Content uploaded by Lena Frischlich
Author content
All content in this area was uploaded by Lena Frischlich on Aug 25, 2014
Content may be subject to copyright.
This book deals with the psychological effects of extremist propaganda videos. It
particularly asks the question how young adults in Germany respond to right-
wing as well as Islamic extremist videos which can be found on the Internet today.
This is not a book about terrorism, but about the potential conditions which might
facilitate a climate of receptivity for radical messages in a young mass audience
with diverging cultural and educational background and different attitudes and
values.
The so called web 2.0, with its mostly unfiltered, user-created content provides
unprecedented opportunities for extremists to present themselves and uncensored
ideas to a mass audience. This internet propaganda is created in order to increase
attention and interest for extremist ideas and group memberships. It also aims to
indoctrinate the recipients and, as a last consequence, to foster radicalization.
The radicalizing potential has been feared by international security agencies and
mass media. Nevertheless, not even the early stage effects of extremist propaganda
in terms of raising attention and interest have yet been analyzed empirically.
They are however necessary preconditions in order for propaganda to envelope
a radicalizing effect.
The current studies close this gap by focusing on this early stage effects. We ana-
lyzed how a non-radicalized audience responds to extremist internet videos. For the
first time, based on a content analysis of actual right-wing and Islamic extremist
Internet videos, our study used state-of-the-art methods from experimental media
psychology for tracking the emotional and cognitive responses of a broad sample
of 450 young male adults. As expected, we mostly found rejection and never
strong acceptance for the extremist videos. Still, specific production styles and
audience characteristics were able to cause at least neutral attitudes underpinning
the strategic potential of internet propaganda. In the end, our studies might result
in more questions than answers. However, we are confident that the conceptual as
well as the methodological way chosen is most promising as to approach a deeper
understanding of the first effects of extremist Internet propaganda.
Diana Rieger · Lena Frischlich · Gary Bente
In cooperation with the Terrorism/Extremism
Research Unit (FTE) of the German Federal
Criminal Police Office (Bundeskriminalamt)
Rieger · Frischlich · Bente Propaganda 2.0
Propaganda 2.0
Psychological Effects of
Right-Wing and Islamic
Extremist Internet Videos
ISBN: 978-3-472-08526-3
www.luchterhand-fachverlag.de
9 783472 085263
08526000_BKA Band 44_Layout 1 08.05.13 15:34 Seite 1
Propaganda 2.0
Psychological Effects of Right-Wing
and Islamic Extremist Internet Videos
Titelei BKA Band 44 neu_Layout 1 11.04.13 14:54 Seite 1
Polizei + Forschung
Bd. 44
Edited by
German Federal Criminal Police Office
Institute of Law Enforcement Studies and Training
Advisory board:
Prof. Dr. Johannes Buchmann
Direktor of the Center for Advanced Security Research Darmstadt
Wolfgang Gatzke
Direktor of the State Office of Criminal Investigation North
Rhine-Westphalia
Prof. Dr. Manfred Hennecke
President of the Federal Institute for Materials Research an Testing
Prof. Dr. Hans-Jürgen Kerner
Director of the Institut for Criminology at the University of Tübingen
Waldemar Kindler
President at the Ministry of the Interior in the State of Bavaria
Klaus Neidhardt
President of the German Police University
Prof. Dr. Peter Wetzels
Professor of Criminology at the School of Law at the Univeristy
of Hamburg
Titelei BKA Band 44 neu_Layout 1 11.04.13 14:54 Seite 2
Diana Rieger · Lena Frischlich · Gary Bente
In cooperation with the Terrorism/Extremism
Research Unit (FTE) of the German Federal
Criminal Police Office (Bundeskriminalamt)
Propaganda 2.0
Psychological Effects of
Right-Wing and Islamic
Extremist Internet Videos
Titelei BKA Band 44 neu_Layout 1 11.04.13 14:54 Seite 3
Citation: Rieger, Frischlich, Bente, Propaganda 2.0 Psychological Effects of
Right-Wing and Islamic Extremist Internet Videos
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche
Nationalbibliografie; detailed bibliographic data are available in the Internet at
http://dnb.d-nb.de.
All publication of the BKA-Series Polizei + Forschung (apart from VS-NfD-
classified volumes) are available as a pdf-file on the Internet on www.bka.de
(Kriminalwissenschaften/Kriminalistisches Institut).
Project management:
Dipl.-Psych. Brahim Ben Slama
German Federal Criminal Police Office
Institute of Law Enforcement Studies and Training
Terrorism/Extremism Research Unit (FTE)
www.wolterskluwer.de
© 2013 Wolters Kluwer Deutschland GmbH,
Luxemburger Straße 449, 50 939 Köln.
Luchterhand – a trademark of Wolters Kluwer Deutschland GmbH.
All rights reserved. No part of this publication may be reproduced or transmitted
in any form or by any means, or stored in any retrieval system of any nature
without prior written permission, except for permitted fair dealing under the
Copyright Act, or in accordance with the terms of a license issued in respect of
photocopying and/or reprographic reproduction. Application for permission for
other use of copyright material including permission to reproduce extracts in
other published works shall be made to the publishers. Full acknowledgment of
author, publisher and source must be given.
The publisher and the authors can accept no liability for errors in the content or
for printing errors.
Typeset by: Satzoffizin Hümmer, Waldbüttelbrunn
Printed by: Paper&Tinta, Nadma
Printed on acid-free and aging resistant paper.
Titelei BKA Band 44 neu_Layout 1 11.04.13 14:54 Seite 4
{luchterh_neu}20130001_BKA_Propaganda/00_titelei.3D 5 08.05.2013 S. 5
Foreword
The radicalization of individuals through extremist propaganda is a phenomenon
which has become increasingly discussed in recent years. Extremist organisations
chiefly use videos and online services on the Internet to win over supporters and to
radicalize individuals. For jihadist organisations, such as Al-Qaida, but also for
right-wing extremists the Internet has become the most important communication
and propaganda tool.
Within the research community there is little agreement regarding the degree of
impact the Internet has on radicalization and recruiting processes. There is, how-
ever, a consensus that within such processes the Internet is deemed, in general, to
exert a considerable influence. Today, the Internet seems to play a key role with
regard to the dissemination of propaganda material by extremist organisations.
The extremist propaganda of jihadist and right-wing extremist organisations has
recently changed not only in quantitative terms. More and more video messages
are being released in German and references are made to current political and so-
cial affairs in Germany. Parallel to this development, the Security Authorities in
Germany have identified numerous travel movements to training camps and com-
bat areas as well as an increased inflow to autonomous right-wing extremists. An
additional factor is the phenomenon of self-radicalization of right-wing and Isla-
mic extremists in the privacy of one’s home, as highlighted in the case of Arid Uka
in Germany or Anders Behring Breivik in Norway.
In order to initiate preventive and repressive counter-measures, a better under-
standing of the effect of extremist propaganda disseminated over the Internet is
required. It is especially important to determine how effective such propaganda
videos really are. Who is susceptible to such propagandistic messages?
In an effort to seek answers to these questions, the German Federal Criminal Po-
lice Office (the “Bundeskriminalamt”), in co-operation with the Dutch National
Coordinator for Security and Counterterrorism within the Ministry of Security
and Justice, invited interested parties to tender for a research project. I am very
pleased that cross-border co-operation with a Dutch Security Agency has been
achieved, focusing on the important question of research in the field of extremism.
This also documents the fact that we are confronted with similar questions in re-
lation to the fight against extremism and that we can pursue the answers to these
questions together. Thus far, research on the impact of extremist propaganda on
the Internet has taken a more theoretical approach, or was aimed at findings de-
rived from qualitative analyses. Systematic, broad-ranging empirical studies have
not been published yet. This experimental study is an initial step towards closing
this research gap.
V
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/00_titelei.3D 5 08.05.2013 S. 6
The main objective behind the research was to identify factors which can influ-
ence the emotional and cognitive effects of Internet propaganda on young – not
radicalized – adults.
I would like to thank the research team led by Prof. Dr. Gary Bente, Head of the
Department for Social Psychology at the University of Cologne, who has carried
out a major empirical study which is the first of its kind. The effects of propaganda
are demonstrated with the help of a combination of various quantitative, qualita-
tive and experimental methods which ensure a broad discussion on the findings,
offering a true insight into relevant interdependencies. At the same time, new
questions are raised. It has to be underlined that the focus is not on conspicuous
individual cases of radicalization, but on the influence of extremist Internet pro-
paganda on inconspicuous individuals. The study is an attempt to identify the ba-
sal effects inherent in such propaganda forming necessary though non-sufficient
pre-conditions for a potential influence on radicalization. .
The findings of the study highlight new avenues for the prevention of radicaliza-
tion over the Internet. With this study, the Bundeskriminalamt and the Dutch Na-
tional Coordinator for Security and Counterterrorism strive to promote further re-
search on the influence of (Internet) extremist propaganda and provide impulses
for the prevention of extremist radicalization.
Jörg Ziercke
President of the Bundeskriminalamt
VI
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/00_titelei.3D 5 08.05.2013 S. 7
Acknowledgements
The current research project was commissioned by the Federal Criminal Police
Office (Bundeskriminalamt, BKA) Research Unit on Terrorism/Extremism and
co-founded by the Expertise and Analysis Department of the Federal Ministry
of the Interior, and the Research Direction of the Dutch National Coordinator
for Security and Counterterrorism. But, in fact, this was not merely a contract.
It was a collaborative effort among principal and agent, an inspiring experience
in which the authors of this book largely benefited from the profound knowledge
and experience in the field provided by the partners in the contracting institutions.
Above all, we have to thank Brahim Ben Slama and Dr. Uwe Kemmesies from the
Terrorism and Extremism Research Unit (FTE) of the Federal Criminal Police Of-
fice (Bundeskriminalamt) who guided us through the complex matters of extre-
mism, providing valuable background information and critical comments on
methods and interpretations without ever interfering with our primarily scientific
goals. This acknowledgement has to be extended to Mr. Jan Kortekaas from the
National Coordinator for Security and Counterterrorism in The Netherlands,
who had provided useful comments throughout the project and had widened
our perspective to the international scenery.
We further gratefully acknowledge the invaluable support received from the Joint
Internet Surveillance Centre in Berlin (GIZ) and the jugendschutz.net with regard
to the collection of our stimulus material. Mr. Khalid Zoubairi and Mr. Stefan Gla-
ser had not only provided us with rich samples of propaganda videos but had also
advised us with regard to the selection and content analytical criteria.
Extending experimental research beyond the usual samples of psychology stu-
dents is certainly a particular challenge. We wish to express our gratitude to the
International Students’ Office (University of Cologne), in particular to Mr.
Karl-Heinz Korn for supporting us in contacting and recruiting students from dif-
ferent cultures and disciplines. Extending our samples beyond student groups
would not have been possiblewithout the support of the vocational schools in Co-
logne. In particular, we gratefully acknowledge the support of the vocational
school “Hans-Böckler” in Cologne Deutz and the support of Mr. Jochen Erpen-
beck, the vocational school for media professions in Cologne Zollstock, espe-
cially Mr. Klaus-Dieter Schulz; and the vocational school Cologne Ehrenfeld,
Mr. Christoph Schneider. They had not only brought us into contact with potential
participants in their schools but had also supported us in conducting the studies
on-site, providing both time and space in this regard.
Last but not least our thanks go to the university team in particular to Dr. Julia
Kneer for her scientific and personal support in all phases of the project. Mr. Tho-
mas Dratsch deserves our gratitude for his thorough reviews and his valuable sup-
port in conducting the group discussions. Further, we thank Mr. Kliment Yanev,
VII
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/00_titelei.3D 5 08.05.2013 S. 8
who took care of the technical solutions needed for the stimulus presentation and
the online data collection, and Mr. Haug Leuschner for his assistance in advanced
statistical analyses.
Neither the development of the project, nor the extensive data collection would
have been possible without the encouraging comments and the support of our stu-
dents and lab assistants. Thanks to Mrs. Julia Tebbe for her diligent contribution
and the incredible effort in finding new participants. Thanks to Mrs. Sutude
Shishechian and Ms. Oana Teodosiu, not only but above all, for their strenuous
efforts in improving our text. Our thanks also go to Mr. Nemezjusz Kasztelan,
Mr. Daniel Piontek, Ms. Victoria Hieb and Mr. Eyüp Aksoy for their special con-
tributions in reviewing and recruiting, but also for their comments during the
whole project.
No psychological study is possible without the support of numerous anonymous
participants. This study would not have been possiblewithout their trust and open-
ness. Their interest and feedback deeply impressed us and encouraged the conti-
nuation of our work.
