Morini, P. (2023b), DEL Dashboard – Germany June 2023.
Development Engagement Lab, London.
Morini, P. (2023c), DEL Dashboard – Germany October 2023.
Development Engagement Lab, London.
Mounk, Y. (2023), The Identity Trap: A Story of Ideas
andPowerinOur Time. Penguin Books.
Mutz, D. C. (2011), Population-based survey experiments.
Princeton University Press, Princeton.

Literature 104
Mullen, B. et al. (1992), “Ingroup Bias as a Function of
Salience, Relevance, and Status: An Integration”, European
Journal of Social Psychology, vol. 22, No. 2, p. 103–122.
Nord, M. et al. (2024), Democracy Report 2024: Democracy
Winning and Losing at the Ballot, V-Dem Institute,
Universityof Gothenburg.
Nosek, B. A. et al. (2018), “The Preregistration Revolution”,
Proceedings of the National Academy of Sciences, vol. 115,
No.11, p. 2600–2606.
ntv (2024), Deutsche Entwicklungshilfe: Das Märchen vom
315-Millionen-Euro-Radweg in Peru, 27 January 2024,
https://www.n-tv.de/politik/Das-Maerchen-vom-315-
Millionen-Euro-Radweg-in-Peru-article24691195.html
(accessed on 22 February 2024).
OECD (2021), OECD Development Co-operation Peer Reviews:
Germany 2021, OECD Publishing, Paris,
https://doi.org/10.1787/bb32a97d-en.
OECD (2022), Development Co-operation Proles, OECD
Publishing, Paris, https://doi.org/10.1787/2dcf1367-en.
OECD (2023), Development Co-operation Report 2023:
Debatingthe Aid System, OECD Publishing, Paris,
https://doi.org/10.1787/f6edc3c2-en.
OECD (2024), ODA Levels in 2023 – Preliminary Data.
Detailed Summary Note. 11 April 2024. OECD, Paris,
https://www.oecd.org/dac/ODA-summary-2023.pdf
(accessed on 13 May 2024).
Oce of the Auditor General of Canada (2023), International
Assistance in Support of Gender Equality –Global Aairs Canada.
Reports of the Auditor General of Canada to the Parliament of
Canada. Report 4. Ottawa, https://www.oag-bvg.gc.ca/internet/
docs/parl_oag_202303_04_e.pdf (accessed on 13 May 2024).
Oldendick, R. (2008), “Question Order Eects”, in Lavrakas,
P. J. (ed.), Encyclopedia of Survey Research Methods,
SagePublications, Thousand Oaks, CA.
Petty, R. E. et al. (2023), “Attitude Strength: What’s New?”,
TheSpanish Journal of Psychology, vol. 26, No. e4, p. 1–13.
Plagemann, J. (2022), Die Ukraine-Krise im globalen Süden: kein
„Epochenbruch“, GIGA Focus Global, 2, German Institute
forGlobal and Area Studies (GIGA), Hamburg, p. 1–9.
Politbarometer (2024),
Pomerantz, E. M. et al. (1995), “Attitude Strength and
Resistance Processes”, Journal of Personality and Social
Psychology, vol. 69, No. 3, p. 408.
Raden, D. (1985), “Strength-Related Attitude Dimensions”,
Social Psychology Quarterly, vol. 48, No. 4, p. 312–330.
Raghavarao, D. et al. (2010),Choice-Based Conjoint Analysis:
Models and Designs. Chapman and Hall/CRC.
Riddell, R. (2007), Does Foreign Aid Really Work?,
OxfordUniversity Press, Oxford.
Rivera-Rodriguez, A. et al. (2022), “Changing Public Opinion
about Gender Activates Group Threat and Opposition to
Feminist Social Movements among Men”, Group Processes &
Intergroup Relations,vol. 25, No. 3, p. 811–829.
Sassenhagen, N. et al. (2023), Feministische Entwicklungspolitik:
Was denkt die Bevölkerung? Ergebnisse einer repräsentativen
Umfrage, DEval Policy Brief 2/2023, German Institute for
Development Evaluation (DEval), Bonn.
Scheyer, V. and M. Kumskova (2019), “Feminist Foreign
Policy: A Fine Line between Adding Women and Pursuing
a Feminist Agenda”, Journal of International Aairs, vol. 72,
No. 2, p. 57–76.
Schick, L. et al. (2022), DEval-Meinungsmonitor Entwicklungs-
politik: Kognitiver Online-Pretest, GESIS Project Report,
version: 1.0, GESIS pretest laboratory, Mannheim/Cologne.
Schneider, S. H. et al. (2019), Meinungsmonitor
Entwicklungspolitik 2019: Medien, Mediennutzung