VIII
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/00_titelei.3D 5 08.05.2013 S. 9
Content
Page
1. Introduction ............................................. 1
Part I Propaganda in the new Millennium ..................... 6
2. What is Propaganda? A Development Towards Modern Media
Applications ............................................. 6
3. Extremism in the 21st century .............................. 10
3.1 Islamic Extremism . . . . . . .............................. 11
3.2 Right-Wing Extremism in Germany . . . . . . . . . . . . .......... 14
4. Virtual Propaganda and Extremist Videos .................... 16
4.1 Right-wing and Islamic Extremists in the Internet . . .......... 16
4.2 Propaganda through Audio-visual Material . . . . . . . .......... 19
5. The Audience of Extremist Propaganda ...................... 28
5.1 Knowing the Audience . . . .............................. 28
5.2 Demographic Factors . . . . .............................. 29
5.2.1 Age and Gender . . .............................. 30
5.2.2 Cultural Background . . . . . . . . . . . . . . . . . . .......... 31
5.2.3 Educational Level and Major Field of Study .......... 32
5.3 Personality Factors and Attitudinal Patterns . . . . . . . .......... 33
6. The Effects of Extremist Propaganda: Rejected Acceptance? ..... 36
7. Research Questions ....................................... 39
Part II Studying the Effects of Right-Wing and Islamic Extremist
Propaganda Videos ................................... 42
8. The Current Study: Exploring the Effects of Propaganda 2.0 ..... 42
9. Research Focus I: Description and Analysis of Right-Wing
Extremist and Islamic Extremist Propaganda Videos ........... 45
9.1 The Categorization System and Frequency of the Formats . . . . . 47
9.2 The Selected Videos for the Studies . . . . . . . . . . . . .......... 51
9.3 Procedure Study I and II . .............................. 56
9.4 The Effects of Propaganda 2.0 – Immediate Evaluations . . . . . . . 58
IX
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/00_titelei.3D 5 08.05.2013 S. 10
9.4.1 Online Measures . . . . . . . . . . . . . . . . . ............... 59
9.4.2 Post-hoc Measures . . . . . . . . . . . . . . . ............... 60
9.4.3 Results for Single Videos . . . . . . . . . . ............... 62
9.4.4 Summary: Research Focus I . . . . . . . . ............... 72
10. Research Focus II: Propaganda and Different Audiences ........ 74
10.1 Characteristics of the Audience . . . . . . . . . . . ............... 74
10.2 Sample . . . .......................................... 77
10.3 Results for Ideology . . . . . . . . . . . . . . . . . . . . ............... 81
10.3.1 Demographics and the Evaluation of Propaganda . . . . . . 81
10.3.2 The Role of Personality and Attitude . ............... 90
10.4 Summary: Research Focus II . . . . . . . . . . . . . ............... 102
11. Research Focus III: Implicit Associations and Propaganda ....... 104
11.1 Sample, Procedure and Material Study III . . . ............... 106
11.2 Results for Implicit Effects . . . . . . . . . . . . . . ............... 108
11.3 Summary: Research Focus III . . . . . . . . . . . . ............... 109
12. Discussion ............................................... 111
References ................................................. 121
Appendix – Supplementary Statistical Material ................... 137
A1. Factor Loadings and Multilevel Factor Analysis . . . . . ........ 137
A2. Pairwise Comparisons for the Effect of Single Videos ........ 142
A3. Pairwise Comparisons for the Interactions between Cultural
Background, Educational Level and Ideology ............... 146
A4. Words Used in the Single Category Implicit Association Test . . . 151
Tables ..................................................... 152
Figures .................................................... 154
X
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 1
1. Introduction
“This is the secret of propaganda: To totally saturate the person, whom the pro-
paganda wants to lay hold of, with the ideas of the propaganda, without him even
noticing that he is being saturated.” (Paul Watzlawick, 1984, p. 112)
The idea of propaganda may be as old as mankind; but its threatening potential is
quite contemporary. In Watzlawick’s terms, the main purpose of propaganda is to
persuade the recipient into adopting the ideas the propagator tries to convey. Con-
vincing the masses of one’s own ideology or dogma has always been part of every
political, religious or societal system. Yet, in our globalized world propaganda has
exceeded national borders or definable audiences. The increasing dissemination
of the internet and its enormous outreach make it the perfect instrument to broad-
cast messages worldwide without temporal or geographical restrictions. Online
messages can be read at the other end of the world within the blink of an eye.
Furthermore, the internet allows the combined usage of stylistic means propa-
ganda so far used separately (texts, videos, and discussion forums) and enables
the masses to actively create their own content.
In particular, extremist groups have discovered these benefits and use the internet
as their new home-base and operational area (Hoffmann, 2006). The two biggest
of these virtual propagators are right-wing and Islamic extremists.1In Germany
alone, there are thousands of online forums, blogs, social media offers and, parti-
cularly, videos that disseminate their ideology (Bundesamt für Verfassungs-
schutz, 2008; Glaser, 2011). By spreading messages via online channels, three
extremist objectives are regulated: (1) to threaten the public (2) to provoke the en-
emy and (3) to mobilize the potential ingroup to join the organization (Münkler,
2002; Wegner, 2009).
Prominent strategists from right-wing as well as Islamic extremism (e.g. Al-Suri,
Milton Kleim Jnr.) pronounce the potential of the internet to form a base for a
“media resistance”. For instance, Al’Qaida’s chief strategist Al-Suri declared
the internet to be an activating tool. It would teach the recipient “to pursue jihad
and resistance in secrecy and alone, [.. .] and [to form a cell] for the individual
jihad.” (Al-Suri cf. Lia, 2007a, p. 393)
It is not surprising that security agencies and mass media have stated concerns that
extremist propaganda would have the power to indoctrinate its recipients and to
push extremist movements (Europol, 2011a). It has even been assumed that a
so-called “turbo-radicalization” (Denso, 2011) among so far non-obtrusive peo-
ple could be possible via single videos. Particularly young adults, digital natives,
1 Left-wing extremist online activities are reported less frequently and differ qualitatively in their
appearance (e.g. Verfassungsschutz des Saarlandes, 2010). Therefore, this book as well as the
conducted studies focused on right-wing and Islamic extremism.
1
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 2
are considered to be susceptible since they can be easily reached via the extremist
online offensive.
Moreover, the lack of geographical boundaries involves the danger of a new type
of indoctrination that differs from the traditional forms as known from history
(e.g. Naroidnaya Wolya) and found in today’s separatist terrorist groups such as
the ETA (Hoffmann, 2006). Traditionally, their members have been raised on-
site and know the causes they fight for by own experiences. They have been taught
to obey the ideology of the terrorist group in their immediate surroundings, a pro-
cess which can be classified as “embedded indoctrination”. However, in the last
decade a new type of terrorist organization has raised the Western’s focus of atten-
tion: The so called “home-grown terrorists” (Wilner & Dubouloz, 2009). In con-
trast to perpetrators, who were socialized and act in current war zones in Iraq or
Afghanistan, home-grown terrorists have grown up in Western societies, and
“autonomously” act according to their extremist ideology (National Coordinator
for Counterterrorism, 2009a, p. 52). We refer to this process as “distributed in-
doctrination” as it has been assumed that it is directed and facilitated by a stream
of extremist propaganda material accessed through the internet (e. g., National
Coordinator for Counterterrorism, 2009a).
Prominent cases of Western terrorists have been interpreted by the mass media as
evidence for this distributed indoctrination process (Flade, 2011). One example is
the case of Arid U., a young German extremist who shot down two U.S. soldiers at
Frankfurt airport (on their way to Afghanistan) in March 2011. Arid U. claimed
YouTube videos featuring U.S. violence and lootings in Afghanistan to be central
triggers for his deed.
Besides serving as possible trigger, the internet also allows perpetrators to present
themselves and their deeds to “the world”. For example, Mohammed Mehra, who
shot down soldiers and Jewish children in Toulouse in March 2012, filmed his at-
tacks in order to “proudly” present them to the global community. This modern
interplay between extremism and propaganda, the so-called propaganda of the
deed (Bueno de Mesquita & Dickson, 2007) is not only found for Islamic extre-
mists. Right-wing extremist examples are Anders Breivic, the self-proclaimed
right-wing crusader who killed 77 mainly young people in Norway in July
2011, as well as the National Socialist underground terrorists who reached public
awareness in Germany a few months later. Both propagated their deeds and mes-
sages via YouTube videos and online publications.
In these anecdotes, in particular extremist propaganda via audio-visuals played a
relevant role in all these cases. From a media psychological perspective, videos
are a perfect tool for propaganda. The technical development makes their record-
ing increasingly easy and the simplicity of uploading them on YouTube allows the
fast contact with a mass audience. Nevertheless, single cases which are retrospec-
tively analyzed after an attack are not able to depict causal factors. Even though in
all cases internet videos were present, research on media effects has constantly
2
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 3
shown media reception to be only one out of many factors affecting attitudes and
violent behaviors (Ferguson, 2007). It is thus implausible that a few videos are
able to transform a so-far normal young adult into a violent terrorist . Nevertheless,
they might contribute to a climate of receptivity for extremist messages.
To makevalid conclusions about the contributions of internet videos, experimen-
tal studies are necessary in order to test the prerequisites leading to an extremist
attitude or even extremist behavior. However, research testing these influences
and conditions is still seldom if existent. To close this gap, the current book ex-
amined extremist propaganda and its effects by applying established communica-
tion, media and social psychological theories and research methods.
In a first step, we conceptualized propaganda as a form of advertising in which an
extremist ideology instead of a product is being sold (O’Shaughnessy & Baines,
2009). As advertisements, extremist internet videos also aim at finding new
“buyers” for the ideology they “sell”. Research on advertising however depicts
the multiple steps a “product” has to climb before it will be bought. One of the
classic models of this research field is the AIDA Model by E.St. Lewis (1903).
He stated that a “product” (such as the extremist ideology) has to first get Atten-
tion, evoke Interest and create Desire before Action (e.g., “buying”) might follow
(E. St. Lewis, 1903). The hierarchical organization of the model implies that each
step has to be preceded by the former(s), but does not necessarily lead to the latter.
Applying this idea to the reception of extremist videos, the ideology first has to get
attention and create interest before the desire to join an extremist group can spread
and finally, an extremist behavior might follow. A direct leap from attention to ac-
tion is–in general-very unlikely. Though this contradicts the medial interpretation
of videos as causal factors in a radicalization process, raising attention and inter-
est are the ground for desire to sprout.
Until now, little is known about the early stage effects of extremist videos and their
potential to increase attention and interest in the underlying ideology. Instead, the
majority of studies focused on the later steps and concentrated retrospectively on
extremist actions. The mechanisms of the preceding steps and their preconditions
remained unclear.
Taking a media and communication psychological perspective, there are many
unanswered questions concerning the effects of extremist propaganda videos
which should be addressed experimentally. Since right-wing and Islamic extre-
mist formats are the most frequently uploaded ideologies online, it is still an
open question whether these two differ systematically with regard to their appear-
ance. Which means do they use to “sell” their ideas? It is still questionable
whether such videos might be able to raise interest or even to influence the atti-
tudes of a broad mass audience. Which emotions and cognitions are sponta-
neously triggered? Which role does the message style play for the effects? Be-
sides, differences between the extremist videos; which factors on the recipients’
side influence the receptivity for extremist propaganda? Is there, for instance, a
3
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 4
difference between recipients with the same cultural background as the propaga-
tor and those with a distinct one (e.g. Germans for German right-wing extremists
vs. for Islamic extremists)? It is as well possible that the educational level of the
recipients influence the potential effect of propaganda videos and that pre-exist-
ing personality and attitudinal variables shape the pattern.
All these questions point to the necessity to analyze extremist internet videos and
their psychological effects from a media and communication psychological per-
spective. In order to understand the impact of a message, the full spectrum of a
communication has to be analyzed. As Lasswell already stated in (1948) it has
to be analyzed “Who says what to whom in which channel and with what effect?”.
The current book and the hereby presented studies were structured according to
this formula: Right-wing and Islamic extremists were identified as main senders
spreading their ideology via online videos to their target audience (the receivers).
The studies aim at identifying the effects of these different aspects. Focusing on
the not (yet) radicalized potential ingroup, the subtle early stage effects of extre-
mist internet videos were analyzed. To make these early stage effects observable
three main guidelines for our research emerged: (1) the systematic description and
analysis of current extremist internet videos, (2) the cautious assessment of rele-
vant factors on the recipients’side, (3) the usage of reliable methods from media,
social and communication psychology allowing to detect even the subtle immedi-
ate effects of extremist internet videos.
Our large-scale study consisted of four consecutive parts. First, we systematically
described the variance in the appearance of right-wing and Islamic extremist vi-
deos. Different production styles and narratives were analyzed, compared and ca-
tegorized into different extremist “genres”. The method of Ethnographic Content
Analysis (ECA, Altheide & Cheney, 2006) allowed us to categorize a large data-
base of extremist propaganda in German or with German subtitles.2We identified
the same four prototypical categories for right-wing and Islamic extremist propa-
ganda depicting the current “genre” spectrum. Prototypical scenes for each of the
categories were then presented to 450 un-radicalized participants.
It is obvious that right-wing and Islamic extremist propaganda differs in terms of
the cultural background of the potential ingroup they address: While German
right-wing extremists try to mobilize Germans (not, i. e.,Turks), Islamic extremist
propaganda tries to recruit Muslims. To account for effects of a potential match
between cultural background of the extremist sender and the recipient, partici-
pants with distinct cultural backgrounds (in terms of national origin and religious
orientation) were invited and were confronted with right-wing as well as Islamic
extremist videos. As education has often been conceptualized as a crucial factor, a
student and a non-student sample were compared. Besides these broad classifica-
2 The database was kindly provided by the Federal Criminal Police Office (more precisely the Joint
Internet Surveillance Centre in Berlin (GIZ)) as well as the Network for Youth Protection (jugen-
dschutz.net).
4
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 5
tions we were also interested in more individual factors of the recipients. In order
to consider potential influences related to personality and attitudinal factors, a
large set of recipient variables was thus included.
We pictured the potential effects by means of an innovative multi-methodological
approach, combining established social psychological measurements and vali-
dated techniques from media psychology to detect subtle immediate effects. Stu-
dies I and II focused on the so-called explicit or conscious effects (during and after
the exposure to the video clips) by combining psycho-physiological and real-
time-response measurements during the reception (Bente, Aelker, & Fürtjes,
2009) and self-reported emotional and cognitive evaluations after each video.
Study III complemented the pattern by using an implicit association test (Kar-
pinski & Steinman, 2006) to investigate the unconscious or implicit effects pro-
paganda videos might have. This methodological combination allowed insights
in the immediate subtle processes triggered by extremist internet videos. Our re-
sults showed that propaganda mostly led to adverse reactions. The potential in-
group was not attached to extremist messages and in turn, evaluated them nega-
tively. However, some risk factors related to propaganda itself, as well as to the
recipient, shaped the amount of rejection.
Using the words of Paul Watzlawick introducing this chapter, we aimed at contri-
buting to an understanding of the “secret of propaganda”. Instead of focusing on
those already “saturated”, as Watzlawick would call it, wewanted to describe the
boundary conditions which might enhance receptivity for extremist messages in a
young mass audience inside Germany.