105Literature
undBerichterstattung zu weltweiter Armut, German Institute
for Development Evaluation (DEval), Bonn.
Schneider, S. H. et al. (2021a), Meinungsmonitor Entwicklungs-
politik 2021: Medieninhalte, Informationen, Appelle und ihre
Wirkung auf die öentliche Meinung, German Institute for
Development Evaluation (DEval), Bonn.
Schneider, S. H. (2021b), “Does the COVID-19 Pandemic
Threaten Global Solidarity? Evidence from Germany”, World
Development, vol. 140, p. 1–7,
https://doi.org/10.1016/j.worlddev.2020.105356.
Schneider, S. H. et al. (2022), Meinungsmonitor Entwicklungs-
politik 2022: Entwicklungspolitisches Engagement in Zeiten
globaler Krisen und Herausforderungen, German Institution
for Development Evaluation (DEval), Bonn.
Schneider, S. H. et al. (2024a), Die Agenda 2030 in der
öentlichen Meinung: Eine Halbzeitbilanz, DEval Policy Brief
1/2024, German Institute for Development Evaluation
(DEval), Bonn.
Schneider, S. H. et al. (2024b), “Who supports feminist
development policy? Evidence from Germany”, European
Journal of Politics and Gender, vol. 7, No. 2, p. 291–295.
Schneider, S. H. and S. H. Gleser (2018), Meinungsmonitor
Entwicklungspolitik 2018: Einstellungen zu Entwicklungs-
zusammenarbeit und nachhaltiger Entwicklung, German
Institute for Development Evaluation (DEval), Bonn.
Schulze, S. (2023), “Zeitenwende ist Entwicklungssache”,
Internationale Politik Online, https://internationalepolitik.de/de/
zeitenwende-ist-entwicklungssache (accessed on 13 May 2024).
Schulze, S. et al. (2023), Entwicklungssache #3: Wie denken
wir Sicherheit neu?, transcript available at: https://www.
bmz.de/resource/blob/135432/transskript-podcast-
entwicklungssache-folge-03.pdf, BMZ, Bonn.
Schulze, S. et al. (2024), Entwicklungssache #16: Krisenregion
Sahel: Was bedeutet das für Deutschland?, available at:
https://www.bmz.de/de/aktuelles/podcast, BMZ, Bonn.
Schuman, H. and S. Presser (1996), Questions and answers in
attitude surveys: Experiments on question form, wording, and
context, Sage Publications, Thousand Oaks, CA.
Scotto, T. J. et al. (2017), “We Spend How Much?
Misperceptions, Innumeracy, and Support for the Foreign
Aid in the United States and Great Britain”, Journal of
Experimental Political Science, vol. 4, No. 2, p. 119–128.
Sides, J. and K. Gross (2013), “Stereotypes of Muslims and
Support for the War on Terror”, The Journal of Politics, vol.75,
No. 3, p. 583–598.
SIPRI (2024), SIPRI Military Expenditure Database, Stockholm
International Peace Research Institute (SIPRI), Stockholm.
Smillie, I. (1999), “Public support and the politics of aid.”,
Development, Vol. 42, Nr. 3, S. 71-76.
Soroka, S. et al. (2019), “Cross-national Evidence of a
Negativity Bias in Psychophysiological Reactions to News”,
Proceedings of the National Academy of Sciences, vol. 116, No.
38, p. 18888–18892.
SPD, Alliance 90/The Greens and FDP (2021), Mehr Fortschritt
wagen: Bündnis für Freiheit, Gerechtigkeit undNachhaltigkeit,
coalition agreement between the SPD, Alliance 90/The
Greens and the FDP, Berlin.
Spurk, D. et al. (2020), “Latent Prole Analysis: A Review
and ‘how to’ Guide of its Application within Vocational
Behavior Research”, Journal of Vocational Behavior, vol. 120,
No. 103445, p. 1–21.
Stagnaro, M. N. et al. (2024), Representativeness versus
Response Quality: Assessing Nine Opt-In Online Survey
Samples, https://doi.org/10.31234/osf.io/h9j2d.
Stephan, W. G. et al. (2015), “Intergroup threat theory”,
Handbook of prejudice, stereotyping, and discrimination,
Psychology Press, p. 255-278.
Stern, M. (1998), Development Aid: What the Public thinks,
UNDP, Oce of Development Studies, Bureau for
Development Policy, New York.