The current book therefore provides a systematic description of extremist propa-
ganda (Chapter 2), its right-wing and Islamic extremist senders (Chapter 3), the
formats or genres of audio-visual propaganda (Chapter 4) and the relevant char-
acteristics of the receivers (Chapter 5) as well as first studies on the effects of such
videos (Chapter 6). Using this theoretical background, or research foci are derived
(Chapter 7). In the second part of the book, a new empirical approach to look at
propaganda effects is presented. We intended to deliver first insights into the psy-
chological effects in terms of emotional and cognitive responses towards right-
wing and Islamic extremist propaganda and the relevant factors shaping them.
We hope to stimulate further research and to give first hints for practitioners de-
veloping ideas for intervention and prevention.
5
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 6
Part I Propaganda in the new Millennium
2. What is Propaganda? A Development Towards Modern Media
Applications
“Propaganda is about talking grievance into people, as Hitler did to the Germans.
There is always the harsh residuum of grievance in any society that can be
exploited by a demagogue. Nor does propaganda necessarily make, or ever did
make, the mistake of asking for belief: it is sometimes merely an invitation to share
a fantasy [. . .]. What has changed of course is the range of mediums colonised by
propagandists [. . .].” (O’Shaughnessy, 2012, p. 38)
The spreading of propaganda messages, as well as the theories which try to ex-
plain their appeal, has a long history. However, especially for extremist groups,
the use of new media and new formats considerably changed the scope and the
possibilities of propaganda during the last decade.
The current understanding of “propaganda” has been developed and shaped
throughout history in interaction with religious, societal and technical changes.
Independent of its current form, propaganda is still rooted in the same principles.
The term itself is derived from the Latin word “propagare” meaning “to spread
plants to new locations” (Rogers et al., 2007, p. 67). It was first used during the
Thirty Year’s War by Pope Gregory the XV. who issued the “Congregatio de Pro-
paganda Fide” (= congregation to spread the Catholic faith) in 1622 as a reaction
to the protestant reformation. During World War II, the term was substantially
coined by the German propaganda minister Goebbels (Taylor, 2003) who stated:
“We have made the Reich by propaganda.” Looking back, the development of pro-
paganda and its application had a clear focus on war and issues related to war. Ber-
nays (1942) even stated that during WWII, “ideas became weapons too” (p. 236).
As O’Shaughnessy (2012) formulated it: “What we see [. . .] is an evolution of pro-
paganda through war, revolution and the retailing of charisma” (p. 31). It is thus
no surprise that propaganda has been associated with psychological warfare. The
“battle of ideas” (Bernays, 1942) was fought by every official institution or gov-
ernmental party during the 20th century, reaching its climax in the era of the Cold
War (e.g.,Taylor, 2003). Politicians were aware of the fact that “winning the minds
of civilians is [. . .] crucial to winning or losing a war” (Finch, 2000, p. 372). The
mainly negative connotation that is inherent in the term “propaganda” to day is due
to this historical association (Rogers et al., 2007).
Although there is no common definition of propaganda, some core ideas derived
from this historical development are consensual: For instance, Lasswell (1927) al-
ready described propaganda as the “management of collective attitudes by mani-
pulation of significant symbols” (p. 627). This aspect of collective manipulation
6
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 7
has been further unfolded in the def inition by Jowett and O’Donnell (2012) we will
refer to throughout this book. They defined propaganda as “the deliberate, sys-
tematic attempt to shape perceptions, manipulate cognitions, and direct behaviour
to achieve a response that furthers the desired intent of the propagandist” (p. 7).
Since the first studies on war propaganda in WWII this desired response remained
the same: “(1) Heighten the moral unity of your own [people], (2) weaken the
moral of your enemy and (3) win over the moral of the neutrals”(Bernays, 1942,
p. 237). From a social psychological perspective this can be understood as an in-
creasing cohesion among one’s ingroup, weakening the cohesion of the outgroup
and win the potential ingroup. Research in the area of Social Identity Theory (SIT
Tajfel & Turner, 1986) has largely described the importance of ingroup/outgroup
categorizations as part of one’s self-concept. These categorizations were found to
influence people’s motivation, emotion and behavior as people prefer to see them-
selves in a positive light (Branscombe & Wann, 1994; Fein & Spencer, 1997).
Wining over the neutral potential ingroup is considered to be of particular interest
for researching the effects of extremist internet videos. Besides the “own people”
who already feel to be part of this extremist ingroup, the worldwide web confronts
numerous so-far neutral recipients with extremist recruiting attempts. Basic stra-
tegies which are used to motivate this potential ingroup remain the same irrespec-
tive of online or offline implementation. As Lasswell (1927) already described it,
four basic strategies can be distinguished: The recipient of the message has to be
convinced that (a) the guilt lies on the side of the enemy (implying that there is a
defined enemy), (b) that this enemy is “incorrigible, wicked and perverse”
(p. 77), and that (c), nonetheless, there is hope of victory when (d) all allies stand
together (Lasswell, 1927). Merten has further structured this basic pattern into
five aspects (Merten, 2000, p. 153):
(1) Unique characteristic: An arbitrary object (idea, action, person or product)
is propagated and gains a unique characteristic.
(2) Behavioral premise: For those who consume propaganda, a behavioral pre-
mise is predetermined which holds a characteristic of exclusiveness.
(3) Exclusiveness (pre-decision): The recipient of these messages is forced to
obey this behavioral premise. This leads to the divestment of freedom to
make own decisions.
(4) Sanctions (future-directed): To have the recipients obey these pre-decisions
unconditionally, the propaganda messages sketch positive and especially ne-
gative sanctions.
(5) Non-verifiable arguments: Propaganda messages should be formulated in a
way that makes them unverifiable. This is mostly realized by emphasizing ac-
tions which may happen in the future – either threats or eschatological wis-
doms.
7
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 8
While the basic aims and the underlying structure of propaganda remained similar
over the time, its presence and face has substantially changed during the centuries.
Throughout history, the possibilities for dissemination and reception of propa-
ganda have evolved with every societal and technical improvement in communi-
cation (O’Shaughnessy, 2012) extending temporal and geographical limitations.
Via the construction of cities and the dissemination of literacy propaganda
reached a larger mass audience. The invention of print further enabled propagan-
distic messages to be available over time. These factors together allowed a fast dis-
tribution to a large geographically and temporally distal audience (Taylor, 2003).
Later, the invention of the telephone and the mechanical recordings extended the
range of propaganda from words and pictures to the human voice.
A qualitatively distinct leap in the distribution of propagandistic messages was the
invention of film and cinema. As O’Shaughnessy (2012) claimed: “Propaganda
which speaks the language of propaganda fails to penetrate cognitive defenses.
The best propaganda is disguised as entertainment”. And “Cinema is manifestly
the finest medium for propaganda” (p. 34). It is not surprising that since the be-
ginning of the cinema era, professional filmmakers such as Leni Riefenstahl used
their influence on a broad audience3for propagandistic purposes. In particular,
Hollywood productions from the 1950s/1960s substantially contributed to the
American fear of a communist infiltration as depicted in movies such as “Amer-
ican Guerrillas at the Philippines” by Fritz Lang (1950) or “the Manchurian
Candidate” (1962) by John Frankenheimer.
Consequently, the development and the dissemination of the internet in the second
half of the 20th century can be regarded as the latest evolution of propaganda. The
internet not only combines all techniques used before, it also eludes national bor-
ders and governmental control, enabling propagators to spread uncensored mes-
sages and pictures to an enormous audience. Furthermore, with the implementa-
tion of Web 2.0 applications, (Münker, 2009) people have got the chance to be
propagators themselves.
By now, this idea has been professionalized: Inter net propaganda changed from an
instrument to convince the masses to an instrument used by the masses
(O’Shaughnessy, 2012). Social network sites (such as youtube), online forums
and blogs enable everybody to produce and upload their own propaganda material
(English, Sweetser, & Ancu, 2011; Holtz, Kronberger, & Wagner, 2012;
O’Shaughnessy, 2012). Even the production of own videos and films has become
increasingly easy with the distribution of computers, cheap cameras and smart-
phones. Furthermore, the seemingly user-generated content in web 2.0 applica-
3 Riefenstahl has been controversially discussed as creator of Hitler’s fascist aesthetic in her films and
because of her closeness to the Nazi regime. See e.g. http://www.abendblatt.de/kultur-live/arti-
cle814253/Ein-Genie-auf-dem-Irrweg.html Retrieved online, 26/06/2012.
8
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 9
tions provides a new form of credibility for professionally produced statements as
they seem to come from the “average users” too.
To sum up, the main aspects, aims and structures of propaganda have remained the
same over time. The guise however has changed with the technical developments
over the centuries and especially during the last decade. In particular, extremist
groups have discovered these benefits for their desired “grass-root movement”
(Sageman, 2008). “Internet is viewed as a perfect medium, one with a far reach
that affords anonymity and where recruitment becomes self-recruitment.” (Seib
& Janbek, 2011). Among them, right-wing and Islamic extremists form the two
biggest groups, particularly when it comes to the production and dissemination
of extremist internet videos (Europol, 2011b; Glaser, 2011; Torres-Soriano,
2010).
9
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 10
3. Extremism in the 21st century
“Extremists conquer the internet” (Pfister, 2012)
The terms “extremism” and “extremist” are considered to be ill-defined. In every-
day language they are often used interchangeably for radicalism, fundamentalism
and even terrorism. From a scientific perspective nonetheless, there are substan-
tial differences between these terms. Thus, for the purpose of a scientific study, it
is necessary to distinguish them. Hence, a short def inition of our understanding is
given so to classify who these “conquerors”, Pfister talked about, are.
In a broader sense, radicalism refers to an attitudinal pattern that aims at chan-
ging societal conditions at their roots (radix = lat. “root”). In a narrower perspec-
tive, radicalism often refers to intolerance towards other attitudes and to the demo-
cratic idea of equality4.Fundamentalism, in contrast, describes the insistence on
firm political, religious or traditional values. Religious traditions are often lit-
erally interpreted and understood as universal solutions for all problems5. Conse-
quently, both terms describe attitudinal patterns which do not necessarily conflict
with a democratic society.
Extremism, derived from the Latin words “extremus/extremitas” (the outmost/
the most dangerous), is often used to describe political attitudes at the uttermost
parts of a left to right-wing spectrum. This implies a certain variance over time
regarding which concrete attitudes or behaviors are classified as extremist. How-
ever, the term also includes actual attempts and behaviors “behind the democratic
line”. In their definition, the German Federal Office for the Protection of the Con-
stitution mainly focuses on extremist efforts against the state, enunciating that ex-
tremism refers to any illegal attempt against the free democratic basic order, ad-
ministration or security (Neugebauer, 2010, p. 5).
In the following we will refer to a definition by Kemmesies (2006) which also in-
cludes the extremist worldview. He specified the term as referring to “any attempt
being in the widest sense politically or religiously motivated and oriented towards
ideologies, that interprets societal conditions in the ‘only truth manner’ with the
intention to radically change the given conditions by accordant strategies” (own
translation, Kemmesies, 2006, p. 11).
Following Jesse (Bundesministerium des Inneren, 2004), extremist groups can be
distinguished by their aims (left vs. right-wing) as well as by their degree of orga-
nization (party vs. single individuals), and the intensity of extremist convictions
and their means (rejection of political violence vs. terrorism). As a rough differ-
4 Retrieved online, http://www.bpb.de/nachschlagen/lexika/politiklexikon/18088/radikalismus,
05/06/12.
5 Retrieved online http://www.bpb.de/nachschlagen/lexika/politiklexikon/17513/fundamentalis-
mus, 05/06/12.
10
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 11
entiation, mostly left-wing, right-wing and Islamic extremists are distinguished as
major threats inside Germany (Otto Schily cf. Bundesministerium des Inneren,
2004)
While the term extremist encompasses those rejecting political violence as well as
those who support political violence, the direct advocacy of terrorism is a special
case. Terrorism, according to Hoffmann (2006), can be understood as “propa-
ganda of the deed” (Bueno de Mesquita & Dickson, 2007). The understanding
of terrorism as a spectacular act of violence pronounces the usage of pictures –
broadcasted via mass media – to be relevant in the context of modern propaganda
(Hoffmann, 2006). For instance, the attacks of 9/11 are the most famous prototype
of propaganda of the deed. “It was a classic asymmetric attack. Moreover, in the
16 minutes between the strikes on the first and second towers, New York’s news-
rooms had scrambled their helicopters and were able to capture the second plane
hitting its target live to a global audience” (Taylor, 2003, p. 315). Even though
terrorists have always tried to communicate their aims via attacks that attract at-
tention (Hoffmann, 2006), the internet has opened up new ways to present them to
the “global community”.
Most importantly, with the new media, the attacks can be followed live and online,
which heightens their impact. Again right-wing and Islamic extremists are the
most salient groups concerning the violence committed, the danger discussed
(Europol, 2011a), and the online distribution of the resulting pictures. Anders
Breivic preliminary announcing his cruel plans online and Mohammed Mehra
filming his kills (see also Chapter 1) are two examples for the interplay of actual
propaganda of the deed and its online distribution.
To our understanding, committing an attack refers to an act of terrorism. In con-
trast, the distribution of such pictures and their propagation is considered to be an
extremist action. In consequence, the book will rely on the term extremism as our
focus lay on the effects of extremist internet propaganda on young adults apart
from the terroristic pole of the extremist spectrum
3.1 Islamic Extremism
“For Islam is power, Jihad and courage, unity and brotherhood, and the path of
Muhammad.” (“Destroy Night’s Dark Injustice” cf. Kimmage & Ridolfo, 2007)
Since the attacks of 9/11 and the resulting “war against terror” declared by former
U.S. president George Bush, the Western world has constantly been informed
about the threat ensuing from Islamic extremists (Frindte & Haußecker, 2010).
Single cases, like Arid U. and Mohammed Mehra, keep the threat salient. They
also lead to some fundamental misinterpretations of Islamic extremism as “typi-
cal for the Islam” (Fischer, Greitemeyer, Kastenmüller, Frey, & Oßwald, 2007). In
sharp contrast to the religion, Islamic extremism is a religiously motivated and
allegedly legitimate form of political extremism (Puschnerat, 2006). Though it
11
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 12
is difficult to clearly define the ideologies and groups that are subsumed under the
term, it basically includes certain fundamental or radical aims and the will to use
extremist means to reach them (see also the definition of extremism by Kemme-
sies, 2006).
Islamic extremist organizations emphasize both the Sharia and the Jihad as cr ucial
components: Originally, the Sharia described the God-established order in Islam.
Islamic extremists misuse the term to militantly claim the implementation of Is-
lamic law (Schubert & Klein, 1997). By referring to it, they create a utopia of one
Islamic state legitimized by Allah and realized by the Ummah (the Muslim com-
munity). With regard to Jihad, the Western media, as well as Islamic extremists
themselves, equate it with the “holy war” although it literally translates into “ef-
fort” and refers to war mainly on a metaphorical level (Burke, 2004; Thackrah,
2004). Generally the “great Jihad” as (peaceful) struggle for a worthy cause
and the “small Jihad” as armed fighters can be distinguished (National Coordina-
tor for Counterterrorism, 2009b). However, the interpretations of the concept are
used in varying ways.
The most prominent representatives of Islamic extremism is Al’Qaeda, (Hoff-
mann, 2006). From their perspective, the “occupation” of the Arab peninsula
(the former territory of the Muslim caliphate) and the spreading of Western values
like individualism or materialism have alienated the Ummah from the rules of Is-
lam. Therefore, they justify violent Jihad as a way back to God (Hoffmann, 2006;
B. Lewis, 2003). The Islamic extremist ideology thus creates a second enemy
stereotype besides the Western infidels – the apostate Muslims (Wegner, 2009).
Islamic extremists seek to cleanse Muslim territories from all Jews and Crusaders
and generally from all Western influence (Hoffmann, 2006). The overall long-
term goal is the establishment of the worldwide Islamic caliphate, middle-term
and short-term goals are the elimination of all corrupt secular regimes and the mo-
bilization of the Muslim community. To do so, Islamic extremists try to elevate
Jihad to a global issue of all believers. However, the concrete implementation
of how the “war of ideologies” should be won is left rather vague (Egerton, 2011).
Al’Qaeda (= arab. the foundation) (Sarangi & Canter, 2009) originated from the
Afghan Service Bureau (Khidmat lil Mujahidin al Arab) but was changed to Al’-
Qaeda in 1988, run by Osama bin Laden. By now, Al’Qaeda is present in many
countries on different continents and is structured in a decentralized way, operat-
ing as a connected network (Thamm, 2005). Local groups act independently and
organize terroristic attacks, such as the attacks in Madrid in 2004 or the bombings
in London one year later. Consequently, Al’Qaeda is also one of the biggest pro-
ducers of “propaganda of the deed” material (see Chapter 3.1). Their attacks aim
at threatening all nations and institutions that do not follow their interpretation of
Islamic law (Thackrah, 2004).
For this global presence it is necessary that Al’Qaeda’s members (inter-) act anon-
ymously and clandestinely within their homelands, thus creating a globalized
12
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 13
grass-root movement of Islamic extremism (Sageman, 2008). Particularly over
the last years, Al’Qaeda has more and more transformed from a clear, identifiable
organization to an ideological movement hard toget hold of (Schneckener, 2006).
The call by the important salafist6strategist Al-Suri for a “Global Islamic Resis-
tance” (Lia, 2007a) is mostly answered by scattered individuals instead of defin-
able groups directly associated with Al’Qaeda (Schneckener, 2006).
He delineates the attacks by Al’Qaeda to “a call for action and martyrdom” (cf.
Lia, 2007b, p. 438) and seeks to persuade every Muslim of the need to proactively
join (Holtmann, 2009). Particularly, the on-going attempt to recruit new followers
who are willing to sacrifice their lives is of major importance. Even though mem-
bers for other purposes are needed, suicide attacks cause the death of the perpe-
trator (see also Hoffmann, 2006) and can therefore be traced back to one single
commitment of the corresponding martyr.
Hegghammer (2006) distinguished three different types of Islamic extremists: (1)
ideologues, whose function is to provide the group with an ideological and argu-
mentative background. Typical examples are Osama bin Laden or Ayman Al-Za-
wahiri (the leader of Al’Qaeda since the death of bin Laden). (2) Lifestyle-Jiha-
dists, who fight in the war or act as commanders in Al’Qaeda training camps
and (3) the martyrs who sacrifice themselves for the goals of the organization.
As their use of violence hits innocent people, (which is clearly not intended in the
Qur’an), Islamic extremists are challenged with gaining acceptance and legitimi-
zation of their actions (Corman et al., 2006). Therefore, they inter pret the holy war
as a defensive fight against a superior adversary. Propagation is the crucial means
to “sell their movement” and gain new followers (Rogan, 2007). This is mostly
achieved via the internet (see Chapter 4.1), Al’Qaeda’s “electronic think tank”
(Musharabash, 2006). Although Sageman (2004) claims personal connections
to Al’Qaeda and Jihadists to be more important to explain why people join
them, successful propaganda in order to recruit new members and to mobilize
them is a necessary component of Al’Qaeda’s personal reproduction (Wegner,
2009).
6 Salafism: Fundamentalist ideology that advocate a traditional interpretation of Islam. New devel-
opments within the Islam are strongly rejected, as are Western values. Retrieved online: http://
www.tagesschau.de/inland/salafismus100.html 24/06/12.
13
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 14
3.2 Right-Wing Extremism in Germany
“Even if all silence, when they drift into their harm. We will protect the county-
national Socialists. Bring the fight to the streets, here and now. National socialism
lives on in us.” (“Haftbefehl” cf. Landesamt für Verfassungschutz, 2003)
Global attention in the last years has surely focused more on the overall threat by
Islamic extremists. However, this bias in media reports and societal awareness
does not reflect well on the problem of right-wing extremism. The case ofAnders
Breivic, and his “crusade against multiculturalism”, and the series of terrorist at-
tacks in Germany discovered a few months later, have once again drawn the atten-
tion of the public towards the outlasting threat right-wing extremism poses.
Besides these single cases, right-wing parties as well as youth organizations are
constantly present in Germany. Their appearance however has changed in recent
years: While “traditional” skinheads have become rare, the autonomous, action-
oriented right-wing youth-culture has constantly grown (Ruhe, 2010)
The definition of right-wing extremism is controversial. There are numerous
terms and diverging definitions, often synonymously used for the radical right,
extreme right or right-wing radicalism. Based on the aforementioned definition
of extremism by Kemmesies (2006), “right-wing extremism” means the co-ap-
pearance of radical attitudes and the usage of extreme means. This is in line
with Heitmeyer’s definition (1992), who also differentiates between right-wing
extremist attitudes and right-wing extremist behavior. Right-wing attitudes are
the advocacy of non-equivalent ideologies. Value differences between people
and groups based on race, nationality or religion can be summarized here, as
well as the marginalization of “un-worthy” others. The affinity for dictatorships,
the downplaying of the Nazi regime, anti-Semitic, anti-Muslim, xenophobic and
social-Darwinist attitudes are included (e.g., Decker, Weißmann, Kiess, & Bräh-
ler, 2010).
For right-wing extremist behavior, the acceptance of violence as a means to solve
conflicts also plays an important role. According to Heitmeyer (2002), the amount
of violence acceptance escalates from the acceptance of the immutability of vio-
lence to the actual execution of accordant behavior.
Both components are clearly anti-democratic and against the constitution. Right-
wing extremism, in thought and deed, is only given when the two co-occur. This
happens either by using right-wing extremist orientation for legitimating an esca-
lating acceptance of violence or by accepting violence as a means to assert the
ideology behind it (Heitmeyer, Borstel, Grau, & Marth, 2010). Both are impossi-
ble to differentiate retrospectively. Nevertheless, it has to be stated that right-wing
attitudes are more frequent than right-wing violence (Decker et al., 2010).
Concerning the actual “offer” of right-wing extremism in Europe, two main
streams within the scene can be distinguished: On the one hand, there are “tradi-
tional” forms of party organized right-wing groups, such as the German NPD or
14
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 15
the Dutch CP’867(Mudde, 1995).They try to reach the middle of society and in-
filtrate the parliaments. They seek to cooperate with other European or even glo-
balized “white-pride” organizations (Langenbacher & Schellenberg, 2011), for
example, with regard to the campaign “Citizens against Islamization” (Langenba-
cher & Schellenberg, 2011). On the other hand, a noticeable part of the neo-Nazis
sharply criticizes the civic strategies of mainstream right-wing parties as being
too soft.
“[. . .] we don’t believe that the capitalistic system can be reformed or improved –
the prevalent system IS the errorand has to be replaced by a new, more liberal, just
and NATIONAL AND SOCIAL form of society.” (Black Block Berlin cf. Bundes-
amt für Verfassungsschutz, 2008).
The increasing number of these autonomous nationalists in Germany (Bundesamt
für Verfassungsschutz, 2008) is far less ideologically rooted but more experience-
oriented. Addressing actual socio-political topics, they offer joint leisure time ac-
tivities which are similar to those of other subcultures, e.g., the spreading of flyers
or the drawing of graffiti.
Particularly, youth-oriented video propaganda is often produced by and for auton-
omous nationalists (Glaser, 2011). Strategies from the black block8and symbols
from left-wing activists are adopted and used for various militant actions. This
“political mimicry” (similar to the strategy of right-wing parties) aims at opening
the way towards the discontent middle of society (Müller & Seiler, 2010). Unsur-
prisingly, the internet plays a major role in propagating these ideas and reaching a
huge audience. It provides an important means for the ongoing presence of right-
wing extremism. As a right-wing extremist formulated it in an Email to Thomas
Pfeiffer in 1996: “The Internet is cheap, fast and clean. We love it” (Pfeiffer,
2003).
7 Party is prohibited since 1998 (Mudde, 2000).
8 Black Block: A group of protesters that seems homogenous because of similar behavior and same-
styled black disguise. The disguising shells hamper identification through safety authorities. (Re-
trieved online http://szenesprachenwiki.de/definition/schwarzer-block/http://szenesprachenwi-
ki.de/definition/schwarzer-block/March, 13th, 2012.)
15
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 16
4. Virtual Propaganda and Extremist Videos
4.1 Right-wing and Islamic Extremists in the Internet
“It is obvious that the media war in this century is one of the strongest methods; in
fact, its ratio may reach90 % of the total preparation for the battles.” (Osama bin-
Laden cf. Corman et al., 2006, p. 3)
The internet encompasses various characteristics that can be used as dangerous
instruments or even weapons of terrorism (Hoffman, 2006). It’s de-centrality
allows access from nearly all over the world, its anonymity hampers legal inves-
tigations, and its global nature distances it from the laws of single countries.
Furthermore, it is incredibly fast and able to transport enormous data packages.
Especially for the younger generations, the digital natives (Ebert, Feierabend,
Karg, & Rathgeb, 2011), it has become a natural living-space, contributing to
own identity development (Schmidt, Lampert, & Schwinge, 2010; Schmidt,
2009)
As was mentioned before, extremist groups have discovered these benefits; espe-
cially because they might be under constitutional supervision (Hoffmann, 2006).
Everybody is able to join via an internet connection; there is no need to leave the
home country or to arrange dubious encounters in order to get in touch with ex-
tremists. In particular, the base right to freely express one’s own opinion can be
acted out without limitations, creating a space of impunity (Inan, 2007). The in-
ternet is used by extremists to communicate with their own people (ingroup), their
enemies (outgroup) as well as to other internet users (potential ingroup and global
community) (Payne, 2009).Ingroup communication, such as training instructions
or planning, is often uploaded in restricted areas such as closed forums (Busch,
2005). For recruiting purposes in contrast, material that is freely accessible on
homepages, social media applications or open discussion forums plays an enor-
mous role. This material is often unobtrusively hidden in “normal” contents
(wolf-in-sheep’s-clothes strategy) and may reach numerous internet surfers who
might stumble over it. In line with this, extremists ’digital communication is often
described as propaganda offensive (Puschnerat, 2006), attention-seeking terror-
ism or the marketing of fear (Weichert, 2007).
This explains why contact with extremist groups is often realized via the internet
(Bundesamt für Verfassungsschutz, 2009). Even in the few cases when public
authorities are able to legally delete extremist content, such content has most
likely already been uploaded to another website (Seib & Janbek, 2011). This
makes the deletion of web content a long and difficult fight (Wortmann, 2007).
Not surprisingly, objective measurements indicate an increasing extremist pre-
sence and activity on the World Wide Web. Two years ago there were already
over 100 English websites featuring Islamic extremist content. Moreover, the pro-
ducers explicitly asked their users to translate these websites into as many lan-
16
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 17
guages as possible (Seib & Janbek, 2011). While in 2001, Al’Qaeda published 19
virtual documents, this figure increased to nearly 1000 documents uploaded by
Islamic extremist groups in March 2007 alone (Torres-Soriano, 2009). In 2010,
the State Office for the Protection of the Constitution in Sachsen noticed approxi-
mately 1000 right-wing extremist websites which were run from inside Germany.