Literature 106
Strabac, Z. and O. Listhaug (2008), “Anti-Muslim Prejudice
in Europe: A Multilevel Analysis of Survey Data from30
Countries”, Social Science Research, vol. 37, No. 1, p. 268–286.
Sundström, M. R. et al. (2021), “Spreading a Norm-Based
Policy? Sweden’s Feminist Foreign Policy in International
Media”, Contemporary Politics, vol. 27, No. 4, p. 439–460.
Taber, C. S. and M. Lodge (2006), “Motivated Skepticism in
the Evaluation of Political Beliefs”, American Journal of
Political Science, vol. 50, No. 3, p. 755–769.
Tagesschau (2023), Wie feministische Politik aussehen soll,
Tagesschau.de, 1 March 2023, https://www.tagesschau.de/
inland/feministische-aussenpolitik-entwicklungspolitik-
baerbock-101.html (accessed on 13 May 2024).
Tagesschau (2024), Millionen aus Deutschland für Radwege
in Peru?, Tageschau.de, 24 January 2024, https://
www.tagesschau.de/faktennder/radwege-peru-
entwicklungshilfe-100.html (accessed on 13 May 2024).
Tajfel, H. and J. C. Turner (1979), “An Integrative Theory of
Intergroup Conict”, in Austin, W. G. and S. Worchel (ed.),
The Social Psychology of Intergroup Relations, Brooks/Cole,
Monterey, CA, p. 33–48.
Tausendpfund, M. (2020), “Niveau und Determinanten
politischen Wissens”, in Tausendpfund, M. and
B.Westle(ed.), Politisches Wissen in Deutschland,
SpringerFachmedien, Wiesbaden, p. 89–126.
https://doi.org/10.1007/978-3-658-30492-8_4.
Thompson, L. and R. Clement (2019), Dening Feminist Foreign
Policy, International Center for Research on Women,
Washington D.C.
Thomson, J. (2020), “What’s Feminist about Feminist Foreign
Policy? Sweden’s and Canada’s Foreign Policy Agendas”,
International Studies Perspectives, vol. 21, No. 4, p. 424–437.
Tourangeau, R. and T. Yan (2007), “Sensitive Questions in
Surveys”, Psychological Bulletin, vol. 133, No. 5, p. 859–883.
Towns, A. et al. (2023), More than a Label, less than aRevolution:
Sweden’s Feminist Foreign Policy.
Expert Group for Aid Studies, Stockholm.
Transue, J. E. et al. (2009), “Treatment Spillover Eects Across
Survey Experiments”, Political Analysis, vol. 17, No. 2, p. 143–161.
Trope, Y. and Liberman, N. (2010), “Construal-Level Theory
ofPsychological Distance”,Psychological Review,117, No. 2, p.
440–463, https://doi.org/10.1037/a0018963.
United Nations (2015), Die Transformation unserer Welt:
DieAgenda 2030 für nachhaltige Entwicklung, Resolution of
the UN General Assembly, No. A/RES/70/1, New York.
United Nations (2020), Ziele für nachhaltige Entwicklung: Bericht
2020, United Nations, New York, https://www.un.org/
Depts/german/millennium/SDG%20Bericht%202020.pdf
(accessed on 16 February 2023).
United Nations (2022), The Sustainable Development Goals
Report 2022, United Nations, New York, https://unstats.
un.org/sdgs/report/2022/The-Sustainable-Development-
Goals-Report-2022.pdf (accessed on 16 February 2023).
United Nations (2023a), Ziele für nachhaltige Entwicklung.
Bericht 2023: Sonderausgabe, United Nations, New York,
https://www.un.org/Depts/german/millennium/SDG%20
Bericht%202023.pdf (accessed on 16 February 2023).
United Nations (2023b), Principles of the Charter of the
United Nations underlying a comprehensive, just and lasting
peace in Ukraine, A/ES-11/L.7, United Nations, New York,
https://documents.un.org/doc/undoc/ltd/n23/048/58/
pdf/n2304858.pdf?token=KqnegdcISXqiDFj79v&fe=true
(accessed on 22 February 2024).
United Nations (2023c), UN General Assembly Calls for
Immediate End to War in Ukraine, 23 February 2023, United
Nations, https://news.un.org/en/story/2023/02/1133847
(accessed on 22 February 2024).