This number has also increased (Europol, 2011a; Glaser, 2011). Extremists them-
selves describe the relevance of a purposeful media application as “a race for the
hearts and minds of our people” (Al-Zawahiri cf. Lia, 2007b).Three main com-
munication goals can be distinguished: (1) Legitimization, (2) Propagation, and
(3) Intimidation (Bockstette, 2008; Corman et al., 2006). While the latter clearly
addresses the enemy, the former two are directed towards the already radicalized
ingroup but also aim at gaining new members and legalizing the fight in the eyes
of the global community (Hoffmann, 2006; Payne, 2009). Weimann (2006) has
complemented this list by adding the possibility of collecting information via
the net (e.g. about the enemy) and easily forwarding it to one’s “colleagues”,
thus facilitating concrete operations. Instructions about how to produce explo-
sives and incendiary agents (Bundesamt für Verfassungsschutz, 2009; Landesamt
für Verfassungschutz, 2003) as well as whole “university like studies on extre-
mism” (Musharabash, 2006) are available.
Via these internet strategies, Al-Suri’s dream of a “grass-root” movement of the
Jihad (Lia, 2007b), a dream also proclaimed by Klu-Klux Klan9Leader Milton
Kleim Jr (cf. Bush, 2008), is realized. “In an individual, secret jihad, the opera-
tional activity also takes place on a global and universal [. ..] horizon. The hor-
izons for this activity open up regardless of borders and countries” (Al-Suri cited
from Lia, 2007b, p. 369).
Tosum up, the internet makes a great contribution to this grass-root movement of
collective thinking and individual acting. According to Sageman (2008), Al’-
Qaeda has already transformed into a social movement with participants rather
than members. Besides providing a terrorist agenda, information, instruction
and exchange, it is the “virtual glue” (Sageman, 2008, p. 144) that is transported
online. The extremist organization itself as well as its affiliated groups became
connected to a global and ideological homogenous Islamic extremist movement,
called the “leaderless jihad” (Sageman, 2008) or “cyber-jihad” (Engelmann et
al., 2010, p. 33). Even though globalized white-power movements are still
more seldom, a development in the same direction is apparent (Verfas-
sungsschutz, 2004). The National Coordinator for Counterterrorism in the Neth-
erlands (2009a) assumed that the Internet “supports the entire radicalization pro-
cess” (p. 52). Four specific goals are inherent in this strategy: (1) Indoctrinate the
youth, (2) foster the meeting with similar others, (3) create a strong community
and (4) plan and commit own actions.
9Klu Klux Klan: A racist secret society in the USA, advocating a mixture of right-wing and reli-
giously disguised messages. The clan is known for his violence (Wagner, 2008).
17
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 18
The indoctrination of the youth is pronounced by both Islamic and right-wing
extremists. For example, Al-Suri claimed a well-considered application of jihadist
propaganda to be able to activate many Muslims (Holtmann, 2009). Particularly,
the youth shall be addressed for an early comprehension of the underlying ideol-
ogy in order to ensure a self-reliant movement (Precht, 2007). Analogously, right-
wing extremists offer entertainment and attract the youth by applying their cul-
tural codes and lifestyle (Becker, 2009), creating a “right-wing extremist world
of experience”. The promotion of subsequent offline activities as political or com-
mon leisure activities (such as concerts, sport events, street festivals) may influ-
ence the socialization of the young generation (Bundesamt für Verfassungsschutz,
2008).
Online communities foster the meeting of others who share their own questions,
sorrows or wishes (National Coordinator for Counterterrorism, 2009a). This is
why Milton Kleim Jr., grand dragon of the Klu-Klux Clan, wrote “Crucial to
our [. . .] campaign is that our message is disseminated beyond ‘our’ groups
[. . .].We MUST move out beyond our present domain, and take up positions on
‘mainstream’ groups” (John Milton Kleim Jnr., 1998).
They want to evoke the interest of potential sympathizers who can then be con-
tacted for persuasive purposes. Similarly, fighters within the media jihad justify
and legitimate their movement online to reach young men as recruits (Heggham-
mer, 2006). Sageman (2008) concluded that the influence of interactive (jihadist)
discussion forums or links may be as big as the influence of radical prayers in real
mosques.
Via these activities, extremists create a strong virtual community (Egerton,
2011; Engelmann et al., 2010). Concerning Islamic extremism, one of the most
attractive aspects of using the internet is the ability to get “a concrete example
of what the utopian Ummah can look like: a community of friends [. . .].” (Sage-
man, 2008, p. 117). This experience is expected to motivate and oblige members
in a very specific way, so the Ummah or “this utopia becomes something worth
fighting for” (Sageman, 2008, p. 117). Similarly right-wing extremists try to cre-
ate an attractive cohesive online community (Glaser, 2011). The 3.992 Facebook
fans of the “Nationalistes Autonomes France10” or the members and fans of cor-
responding German sites (Glaser, 2011) seem to depict the success of this aim.
Finally, interested users shall commit own actions with the internet providing a
gateway function. Often, neither the materials uploaded on websites like Wikipe-
dia,youtube,twitter,MySpace or facebook, nor the actions demanded are directly
10 Retrieved online www.facebook.com, 05/05/12.
18
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 19
recognizable as extremist (Schürmann11, 2012 ZDF Mediathek). For example
“liking” fan-sites against pedophilia on Facebook may instead be “liking” the Na-
tional Socialist organization behind it, without (the user) even noticing (Schür-
mann, 2012 ZDF Mediathek). Web 2.0 applications (e. g. music communities,
regional parties, communication networks) therefore perfectly fit the “take up po-
sition” strategy and create easy “take-part” opportunities (Busch, 2008). Content
generated by the individual user, whether it consists of texts, pictures, tones, mu-
sic or even videos, can be produced and uploaded without a sophisticated knowl-
edge (Winn & Zakem, 2009).
This is where audio-visual propaganda gets its outstanding function. Especially,
videos that are uploaded on platforms like YouTube or shared via mobile phones
play a significant role within the extremist web 2.0 offers. Google (2012) stated
that about 60 hours of videos are uploaded every minute on YouTube and a sub-
stantial amount of them (though the actual number is unknown) has been created
by extremist individuals or groups. Glaser (2011) reported that single right-wing
extremist video clips can easily exceed 170.000 clicks.
Audio-visual propaganda appears to be very attractive (see also Chapter 2) and is
able to reach high numbers of recipients (Seib & Janbek, 2011). It can even get the
attention of news broadcasting companies like CNN or Al-Jazeera. For instance,
scaring effects can be achieved via brutal film sequences (e. g. the pictures of
9/11). By distributing them online, these videos get the attention of mass media
broadcasting and thus may frighten the enemy or society even more (Bockstette,
2008).
4.2 Propaganda through Audio-visual Material
“It is a new version of an old activity, terrorism and its novelty is characterized by:
First, the extremism of the terror act itself; second the global reach of such acts
and third, the ability to elucidate the message of those acts both verbally and vi-
sually on a global scale via the universal reach of the internet, [.. .]” (O’Shaugh-
nessy, 2012, p. 37)
Audio-visual propaganda may be the universal tool for transporting the propaga-
tor’s desires. Therefore, although web 2.0 is open to everybody, it is not surprising
that video propaganda is often produced by professional media companies in
order to promote extremist ideologies. Beyond user generated content as for ex-
ample the videos uploaded by Anders Breivic (see Chapter 1) there are numerous
systematically produced and distributed videos conveying a right-wing or Islamic
extremist ideology. In particular, Islamic extremism has highly professionalized
structures (National Coordinator for Counterterrorism, 2009a). In general, there
11 Retrieved online. http://www.zdf.de/ZDFmediathek/beitrag/video/1493124/Netzschau-Subtile-
Rekrutierung#/beitrag/video/1493124/Netzschau-Subtile-Rekrutierung, 24/04/2012.
19
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 20
are three units with distinct functions: (1) militant groupings as footage providers
or creators, (2) media production units, and (3) media distribution units (Kim-
mage & Ridolfo, 2007; Kimmage, 2008). In some cases this segmentation is
not distinctive. For instance Al-Fajr, As-Sahab and the Global Islamic Media
Front (GIMF) are considered to be global operating production and distribution
centers. For right-wing extremism, professional production companies such as
“media-pro-patria” (Müller & Seiler, 2010) also exist.
The enormous amount of extremist videos online has led to a broad variety of ex-
tremist propaganda genres and topics addressed. Former studies have tried to de-
scribe how propaganda 2.0 looks like mostly focusing on descriptions of Islamic
extremist material. Besides first exemplary descriptions of right-wing extremist
videos, formal categorizations have not yet been provided here.
On the one hand, descriptive analyses of Islamic extremist material can be distin-
guished on the basis of content analytical approaches aiming at identifying genres
of extremist audio-visuals (thus encompassing typical plots, styles and supposed
aims) as, for example, provided by Finsnes (2010) the Intel Center (2005, 2006) or
Salem, Reid and Chen (2008).
On the other hand, more detailed analyses of the narrative structure concerning the
symbols used and the messages transported are also available (Combating Terror-
ism Center at WestPoint, 2006; O’Shaughnessy & Baines, 2009; Payne, 2009).
The description of Jihadist genres has mainly been conducted by governmental
organizations aiming at providing operator communities working on counter-ter-
rorism with a professional categorization. For example, the Intel Center (2005,
2006), and the Terrorism Research Group of the Norwegian Defense Research Es-
tablishment (Finsnes, 2010) both reported comparable genres among their large
databases. Accordant with extremist aims, different genres satisfy different pur-
poses. Still, one video can encompass different genres and usually fulfills differ-
ent purposes.
As already described in Chapter 2, propaganda can be directed towards the extre-
mist ingroup, for instance in order to give instructions. It can also address the en-
emy or outgroup in order to intimidate, and the so-far neutral potential ingroup in
order to propagate the own ideas and legitimate the own deeds (Bernays, 1942;
Lasswell, 1927; Payne, 2009). These three basic goals of propaganda (intimidate,
propagate and legitimate) of course also apply to extremist internet videos. For
instance, within former categorization approaches (e.g., Intel Center, 2005) these
goals can also be recognized. The following two genres preliminarily address the
extremists’ outgroup in order to intimidate, whereas the third additionally ad-
dresses their ingroup.
(1) Operational (Documentary) Videos show assaults and assassinations aiming
at threatening the enemy. They usually last only a few minutes (1–8min.) and
are confined to the depiction of attacks. For instance, sequences from remote-
20
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 21
controlled charges that explode when a military vehicle passes by. These se-
quences are often repeated in slow motion. Image marks, like red arrows, are
added to emphasize the detonation. Anasheed (religious battle songs) accom-
pany the images to praise the jihad.
(2) Intimidation is achieved through Hostage Videos and Beheading Videos
which demoralize and deter the enemy. These videos serve as primary tools
of psychological warfare. Through their brutality they warrant extreme media
exposure. The militant group threatens with consequences so that the hosta-
ges set requirements to their governments.
(3) Statement Videos quote high-ranked jihadists. These videos refer to actual
events. The speakers threaten their enemies by emphasizing future attempts
or offering exemptions if certain claims are accomplished. These videos
are directed towards the enemy as well as towards financial donators, poten-
tial recruits and insiders.
The second important aspect is the legitimation of the own deeds, which plays an
important role in all video clips and is not clearly distinguishable from videos with
propagandistic purposes. The subsequent categories encompass this legitima-
tion aspect in addition to propagandistic purposes.
(4) Propaganda/Promotion Videos aim at communicating goals, legitimating the
assaults, propagating the ideology, recruiting new holy fighters and at intimi-
dating the enemy. The speakers are mostly hooded and comment on images
from mass media coverage which show humiliated and abused Muslims,
ruined homes and sequences of fights. The speakers refer to conspiracy the-
ories against Muslims.
(5) Humiliation Videos resemble the Propaganda category in terms discrediting
and taunting the enemy. Similar categories are Ideology and Statement.
(6) Ideology Videos serve as ideological presentations through the speeches of
important leaders. They can last five to 90 minutes. In many of them, only
one subtitled speaker is shown. The argumentation is often based upon poli-
tical conflicts and the individual obligation of a Muslim to participate in the
Jihad. The Qur’an is cited to fundamentally root the argumentation.
(7) Recruitment Videos can be considered rather as more of a function of several
categories than as a separate genre and represent the explicit call to militant
Jihad.
The next genres primarily address those already convinced of the ideology: The
ingroup. They offer concrete instructions or advocate the most extreme form
of personal involvement with Jihad: Sacrificing oneself.
(8) Praising/Tribute Videos serve as commemoration of fallen jihadists and as
an honor of both important leaders who are still alive or of major group los-
21
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 22
ses. Martyr Videos often show the planning of an attack, preparation of the
vehicle and pictures of the final destruction. The martyrs read out their tes-
taments, send their dying words to their families and encourage recruits by
emphasizing the importance of self-sacrifice.
(9) Instructional and Training Videos are produced less frequently. They pro-
vide manuals, instructions and training for the use of arms or preparation
of crafted dynamite.
The intention is not to replace real existing training camps but rather to use
the internet as an additional instrument for self-radicalization. Since the ob-
ligation to (militant) Jihad includes appropriate preparation, training videos
are often used for internal documentation of the drill.
To conclude the categorization, professional style video clips have to be
mentioned as they are the most cinema-like productions available and
thus the best to “disguise propaganda as entertainment” (O’Shaughnessy,
2012, p. 38). These productions however are still seldom compared to the
other formats (Intel, 2006).
(10) These Hollywood-like Produced Videos, are produced on a highly profes-
sional level. They mainly consist of a mix of several genres and sequences
and last between 1–2 hours. Extremist groups produce only a few of those
videos per year and time their release (anniversaries etc.).
Based on the content analysis of the Intel Center (2005, 2006), Salem, Reid and
Chen (2008) developed a two-dimensional approach to depict the dimension un-
derlying these classifications of extremist videos: One dimension covers the
speaker orientation and the other one the depiction of violence (operational versus
non-operational). The speaker orientation can be either individual- or group-or-
iented to emphasize whether the destiny of several individually distinguishable
mujahedin is displayed or whether the video focuses on a group as a whole (see
comparisons of all three category systems in Table 1).