107Literature
UN Women (2015), Preventing Conict, Transforming Justice,
Securing the Peace. A Global Study on the Implementation of
United Nations Security Council resolution 1325, UN Women,
New York.
UN Women (2020), Pekinger Erklärung und Aktionsplattform,
asat October 2020, https://unwomen.de/pekinger-
erklaerung/ (accessed on 13 May 2024).
UN Women (2023), Progress on the Sustainable Development
Goals: The Gender Snapshot 2023, UN Women, New York,
https://www.unwomen.org/sites/default/les/2023-09/
progress-on-the-sustainable-development-goals-the-gender-
snapshot-2023-en.pdf (accessed on 13 May 2024).
Verbrugge, B. and H. Huyse (2020), “Civil Society Funding
by OECD-DAC Governments: No More Politics as Usual?”,
Development in Practice, vol. 30, No. 6, p. 751–762.
Vehrkamp, R. and W. Merkel (2019), Populismusbarometer
2018, Bertelsmann Stiftung, Gütersloh.
Vermunt, J. K. and J. Magidson (2002), “Latent Class Cluster
Analysis”, in Hagenaars, J. A. and A. L. McCutcheon (ed.),
Applied Latent Class Analysis, Cambridge University Press,
Cambridge, p. 89–106.
Vogel, T. K. et al. (2016), Attitudes and attitude change: second
edition, Psychology Press, New York.
Wencker, T. and I. Verspohl (2019), German Development
Cooperation in Fragile Contexts, German Institute for
Development Evaluation (DEval), Bonn.
Wietholter, C. and A. L. Hillard (2021), “Feminism = Women’s
Movement? The Eects of Gender and Terminology on
Endorsement of Feminism”, The Undergraduate Journal of
Psychology, vol. 23, No. 1, p. 14–20.
Wittkopf, E. R. (1986), “On the Foreign Policy Beliefs of
the American People: A Critique and Some Evidence”,
International Studies Quarterly, vol. 30, No. 4, p. 425–445.
World Food Programme (2022), WFP Annual Review 2022,
World Food Programme, Rome.
Yang, K.-C. et al. (2020), “Scalable and Generalizable Social
Bot Detection through Data Selection”, Proceedings of
the AAAI Conference on Articial Intelligence, 34, No. 1, p.
1096–1103. https://doi.org/10.1609/aaai.v34i01.5460.
YouGov (2017), YouGov Panelbook 2017, London.
Zagea, H. et al. (2011), „Donating to Disaster Victims: Responses
to Natural and Humanly Caused Events”, European Journal
ofSocial Psychology, vol.41, No. 3, p. 353–363.
Zhukova, E. (2023), “Postcolonial Logic and Silences in
Strategic Narratives: Sweden’s Feminist Foreign Policy in
Conict-Aected States”, Global Society,37, No. 1, p. 1–22.
Zilla, C. (2022), “Feministische Außenpolitik: Konzepte,
Kernelemente und Kontroversen”, SWP-Aktuell, No. 50,
Stiftung Wissenschaft und Politik (SWP), Berlin.
Zille, H. et al. (2023), Entwicklungspolitische Einstellungen im
Kontext des Krieges gegen die Ukraine, DEval Policy Brief
9/2023, German Institute for Development Evaluation
(DEval), Bonn.
Zuell, C. and E. Scholz (2019), “Construct Equivalence of Left-
Right Scale Placement in a Cross-National Perspective”,
International Journal of Sociology, vol. 49, No. 1, p. 77–95.