22
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 23
Table 1. Categorization Approaches of Islamic Extremist Material
The specific implementation of these genres is strongly transported by subtle cues
as symbols and music. The Combating Terrorism Centre at West Point of the Uni-
ted States Military Academy has published a comprehensive analysis of the used
imagery in the jihadist internet propaganda. It aimed at making the subtle and
non-verbalized messages in propaganda footage visible. The catalogue reaches
from symbols of nature, geography and politics to people, weapons, warfare
and afterlife, to gestures and colors. Many motives are immanent in the Islamic
culture and combined with selective picture elements which are used by the jiha-
dists to suggest a certain extremist interpretation. The Imagery Project reveals the
attempt to elicit a required reaction through emotional or historical memories. For
instance, the depiction of a waterfall symbolizes spiritual progress, the spirit
which is ultimately raised to paradise and is therefore often used in conjunction
with martyrdom (Combating Terrorism Center at WestPoint, 2006).
Similarly, music plays an important role: Kimmage and Ridolfo (2007) mention
the importance of audio visuals consisting of war or attack pictures accompanied
by songs which “are the ideal conduit for an ideological message. The message is
strikingly uniform and primarily jihadist.” (p. 32).
A closer look into the specific narrations of Islamic extremist propaganda has
been provided by narrative analyses. Similar to conventional news reports, gener-
ally jihadist reports comprise a structured framing, consisting of a problem defi-
nition, causal interpretations, moral judgments and guidance (Rogan, 2007). Ac-
cording to Bandura’s concept of moral disengagement, the videos employ, (a)
“displacement of responsibility”, (b) “diffusion of responsibility”, (c) “dehuma-
nization of targets”, (d) “use of euphemistic language”, (e) “advantageous com-
23
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 24
parisons”, (f) “attribution of blame” and (g) the “distortion of the sequence of
events” (Weimann, 2006).
In a detailed analysis of the symbolic and semiotic level of communication in Ji-
hadist video clips, O’Shaughnessy and Baines (2009) established a three-way-in-
teraction model: The positioning triad. They differentiated between the Item (the
issue or text the communication is about), the Symbol (the intended message) and
the Meaning (the interpretation of the symbol within a cultural context). They
found that in all videos the West was depicted as the permanent aggressor
(Item), standing in contrast to the unfairly treated Muslims (Meaning) who
were represented through several cultural codes (Symbol). O’Shaughnessy and
Baines (2009) summarized that terrorist positioning of fear, hatred, death or jihad
is similar to a marketing strategy to the extent that terrorists “have target markets,
use communications, access channel of distribution and have a market differen-
tiation strategy” (p. 239).
In line with this, Payne (2009) also ascribes jihadist propaganda a special rhetoric,
using the idea of a global conspiracy against Islam and the obligation of a Muslim
to defend himself and his family as religious or social justification for the fight. In
particular, Al’Qaeda is famous for the narratives it invents. By distributing the vi-
deos they are successful in building a kind of brand that can appropriate for any act
of jihadist terror.
“Al’Qaeda’s ideological narrative has at its heart the desire for a return to Islamic
purity, and the division of the world into two distinct groupings, an ‘us’ and a
‘them’. On the ‘us’side of the equation are a vanguard of virtuous Muslims, con-
scious of the true Islam. The ‘them’ is an unholy alliance of Crusader-Zionists,
with a centuries-long record of scheming to do down Islam, allied with the corrupt
and degenerate political leadership of the Muslim world, who are thwarting the
desire of Muslims to return to their true faith.” (Payne, 2009, p. 111)
In typical Al’Qaeda propaganda messages, there appear to be four different kinds
of narratives: (1) An Islamist Utopia, (2) the notion “they are attacking us”, (3) the
message “Jihad is the only just response” and (4) the glory of martyrdom as a con-
sequence of the notion that terrorism is a legitimate tactic in jihad. By analyzing
these narrative structures, Payne (2009) clarifies that Al’Qaeda’s propaganda
message features death and tries to advance a wholly utopian ideal. “To do so,
it employs a simple, powerful narrative of revolution and resistance, good versus
evil, honor and retribution; all grounded in a particular interpretation of Islam.”
(Payne, 2009, p. 114)
As for the Islamic extremist video clips, some studies have been conducted in the
field of right-wing extremism. The production style of right-wing extremist pro-
paganda can be characterized as being provocative, containing a revolutionary
pathos and a rebellious appearance (Schedler, 2009).
24
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 25
Fewauthors name the most typical formats of right-wing extremist audio-visuals.
Busch (2008) described music videos with right-wing rock music to be the most
frequent format. Other videos contain scenes from demonstrations, presentations
of certain groups or talented speakers who present the right-wing extremist scene,
implying here also the existence of different extremist genres.
One exemplary narration analysis was conducted by Müller and Seiler (2010).
They analyzed both a right-wing extremist website (www.media-pro-patria.net)
and one propaganda video clip (“Mahngang des Gewissens”12). Regarding the
website, the authors noticed a simple-structured design and the usage of very
subtle means in order to appeal to the user. The intention of right-wing extremist
producers is to appear as serious and informational.
There are no significant features, symbols, paroles or pictures that clearly point
towards the right-wing extremist scene. The dominating aspects are calls to fight
against capitalism, globalization and Americanism. Social problems like child
poverty, child abuse and animal abuse are topics chosen to evoke sympathy due
to the right-wing extremist’s caring image (Müller & Seiler, 2010). Speaking
about the propaganda video, it directs criticism towards capitalism. The “System”
is found to be guilty for all social and political misfits. The authors state that the
video has a very modern, professionalized presentation format. There is no expli-
cit reference to a glorification of violence or the National Socialist ideology. Re-
lying on the general structure of propaganda described in Chapter 2 (Merten,
2000), the structure of Islamic and right-wing extremist propaganda can be sum-
marized as shown in Table 2.
12 “Mahngang des Gewissens” literally translates into “admonition of the conscience”.
25
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 26
Table 2. Structures of Islamic and Right-wing Extremist Propaganda
Structures Religious propaganda
Islamic Extremism
Political propaganda
Right-wing Extremism
(1) An arbitrary object (idea,
action, person or product) is
propagated and gains a un-
ique characteristic
God/Allah
The “Ummah”
Charismatic figure (“der
Führer”)
The “Völkische” Commu-
nity
(2) For those who consume
propaganda, a behavioral
premise is predetermined
that holds an exclusiveness
characteristic
Believe in Allah and/or
Follow his rules
Jihad is a duty
Obey the leader/the organi-
zation
Nationality defines who you
are
Defense of the Nation is a
duty
(3) The recipient of these
messages is forced to obey
this behavioral premise. This
leads to the divestment of
freedom to make own deci-
sions and creates exclusive-
ness
“Thou shalt have no other
gods before me” [Qur’an]
Encompasses the commit-
ment to the uniqueness of
god for which martyrdom is
the ultimate worship (Na-
tional Coordinator for
Counterterrorism, 2009b)
The leader/the organization
is always right, more vague
than in religious propaganda
but inherent in actions such
as stigmatization after dis-
obedience (heretic, deviant,
dissident)
(4) To have the recipients un-
conditionally obey these pre-
decisions, the propaganda
messages sketch positive and
especially negative future-
directed sanctions
Negative sanctions: the in-
vention of “the evil” and its
location: those who don’t
believe and disobey will go
to hell
Positive sanctions: those
who believe and obey will
go to heaven/paradise, will
have eternal life
Negative sanctions: actions
under massive audiences
(show trials, combustions,
manifestations, arrests)*
Positive sanctions: the idea
of an ultimate victory, revo-
lution of the world, the
“1000years Reich”
(5) Propaganda messages are
best formulated in a way that
makes them un-verifiable.
This is mostly realized by
emphasizing actions that may
happen in the future
Sanctions are non-verifi-
able since they are only ef-
fective after death
Reference to a better future
Note. * Political propaganda relies more on negative sanctions. Generally Islamic extre-
mism refers also to political topics. Nonetheless it is also a classical example for religious
propaganda (see Hoffman, 2006).
Tosum up, Islamic as well as right-wing extremist propaganda videos include dif-
ferent genres. As former approaches have not yet combined insights from the de-
scriptions of Islamic and right-wing extremist propaganda videos, we decided to
develop a categorization system encompassing both ideologies. Thus we focused
on prototypical genres asking whether right-wing and Islamic propaganda can be
described in comparable terms. The common root strategies depicted in Table 2
are comparable between the two ideologies (though distinctively realized).
26
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 27
The description of style and message in propaganda 2.0, however, is only the first
component in understanding its effects. As O’Shaughnessy and Baines (2009) sta-
ted, the symbolization of a visual image is based on the interaction between an in-
tended meaning and the interpretation by the audience. Without the recipient,
neither genre, nor symbol, nor message can unfold any persuasive power.
27
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 28
5. The Audience of Extremist Propaganda
“Internet propaganda is immensely dangerous, falling on a fertile ground it leads
to radicalization”. (German Federal Minister of the Interior Hans Peter Friedrich
RTL.de 27.03.2012)
Propaganda seeks a clear audience. Without a listener or recipient the message
would go unheard, unable to achieve its purpose. From the propagators’ perspec-
tive, four groups of recipients can be distinguished (Payne, 2009). Lying on a con-
tinuum, they range from those who would totally reject the presented messages to
those who would fully accept the portrayed ideology.
First, there are those already convinced of the transported ideology, the extremist
ingroup. The second group consists of those who are attacked by the threat, the
enemy or outgroup. Third, the group in between both ends of this continuum fea-
tures recipients who are not (yet) in support of the ideology or susceptible to the
conveyed messages. According to Payne (2009), they can be further differentiated
into (1) those who are to be convinced, and (2) the “global community”, those who
are able to classify the sender as “terrorist” or “freedom fighter”. Together they
form the potential ingroup or outgroup. They can sway to acceptance and become
radicalized ingroup members, as well as harshly distance themselves from the
ideology and turn to an outgroup. The resulting question “Under which condi-
tions do recipients of extremist propaganda offer ‘fertile ground’?” is not only im-
portant for possible prevention options but this also lies in the focal attention of
the presented study.
5.1 Knowing the Audience
Former research on the audience focused on the extreme points of the ingroup-
outgroup continuum, examining either one or the other. Research on the outgroup
(mainly in Western societies) showed us how terrorist messages can increase the
preference for one’s own culture (Liu & Smeesters, 2010) and the support of harsh
punishments against those who are rejecting their own norms (Fischer et al.,
2007). The research conducted after terrorist attacks and the subsequent media
coverage found that society reports higher levels of stress and anxiety (Marshall
& Galea, 2004; Propper, Stickgold, Keeley, & Christman, 2007). It was also found
that the overall acceptance of terrorism is low (Frindte & Haußecker, 2010), in-
dicating that the “global community” would rather reject extremist propaganda
when it has been identified as terroristic. However, this research does not yet an-
swer the question regarding the conditions under which this rejection could be
suspended.
Research on the already radicalized ingroup faces many difficulties. For example,
finding interview partners willing to talk is challenging and they may substan-
28
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 29
tially differ from those who do not. Researchers first have to gain access to prisons
(in order to interview imprisoned terrorists) or have to deal with injuries or even
the death of possible interview partners (particularly self-bombers in Islamic ex-
tremism, see Loza, 2007). This partly explains why the vast majority of studies
chose a retrospective approach and used secondary sources such as friends and
families (Bakker, 2006; Hegghammer, 2006; Sageman, 2004) or document ana-
lyses (e.g., Jenkins, 2010), mostly resulting in “anecdotal evidence”(Munton et
al., 2011, p.1). These retrospective analyses of extremist perpetrators can be sub-
stantially distorted. Besides the problems of sample selection, psychological
biases may veil the original motivations even for the perpetrators themselves. Hu-
man memory tends to be pretty flawed (Zhu et al., 2012) and a consistent positive
self-view is a strong motivator (Aronson, 1992), able to enormously influence
post-hoc interpretation of one’s own behavior. As Heitmeyer, Borstel, Grau, &
Marth (2010) have already formulated for the case of right-wing extremism, it
is impossible to define what came first: The acceptance of violence or the ideol-
ogy advocated (see also chapter 3.2).
The patterns described in all these studies may however serve as a starting point to
further investigate the audience. Although they lack in causal evidence, certain
factors mentioned in these analyses could have influenced an orientation towards
extremism (pre-radicalization phase, Precht, 2007) among the potential ingroup.
For the presented study, demographic characteristics were retrieved from relevant
research as they do not demand post-hoc interpretation (see Chapter 5.1). In a sec-
ond step, personality variables and attitudinal patterns were identified as possible
moderators of the found effects and their potential to influence the recipients’ re-
action was analyzed (see Chapter 5.2).
5.2 Demographic Factors
In a series of interviews with political inmates from right-wing and Islamic extre-
mist groups, Lützinger (2010) reported more commonalities than differences be-
tween extremist ideologies. Munton and his colleagues (2011) also mention that
those demographic factors which were found among right-wing as well as among
Islamic extremists seem to have the potential to influence the general reaction to-
wards extremist propaganda. Former studies analyzed the structure of Islamic ex-
tremist networks and found the members to be relatively young and mostly male.
Often, they were immigrants with a Muslim cultural background, and had to cope
with either broken educational careers or a frustrating professional situation (Bak-
ker, 2006; Lützinger, 2010; Sageman, 2004).
Although immigration was not found to be influential among right-wing extre-
mists, a general feeling of marginalization, which is present among many mi-
grants, was also found to be a factor in right-wing radicalization (Fuchs, 2003;
Heitmeyer, 2002; Terwey, 2000). According to Precht, “This [. . .] process [of rad-
29
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 30
icalization] is largely influenced by factors which are unique to every individual
but often starts by persons being frustrated with their lives [. ..]. They are search-
ing for an identity and a cause and often extremist Islam offers the solution to their
quest” (Precht, 2007, p. 35).
This “search for an identity” is strongly influenced by groups people feel af-
filiated to. Demographic factors, such as gender or culture, are easily accessible
categorizations which have the potential to create a feeling of being part of a cer-
tain group (Tajfel & Turner, 1986). Social psychological research has shown how
our identity, emotions, cognitions and behaviors are substantially influenced by
this need to belong.
Social Identity determines the direction, being, “[. . .] part of an individual’s self-
concept which derives from his knowledge of his membership of a social group (or
groups) together with the value andemotional significance attached to that mem-
bership” (Tajfel, 1982, p. 24).This is why extremists try to activate culture as pri-
mary group-identity (see Chapter 4.1) and foster outgroup comparisons. Based on
these observations, age and gender, cultural background, and educational level are
described in more detail below.
5.2.1 Age and Gender
In his classical work “Understanding Terror Networks”, Sageman (2004) ana-
lyzed 172 cases of Islamic extremists and reported the average age to join the jihad
to be 26 (see also, Bakker, 2006). For right-wing extremists, the first contact with
the scene happens between the age of 12 and 16 (Claus, Lehnert, & Müller, 2010).
Internet propaganda has a high chance to be accessed by adolescents. The so-
called digital natives are especially susceptible to online influences (Schmidt,
2009; Schmidt et al., 2010). According to the latest JIM Study (2011), practically
every teenager in Germany has access to a computer or notebook at home (Ebert,
Feierabend, Karg, & Rathgeb, 2011, p. 5). In 2011, 89% used the internet multi-
ple times a week or even daily and 68 % of them used video-platforms such as You-
Tube on a regular basis. Although there is no gender difference with regard togen-
eral internet usage, boys and girls still differed in what they did online and how
they evaluated it. For example 16 % of the boys and only 11 % of the girls rated
the internet as a credible source of information (Ebert et al., 2011).
Since extremism is still mainly a male phenomenon (e.g., Bakker, 2006; Claus et
al., 2010), extremists can be assumed to address mainly young males via online
propaganda. For the presented study we thus decided to focus only on young
male adults.
30
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 31
5.2.2 Cultural Background
“That was how it started with the right-wing extremist attitudes, because I’ve
grown up in an estate where foreigners were foreigners and Germans were Ger-
mans” (Right-wing extremist “Ri21”, cf. Lützinger, 2010, p. 171)
Whenever dealing with any kind of extremist ideology, it is obvious that the po-
tential ingroup addressed is often a cultural one.
While right-wing extremists in Germany refer to the German “blood and history”,
Islamic extremists point towards their specific interpretation of being “Muslim”.
Culture, in both cases, is formed by national as well as religious influences.
While Germany has developed with an atheistic/Christian (occidental) back-
ground, Muslim tradition and identity has been formed by Islam and by oriental
influences (e.g., Lewis, 2003). In both ideologies, the striving for “purity” (Ger-
man nation respectively Muslim caliphate) free of foreign influences lies in focal
attention (see Chapter 3.1 and 3.2). Entering a right-wing extremist group in Ger-
many being a migrant is practically impossible.
Having parents from a Muslim country has been reported to be a characteristic
among the members of Islamic extremist networks. For example, Bakker
(2006) found that especially second generation migration was a common factor
in nearly 300 cases of European “home-grown” jihadists. The same was reported
for jihadists socialized in the USA or for converted Islamic extremist (Jenkins,
2010).
Although Islamic extremist propaganda can be classified as religious (Merten,
2000, see Chapter 2), Sageman (2004) pronounced that the causality between ra-
dicalization and religion is unclear. Only a small percentage of jihadists in his ana-
lysis had a primary and secondary Muslim education. Though religious identifi-
cation increases during the radicalization process, it does not necessarily trigger it
(Munton et al., 2011; Precht, 2007).
In line with this, the Muslim educational scientist Mouhanad Khorchide said in an
interview with the German “Zeit”(Khorchide & Schenk, 2012) that the message
transported by salafists (see Footnote 6, Chapter 3.1) in Germany is not really
about religion but about being rejected by the main society for being Muslim.
Perceived discrimination and a feeling of marginalization have constantly been
discussed as the underlying factors in the Islamic radicalization processes (Baines
et al., 2006; Wilner & Dubouloz, 2009). Similarly, feelings of marginalization and
of non-acceptance by society have been found to be risk factors for right-wing ex-
tremism (Rippl & Seipl, 2002). Internet communities can contribute to identity
development particularly among marginalized groups (McKenna & Bargh,
1998) and extremists try to satisfy this need (Precht, 2007). “In the organization,
colors and races dissolve. There are no foreigners among us. We are all brothers,
believers and have a close relationship. [. . .].” (Al-Ablaj cf. Aaron, 2008, p. 263).
31
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 32
However, the “cultural” group-identity extremists address has to be isolated from
the general national identity. The extent to which someone identifies with the na-
tional or majority culture (national identity self-concept) should thus be re-
garded separately (Maehler & Schmidt-Denter, in press). The majority of Mus-
lims and Germans harshly reject extremism (Frindte, Boehnke, Kreikenbronn,
& Wagner, 2012; Jenkins, 2010). The actual number of Muslims in Germany
(3.8 to 4.3 million, see Frindte et al., 2012, p.16) is nearly 40.000 times higher
than the number of actual Islamic extremists (about 130, Friedrich, RTL.de,
march 27th, 2012). In line with this, radical right-wing parties in Germany are
voted for only by a minority (see Bundeszentrale für Politische Bildung13) (see
Jenkins, 2010 for US numbers). Additionally, research has shown that negative
ingroups motivate people to distance themselves from them (Arndt, Greenberg,
Schimel, Pyszczynski, & Solomon, 2002). In general, there are specific factors
which make people suspend their primary rejection of extremist attitudes or be-
haviors. Bandura (cf. Weimann, 2006) claims “moral disengagement” to be ne-
cessary to overcome this rejection. It seems that addressing the cultural back-
ground as a characteristic of the potential ingroup may be ambiguous.
A shared culture may lead to a feeling of belongingness to the “Ummah” or the
“German nation” as well as to sharply distance from the ideology promoted.
For the presented study, different ingroup and outgroup conditions were realized
by inviting Germans, Muslims and other migrants and confronting them with
right-wing as well as Islamic extremist propaganda.
5.2.3 Educational Level and Major Field of Study
Feelings of social marginalization have already been discussed as risk factors for
radicalization (e.g. Decker,Weißmann, Kiess, & Brähler, 2010; Heitmeyer, 1992;
Precht, 2007). Low educational status can also foster feelings of being separated
(Rippl & Seipl, 2002). Not surprisingly, low educational status or a gap between
education and profession has frequently been found among Islamic as well as
right-wing extremists.
Lützinger (2010) reported broken educational careers as communality among the
political inmates in her study. Similarly, Bakker (2006; see also Sageman, 2004)
report a mismatch between educational capabilities and work situation. Bakker
(2006) attributed this to the difficulties immigrants have to face in working situa-
tions (e.g.,Watson, Appiah, & Thornton, 2011). This explains why only few Isla-
mic extremists, especially home-grown terrorists in Western countries, had com-
pleted an academic career. Professional frustration has also been discussed to be
an underlying factor (Gambetta & Hertog, 2007). For biographies of extremists
who are located on the Arabian Peninsula, Gambetta and Hertog (2007) reported
13 http://www.bpb.de/politik/extremismus/rechtsextremismus/41904/rechtsextreme-parteien. Re-
trieved online, 22/06/12.
32
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 33
a strong overrepresentation of engineers which are frequently confronted with un-
employment there, among the already radicalized ones. Educational level may
thus play a role in the reaction towards extremist propaganda. Therefore, the sam-
ple for the presented study included different educational levels; namely students
and pupils from vocational schools and with regard to the findings of Gambetta
and Hertog (2007) with different major fields of study. Similar to culture (see
Chapter 5.2.2) as a potential influence, claiming a causal relation between educa-
tion and the effects of propaganda would do injustice to the understanding of the
process of radicalization. Unsurprisingly in that respect, research has tried to
identify personality factors and attitudinal patterns that could moderate the rela-
tionship between certain superficial classifications as preconditions and radical
attitudes.
5.3 Personality Factors and Attitudinal Patterns
Personality factors influencing extremist biographies in context of Islamic extre-
mism have also been researched retrospectively, trying to interpret or even psy-
choanalyze the entire personality of already radicalized individuals.
“One finding is that, unfortunately for profiling purposes, there does not appear
to be a single terrorist personality. This seems to be the consensus among terror-
ism psychologists as well as political scientists and sociologists. The personalities
of terrorists may be as diverse as the personalities of people in any lawful profes-
sion.” (Hudson, Majeska, & Metz, 1999, p. 60).
Hudson’s (1999) findings explain why broad personality factors such as the Big
Five (Costa & McCrae, 1992) have failed to reliably predict prejudice or discri-
mination (Ekehammar, Akrami, Gylje, & Zakrisson, 2004). Accordingly,psycho-
pathological approaches generally could not find a significantly higher preva-
lence of personality disorders among terrorists compared to others (Munton et
al., 2011; Sageman, 2004). “This is not to deny, however, that certain psycholo-
gical types of people may be attracted to terrorism” (Hudson et al., 1999,
p. 60). For example, search for excitement and hunger for stimulation have fre-
quently been discussed (e.g., Hudson et al., 1999) to increase the attraction of ter-
rorism. Lützinger (2010) reported thrill and adventure as well as experience seek-
ing motives to influence right-wing and Islamic extremists. Baines and colleagues
(2010) supposed that an adventure motivation could contribute to a positive eva-
luation of Islamic extremist propaganda. Similarly, modern right-wing extremists
(see Chapter 3.2) offer an action-oriented group image (Schedler, 2011). These
features indicate characteristics of the sensation seeking personality. Sensation
seeking (Zuckermann, 1979) describes the search for stimulating experiences
motivated by a low level of general arousal. This fits Lützinger’s finding (2010)
that boredom was negatively perceived by her interview partners. Regarding radi-
calization processes, sensation seeking has not yet been analyzed experimentally.
33
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 34
The effect it might have on the evaluation of extremist messages has yet to be
tested.
In addition to personality factors, the definition of extremism as “a combination
of radical attitudes and the choice of extreme means” (see Chapter 3) implies the
importance of attitudinal patterns. Lützinger (2010) reported a strong political ar-
gumentation among right-wing as well as Islamic extremists. Munton et al. (2011)
also found political attitude to be a strong motivator. In particular, right-wing ex-
tremist propaganda can be classified as political.
As inherent in the relation of cognitive abilities and the preference for a simple
worldview, Lützinger (2010) found an increased search for order and structure
and a simplified worldview mentality among all types of extremists.
“System extends into all aspects of life; it discusses all minor or major affairs of
mankind; it orders man’s life[ . . .] People should devote their entire lives in sub-
mission to God, should not decide any affair on their own [. . .]” (Sayed Qutb,
1964, p. 34–36).
Authoritarianism is strongly related to this search for order and also to moderate
the relation between cognitive abilities and racism (Hodson & Busseri, 2012).
Authoritarianism, as an attitudinal pattern, is the desire to submit under authori-
ties and the demand for punishment of those who do not. It has been found to be a
predictor for outgroup derogation, defense of the own cultural worldview (Fuchs,
2003; Greenberg et al., 1990), as a mediator of the relationship of low feelings of
control and prejudice (Agroskin & Jonas, 2010) and to correlate with right-wing
political attitudes (Larsen & Schwendiman, 1969). Due to the inherent orientation
towards the ingroup and the need for social order, cohesion and stability (Duckitt,
2011; Duckitt & Sibley, 2010), authoritarianism might be a predictor for radical
right-wing and Islamic attitudes (Altemeyer, 1988; Frindte et al., 2012; Fuchs,
2003). It also correlates with low educational level and fundamentalist religious
beliefs (e.g., Altemeyer, 1988). Furthermore, it was found to moderate the nega-
tive relationship between religiosity and the endorsement of democratic values
(Canetti-Nisim, 2004). High levels of authoritarianism could also positively influ-
ence the evaluation of extremist messages.
According to Precht (2007), the second phase in a radicalization process is char-
acterized by the acceptance of extremist means. Accordingly, a general accep-
tance of violence is the second component of right-wing extremism (Heitmeyer,
2002). This may also be relevant in the context of Islamic extremism, as can be
derived from the harsh rhetoric used in many propaganda clips: “Strike the
long-awaited blow and kill as many infidels as you wish. Turn my lands into a
grave” (cf. Kimmage & Ridolfo, 2007). As described by Heitmeyer et al.
(2010), the causality between ideological justification and acceptance of violence
is unclear, suggesting that specific justif ications of terrorism may also influence
the evaluations of extremist propaganda. It has to be noted, however, that a general
34
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 35
acceptance of both terrorism and violence has usually been found to be low (see
Frindte et al., 2012; Heitmeyer, 2002).
35
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 36
6. The Effects of Extremist Propaganda: Rejected Acceptance?
“They would invite pupils to evening lectures and social occasions and give them
pamphlets to read. Then they would show jihad videos from Bosnia, Chechnya and
elsewhere, and thus motivate people to travel to Afghanistan.” (Former militant
describes his initiation process cf. Hegghammer, 2006, p. 50)
The senders of propaganda clips aim to: “shape perceptions, manipulate cogni-
tions and direct behavior” (Jowett & O’Donnell, 2012, p.7). The potential in-
group is to be persuaded of the ideology transported, the “corporate identity” cre-
ated and the means derived. As of yet, research examining if propagandistic video
clips are able to do so, is scarce.
Former research on the effects of extremist propaganda has mainly shown that un-
obtrusive participants did not react with “desire” but with rejection. The degree of
rejection varied depending on style and content of the message as well as depend-
ing on sender and recipient. For example, messages not clearly identifiable as pro-
paganda (Allen, 1991) or with credible speakers (English et al., 2011) were more
persuasive than others. However, research on the effects of extremist propaganda
clips has been practically nonexistent until now. The only exception is the work by
Baines and his colleagues (Baines et al., 2006, 2010). They showed first hints on
differences between Islamic extremist clips depending on their content and style.