7.
ANNEX

109Annex
7.1 Data sources
Table 1 Surveys: Aid Attitudes Tracker and Development Engagement Lab
No. Designation Data collection Respondents (total)
1AAT wave 1 December 2013 5,700
2AAT wave 2 June 2014 6,170
3AAT wave 3 December 2014 5,914
4AAT wave 4 June 2015 6,059
5AAT wave 5 December 2015 6,027
6AAT wave 6 June 2016 6,049
7AAT wave 7 December 2016 6,131
8AAT wave 8 July 2017 6,096
9AAT wave 9 December 2017 6,108
10 AAT wave 10 July 2018 6,039
11 DEL panel wave 1 September 2019 6,004
12 DEL tracker wave 1 January 2020 1,141
13 DEL tracker wave 2 June 2020 1,025
14 DEL panel wave 2 September 2020 6,000
15 DEL tracker wave 3 January 2021 1,004
16 DEL tracker wave 4 June 2021 1,001
17 DEL panel wave 3 September 2021 6,000
18 DEL tracker wave 5 January 2022 1,015
19 DEL sandbox June 2022 2,059
20 DEL tracker wave 6 June 2022 1,038
21 DEL panel wave 4 October 2022 6,008
22 DEL tracker wave 7 January 2023 1,100
23 DEL tracker wave 8 June 2023 1,067
24 DEL panel wave 5 October 2023 6,050
25 DEL tracker wave 9 January 2023 1,021
Source: DEval, own visualisation.

Annex110
Table 2 Respondi/Bilendi surveys
No. Time period Respondents
(total)
Contents Remarks
124 January –
31 January 2023
2,000 Core questionnaire; module on
cooperation with the private sector
229 June –
12 July 2023
2,050 Foreign policy attitudes, emergency
situations, support for Ukraine
311 July –
9 August 2023
4,050 Survey tool for attitude strength;
strength of attitudes towards
development policy and feminist
development policy
Half of respondents had to answer the block on
feminist development policy rst, followed by the
block on development policy; the other half had to
complete the survey the other way round.
425 July –
8 August 2023
2,000 Core questionnaire
54 August –
22 August 2023
2,000 Knowledge and understanding of
feminist development policy
612 October –
24 October 2023
3,047 Conjoint experiment on prioritising
partner countries
The survey contained a module for which the
respondents were redirected to the survey website
SoSci-Survey and had to answer the conjoint
component there.
727 October –
13 November
2023
1,628 Survey experiment on feminist
development policy: content v. label
The sample was divided into four experimental
groups. Attention check: respondents who did not
answer a content-related attention check question
correctly were excluded.
819 December
– 28 December
2023
2,146 Survey experiment on feminist
development policy in a time of
globalcrises
The sample was divided into four experimental
groups. Attention check: respondents who were
unable to answer a low-threshold content-related
question about the treatments were excluded.
911 January –
19 January 2024
2,046 Survey experiment on support for DC in
the context of wars against Ukraine and
economic troubles
The sample was divided into four experimental
groups. Attention check: respondents who were
unable to answer a low-threshold content-related
question about the treatments were excluded.
10 10 January –
17 January 2024
2,101 Core questionnaire; questions about
expenditure on development policy in
comparison with other policy areas;
development policy education
Source: DEval, own visualisation.

111Annex
7. 2 Supplementary analyses and gures
Figure 44 LayoutSupport for development cooperation/ODA expenditure by party voted for in the last Bundestag elections
33
16
40
11
27
17
45
11
27
17
46
10
31
19
41
10
22
17
53
8
30
14
46
11
37
16
40
7
21
26
44
9
20
27
44
9
25
16
48
12
22
18
49
11
25
23
42
10
24
20
45
11
13
42
38
7
23
21
46
10
22
18
51
9
21
19
52
8
29
16
48
7
18
27
49
6
19
21
55
5
20
19
54
7
20
25
47
8
16
34
42
8
19
18
55
7
26
22
43
9
18
31
45
6
23
29
38
10
12
39
36
12
20
25
42
13
23
27
39
12
22
28
39
12
23
21
35
21
8
20
55
16
26
25
28
21
31
19
34
15
15
32
40
13
16
38
35
11
27
22
41
10
14
33
37
17
14
29
48
9
22
35
28
15
4
44
36
16
41
13
36
10
41
9
41
9
39
9
42
10
44
9
44
3
45
8
37
10
47
3
36
13
46
7
34
13
38
9
44
9
30
12
46
12
38
5
43
13
46
7
32
15
46
12
34
9
37
13
38
12
22
24
42
11
11
47
32
11
8
52
31
9
8
58
28
6
17
47
29
8
14
44
30
12
12
53
25
9
5
51
32
11
4
66
23
7
6
67
22
5
5
61
21
13
4
59
25
12
8
66
20
6
6
64
24
6
1
71
18
10
15
28
35
22
16
27
37
20
15
30
39
16
19
30
32
18
12
37
36
16
18
21
39
22
13
27
43
17
14
36
30
20
13
39
32
16
19
22
42
17
18
28
30
24
14
29
32
25
14
39
32
16
12
48
28
12
19
21
52
9
19
21
54
7
20
23
48
9
14
26
52
8
22
29
50
17
20
58
5
16
18
57
8
11
34
45
10
10
40
39
10
17
18
57
8
16
14
55
15
12
31
47
9
18
35
41
6
11
51
37
2
35
17
35
12
35
16
39
10
32
18
41
9
44
9
35
12
32
16
44
7
36
17
41
6
33
22
38
6
28
18
41
13
24
28
34
13
50
9
37
4
26
15
46
13
22
21
36
22
35
12
38
15
27
22
46
5
Other
Non-voter
CDU/CSU
AfD
The Left
SPD
Alliance 90/The Greens
FDP
9/2019
1/2020
6/2020
9/2020
1/2021
6/2021
9/2021
1/2022
6/2022
10/2022
1/2023
6/2023
10/2023
1/2024
9/2019
1/2020
6/2020
9/2020
1/2021
6/2021
9/2021
1/2022
6/2022
10/2022
1/2023
6/2023
10/2023
1/2024
9/2019
1/2020
6/2020
9/2020
1/2021
6/2021
9/2021
1/2022
6/2022
10/2022
1/2023
6/2023
10/2023
1/2024
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
Date of survey
Percentage of respondents
Increase slightly/a great deal Do not change Reduce slightly/a great deal Don’t know
Source: DEval, own visualisation. Source of the data: DEL tracker and DEL panel 2019 -2024. Weighted data. The question was: “From its total budget of approximately EUR 360 billion, the German Federal
Government currently provides X.X percent – EUR XX. X billion – (note: the gures are updated each year) to poor countries for development cooperation. Do you think that the government should increase or
decrease the amount of money that it spends on development cooperation?” The response options were: “increase a great deal”, “increase somewhat”, “don’t change”, “decrease somewhat”, “decrease a great
deal”. The categories were conated for the gure.

Annex112
Figure 45 Policy area in which respondents would be most willing to cut expenditure
2
3
3
10
3
9
11
5
13
21
21
Education
Internal security
Health
Pensions
Transport
Digitalisation
Defence
Environmental protection
Economic promotion
Unemployment benefits
Development cooperation
0 10 20 30 40 50
Percentage of respondents
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2024. N = 2,101. The question was: “Imagine that
government expenditure urgently needs to be cut. In which of the following policy areas would you be most willing to cut expenditure? Please select a policy area.”
Thequestion is based on the policy areas used in the survey carried out by the ZMSBw (Graf, 2022, p. 5).

113Annex
Figure 46 Preferred policy area for cuts by party identication
3
2
3
7
3
5
13
4
12
23
26
2
6
4
15
6
12
4
5
29
6
13
2
4
7
0
4
9
20
2
4
26
22
2
3
3
12
3
7
9
5
19
19
19
0
1
5
8
3
8
8
6
12
19
30
3
1
5
23
0
8
9
6
13
21
11
3
3
0
20
3
14
7
10
17
10
12
3
2
5
7
3
7
20
3
4
16
30
0
2
2
10
2
10
10
5
13
22
24
Free Voters
Bündnis Sarah Wagenknecht (BSW)
Not affiliated to a party
Alliance 90/The Greens
FDP
AfD
CDU/CSU
SPD
The Left
0 10 20 30 0 10 20 30 0 10 20 30
Pensions
Education
Health
Internal security
Environmental protection
Development cooperation
Transport
Unemployment benefits
Digitalisation
Economic promotion
Defence
Health
Education
Internal security
Transport
Economic promotion
Pensions
Defence
Digitalisation
Unemployment benefits
Environmental protection
Development cooperation
Education
Health
Internal security
Pensions
Transport
Defence
Digitalisation
Environmental protection
Economic promotion
Development cooperation
Unemployment benefits
Education
Health
Internal security
Pensions
Transport
Digitalisation
Environmental protection
Defence
Development cooperation
Economic promotion
Unemployment benefits
Education
Health
Internal security
Pensions
Transport
Defence
Digitalisation
Environmental protection
Economic promotion
Development cooperation
Unemployment benefits
Internal security
Health
Education
Pensions
Transport
Digitalisation
Environmental protection
Unemployment benefits
Economic promotion
Development cooperation
Defence
Health
Education
Internal security
Pensions
Transport
Digitalisation
Defence
Economic promotion
Environmental protection
Development cooperation
Unemployment benefits
Education
Environmental protection
Pensions
Transport
Development cooperation
Health
Internal security
Digitalisation
Unemployment benefits
Defence
Economic promotion
Defence
Education
Transport
Economic promotion
Health
Internal security
Pensions
Digitalisation
Environmental protection
Unemployment benefits
Development cooperation
Percentage of respondents
Source: DEval, own visualisation. Source of the data: DEval tracking, carried out by Respondi/Bilendi in January 2024. N = 2,101. The question was: “Imagine that government expenditure
urgently needs to be cut. In which of the following policy areas would you be most willing to cut expenditure? Please select a policy area.” Thequestion is based on the policy areas used in the
survey carried out by the ZMSBw (Graf, 2022, p. 5).

Annex114
Figure 47 Strength of the correlation between the assessment of the economic situation and support
for development cooperation in dierent years
0
1
2
3
4
5
6
7
8
9
10
-2 0 2
Assessment of the national economic situation
General support for DC
2019
2020
2021
2022
2023
0
1
2
3
4
5
6
7
8
9
10
-2 0 2
Assessment of one’s own personal economic situation
General support for DC
2019
2020
2021
2022
2023
Source: DEval, own visualisation. Source of the data: DEL panel wave 1 to 5; N per wave ≈ 6,000. Multi-level model for longitudinal data (random intercepts and
random slopes for the assessment of the economic situation). The gure is based on models M4 and M5 in Table 24 in the online appendix.

115Annex
Figure 48 Strength of the correlation between the assessment of the economic situation and public support
forthecurrent development cooperation/ODA expenditure in dierent years
-2
-1
0
1
2
-3 -2 -1 0 1 2 3
Assessment of the national economic situation
Support for the current DC/ODA expenditure
2019
2020
2021
2022
2023
-2
-1
0
1
2
-2 0 2
Assessment of one’s own personal economic situation
Support for the current DC/ODA expenditure
2019
2020
2021
2022
2023
Source: DEval, own visualisation. Source of the data: DEL panel wave 1 to 5; N per wave ≈ 6,000. Multi-level model for longitudinal data (random intercepts
andrandom slopes for the assessment of the economic situation). The visualisation is based on models M4 and M5 in Table 25 of the online appendix.

Annex116
7.3 Study schedule
Time frame Tasks
January 2023 – March 2023 Conception and preparation of the study concept
March 2023 First reference group meeting to discuss the study concept
April 2024 – May 2024 Revision of the study concept
May 2023 – February 2024 Data collection, data analysis, report writing
March 2024 – May 2024 Internal and external peer review; revision of the draft report
May 2024 First reference group meeting to discuss the draft report
June 2024 – July 2024 Revision of the draft report
July 2024 – October 2024 Editing, layout and publication
7. 4 Study team and contributors
Core team Function CRediT-Statement74
Dr Sebastian H. Schneider Team leader Conceptualization, Methodology, Formal analysis, Investigation,
Writing – Original Draft, Writing – Review & Editing, Visualization,
Supervision, Project administration
Dr Alexandra Gödderz Evaluator Conceptualization, Methodology, Formal analysis, Investigation,
Writing – Original Draft, Writing – Review & Editing, Visualization
Dr Helge Zille Evaluator Conceptualization, Methodology, Formal analysis, Investigation,
Writing – Original Draft, Writing – Review & Editing, Visualization
Dr Martin Bruder Head of department Conceptualization, Writing – Review & Editing, Supervision
Merle Gmeineder Project administrator Project administration
Caroline Orth Project administrator Project administration
Contributors Function
Prof. Jörg Faust Internal peer reviewer
Heike Steckhan Internal peer reviewer
Dr Markus Steinbrecher External peer reviewer
Simon Becker Student employee
Hanna Link Student employee
Carla Vasco Páez Student employee
Responsible
Dr. Martin Bruder Head of department
74 The CRediT statement (Contributor Roles Taxonomy, https://credit.niso.org/) indicates the roles of the authors of this evaluation report in the evaluation. The CRediT
taxonomy distinguishes between 14 dierent roles to show the specic contribution of the individual authors.

Annex90
German Institute for
Development Evaluation (DEval)
Fritz-Schäer-Straße 26
53113 Bonn, Germany
Phone: +49 (0)228 33 69 07–0
E-Mail: info@DEval.org
www.DEval.org