Focusing on the potential ingroup, they analyzed a non-radicalized sample of
British Muslims.
Each of their participants watched five Islamic extremist clips which differed in
terms of the style and topics they dealt with. Two of them were movie-like formats:
Amusical clip about a suicide attack and a comic showing a Palestine boy fighting
an Israeli soldier. Furthermore, a realistic clip – a CNN-style news broadcast –
about how Al’Qaeda sees the world, and two “talking heads” clips were included.
One showed Al-Zawahiri, an Ideologue and the other one featured Muhammed
Sidique, a Martyr. Both explained how they justified Jihad by addressing Western
failures and Muslim victims.
Following reception, participants were asked to discuss their impressions in small
focus groups. One of the two male focus groups clearly rejected the clips and
thought them to be “irresponsible and childish” (Baines et al., p. 6). They eval-
uated them as “unlikely to stimulate respondents to act in the name of Allah”
(p. 6). The other male group, however, showed a less pronounced rejection.
Though not stating own susceptibility, the participants claimed that others might
copy the acts, reacting to the message that “the Western world is corrupt” (p. 10).
In both cases, the speech of Al-Zawahiri about the number of young Muslim chil-
dren dying due to the West raised understanding. “Up to [a] point, most of what he
[Al-Zawahiri] said made sense” (p. 10). Furthermore, the authors suggested that
“stimulating” clips (such as the movie-style ones) could raise greater interest in
36
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 37
the Jihad. Having watched the videos, however, none of the groups expressed an
overt sympathy for the radical means proposed.
The study implies that comprehensibility of the arguments plays a role. Compre-
hensibility might be influenced by pre-existing attitudes as well as the context in
which the arguments are embedded. “Communication effects are the greatest
when the message is in line with relevance, existing opinions, beliefs and disposi-
tion of the receiver” (Jowett & O’Donnell, 2012, p. 209).
Ramírez and Verkuyten (2011) analyzed the effect of pre-existing values and mes-
sage context. Their participants read a newspaper article about a planned extre-
mist speech in their hometown. They evaluated whether the speech should be tol-
erated and how participants would feel about the extremists. The speech was
either classified as right-wing or as Islamic extremist. The context was manipu-
lated by presenting the article either as discussing public order and stability or
as dealing with freedom of speech. Participants were also asked about their pre-
ference when forced to choose between these two values. Similar to Baines et
al. (2010), the vast majority of participants had negative feelings towards the ex-
tremist groups. Tolerance for the speech nevertheless varied with individual va-
lues and framed context. Participants preferring order and stability showed less
tolerance, except when freedom of speech was framed.
For those already valuing freedom of speech, framing public order did not lead to
a mirrored pattern. In contrast, they showed a boomerang effect (Feather, 1990),
being even more tolerant. Thus, personal values and the “label” of presentation
interacted to produce the effects. The ideology also influenced the results. Fram-
ing freedom of speech increased tolerance for right-wing but not for Islamic ex-
tremists.
Roessing and Siebert (2006) also found differences depending on the portrayed
ideology. They presented four extremist documents (e. g. screenshots of an extre-
mist forum or instructions on how to craft Molotov cocktails) to German students.
The material was either framed as right-wing or as left-wing extremist. Results
showed that material presented as right-wing was perceived as more threatening
than the left-wing material.
All of these studies imply that the extremist aims to “shape perceptions, manip-
ulate cognitions and direct behavior” (Jowett & O’Donnell, 2012, p.7) are not
automatically fulfilled by propaganda. However, they cannot answer the question
under which conditions non-obtrusive recipients could be susceptible. Different
video styles (Baines et al., 2010) and different ideologies (Ramírez & Verkuyten,
2011; Roessing & Siebert, 2006) have not yet been combined in experimental ap-
proaches.
The presented study aims at closing this gap. As a first step, we explored the re-
actions of unobtrusive recipients, confronting them with varying propaganda
from their cultural ingroup or outgroup. We aimed at investigating immediate ef-
37
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 38
fects, as they influence the probability offurther processing of accordant material.
As McGuire (1968) and Zimbardo and Leippe (1991) formulated it: The recipient
has to be exposed to a certain message; the message has to create attention and
evoke interest. [.. .] Then [it] can create retention, which might lead to a transla-
tion of attitude to behavior and finally to action.
However, the effects are also substantially influenced by the recipient’s motiva-
tion. Several two-process models of persuasion distinguish between an accuracy
motivation (leading to a detailed consideration if the individual feels involved),
defense motivation (evokes when attitude change would threat the self-concept)
and an impression motivation (drives the effects if the attitude matches social
norms (Chen, Duckworth, & Chaiken, 1999).
Defense motivation allows only for the information supporting the pre-existing
attitudes to be processed. Impression motivation, in contrast, leads to a reporting
of only socially desirable attitudes.
Summarizing the immediate effects of reception can be described as spontaneous
evaluations or attitudes towards the clips. As implied in the definition of propa-
ganda (Jowett & O’Donnell, 2012, p. 7), attitudes contain cognitive, emotional
and behavioral components. Predominantly cognitive attitudes refer to known ob-
jects; their positive and negative features end the resulting expectancies. Emo-
tional attitudes, in contrast, refer to personal values and “gut feelings”. They ex-
press the own (cultural) worldview (Katz, 1960). Behavioral attitudes emerge
after the execution of a certain behavior and justify it (Bem, 1972). Thus they ren-
der an implausible explanation when someone is confronted with a video for the
first time.
The presented study focused on immediate emotional and cognitive responses. As
shown in former studies, we expected overall negative emotions and cognitions.
According to them, we also expected different degrees of rejection. The style and
content of the clips, the ideology behind them and factors on the side of the reci-
pients (e.g. attitudes or shared cultural background) were expected to shape the
pattern.
38
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 39
7. Research Questions
In the last chapters, research on propaganda itself, the senders and possible char-
acteristics of the recipients, as well as potential effects of such material were pre-
sented. Relying on Lasswell’s (1948) formula “Who (sender) says what to whom
(recipient) in which channel and with what effect” (see Chapter 1), the first part of
this book aimed at describing the sender, channel and recipient of extremist pro-
paganda 2.0.
Although propaganda can appear in many possible ways, predominantly the inter-
net was presented as increasingly important for modern propaganda (see Chapter
2). In particular, online distributed audio-visuals evoke special interest as they
combine the entertaining potential of movies (O’Shaughnessy, 2012) and the
wide dissemination and take-part opportunities of the social web (Schmidt et
al., 2010). It was depicted that mainly right-wing and Islamic extremists use
the internet to an alerting extent (see Chapter 3). Over the last years these two ex-
tremist groups expanded their online presence in order to propagate their ideol-
ogy, legitimate their goals or actions as well as intimidate their “enemies” (Cor-
man et al., 2006; Payne, 2009). For this purpose Islamic and right-wing extremists
produce a large variety of video formats. The current research therefore focused
on right-wing and Islamic extremist propaganda videos. Comparable formats
within both ideologies were expected. From this literature review the first of
our research foci emerged:
R1: Description of right-wing and Islamic extremist propaganda videos and ana-
lysis of their effects.
Q1: Which prototypical formats of right-wing and Islamic extremist propa-
ganda in Germany can be distinguished? Which messages are transported?
Q2: Do distinct production formats and message styles differ with regard to
their potential to create interest or to trigger rejection?
Among the goals of propagation, legitimation and intimidation, different addres-
sees of extremist propaganda were distinguished (see Chapter 5). The potential
ingroup in terms of demographic, personality and attitudinal patterns was identi-
fied as the group of interest for the current study. Former research had mainly fo-
cused on the already radicalized ingroup respectively the rather uninvolved out-
group. However, the factors which could contribute to a f irstapproach towards the
ideology when stumbling over such video material have not been tested experi-
mentally before.
Relying on factors discussed among actual ingroup members, age,gender,cul-
tural background,educational level and major field of study emerged as pos-
sible explanative demographic factors. National identity has also been discussed
as a possible influence. Furthermore, broader personality characteristics, such as
39
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 40
the big five and sensation seeking, were assumed to contribute to a first under-
standing of why propaganda might trigger a slight approach. Above all, however,
attitudinal patterns, such as authoritarianism,feeling of control,political atti-
tude and acceptance of violence, were identified as potential influences on the
effects of propaganda.
Our second research focus was derived from the analysis of these potential factors
influencing the effects of propaganda:
R2: Analysis of the immediate explicit effects of right-wing vs. Islamic extremist
propaganda on different audiences.
Q3: Do recipients react differently to right-wing than to Islamic extremist
propaganda?
Q4: Does this reaction depend on the cultural background or other demo-
graphics of the participant?
Q5: Are personality factors and pre-existing attitudinal patterns able to in-
fluence the evaluation of extremist propaganda?
Former research on the effects of audio-visual and extremist propaganda on nor-
mal young adults predominantly reported adverse reactions (see Chapter 6). How-
ever, the effects depended on characteristics of the material, the ideology, and the
participant.
Most research so far focused on explicit14 reactions towards propaganda. Since
the topic itself can be regarded as adverse and thus might evoke socially desirable
answers (King & Bruner, 2000), it is important to take implicit reactions towards
the topic into account. Our third research focus thus additionally addressed the
question of how implicit associations were affected by the exposure to propa-
ganda material.
R3: Analysis of the implicit effects of selected propaganda videos.
Q6: Do different propaganda videos have the potential to shape the implicit
associations of extremism in its most extreme form?
Figure 1 displays the factors which were expected to influence the immediate eva-
luation of extremist propaganda videos.
14 Explicit= verbalised, conscious. Implicit= non-verbalised, unconscious (Schnabel et al., 2006)
40
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 41
Figure 1. Assumed factors influencing the immediate evaluations of extremist propaganda.
41
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 42
Part II Studying the Effects of Right-Wing and Islamic Extremist
Propaganda Videos
8. The Current Study: Exploring the Effects of Propaganda 2.0
To address the research questions (see Chapter 7), a large-scale study with four
consecutive parts was conducted at the University of Cologne in 2011. This study
and its results are described in this second part of the book. We aimed to balance
readability and precise presentation and therefore we explain the statistical ana-
lyses briefly in footnotes for the non-trained reader. Meanwhile, the appendix pre-
sents supplementary statistical analyses for trained readers. For further questions
concerning the statistical details, please contact the authors.
Based on the theoretical background displayed in the first part of this book, we
made the following decisions for our study:
Comparison of right-wing and Islamic extremist video clips in order to under-
stand current propaganda 2.0 in Germany
Description of similarities and differences between videos from both ideolo-
gies concerning production style and plot
zDevelopment of an integrative categorization system allowing systematic
comparisons of single clips within an ideology as well as between ideologies
zIdentification of prototypical example videos (stimuli) in order to expose
participants to them during the experiments
Invitation of male young adults differing in educational level and cultural back-
ground
Assessment of further demographics, personality factors and attitudinal pat-
terns, potentially influencing the evaluation of extremist video clips
Realization of a multi-methodological approach in order to display explicit and
implicit effects of propaganda.
Table 3 gives an overview of the deducted research foci and their realization
throughout the project.
42
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 43
Table 3. The Research Foci and their Realization during the Project
Research Focus Step Aim Methods
I. Description of right-
wing extremist and Is-
lamic extremist propa-
ganda videos and ana-
lysis of their effects
(see Chapter 9)
Research Questions 1
and 2
1. Content
Analysis
Development of a categorization
system describing the prototypical
formats of extremist video clips
Selection of prototypical video
clips for these categories as stimu-
lus material
Ethnographic
Content Ana-
lysis (ECA)
2. Study I Development of factors explaining
the explicit post-hoc evaluation of
propaganda
Description of online and post-hoc
evaluations of the videos
Further selection of affecting video
clips as a more precise stimulus
material for further processing
Integrated-
Reception
Process Ana-
lysis (IRP)
Self-reported
post-hoc eva-
luations
II. Analysis of the im-
mediate explicit effects
of right-wing vs. Isla-
mic extremist propa-
ganda on different
audiences (see Chapter
10)
Research Questions
3–5
2. Study I
and
3. Study II
Testing the explicit effects of var-
ious propaganda videos within one
student and one non-student sam-
ple (pupils from vocational
schools)
Identification of affecting factors
on the side of the recipient influ-
encing these effects
III. Analysis of the im-
plicit effects of selected
propaganda videos (see
Chapter 11)
Research Question 6
4. Study III Testing the implicit effects of se-
lected video clips on the positive vs.
negative association of the concept
“terrorism”
Single Cate-
gory Implicit
Association
Test (SC-IAT)
As depicted in Table 3, we used a multi-methodological approach to investigate
the various aspects of propaganda videos. A total of four studies were conducted
in order to answer our research questions (see Chapter 7).
At first, a content analysis was conducted so as to cautiously describe the available
video material. After we had identified distinguishable prototypes (see Chapter
9.1), we chose representativevideos out of these for further processing (see Chap-
ter 9.2).
As presented in Table 3, the first two studies focused on explicit measures and
Study III addressed the implicit ones. Study I and II followed the same procedure,
which is described in Chapter 9.3. In order to measure different aspects of the eva-
luation of the presented videos, several complementing methods were used. For
the implicit effects, real-time response measurements were assessed during recep-
43
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 44
tion (psycho-physiological arousal and real-time evaluations). On the other hand,
questions following each clip addressed participants’ immediate emotional and
cognitive reactions (post-hoc evaluations) (see Chapter 9.4.1.).
For these complementing effect measures, the data of study I wasfirst analyzed in
an explorative manner. We examined which videos evoked which reactions in the
participants and identified underlying factors in the evaluation. The results
showed differences in the affecting potential of different video clips (see Chapter
9.4.2). Based on these results, the sample of videos was reduced. Only videos
which clearly evoked a reaction in the participants were included in Study II
(see Chapter 9.4.2).
The stimuli selected for Study II were then presented to a broader range of reci-
pients. Besides students, a sample of pupils from vocational schools was included.
The statistical analyses15 confirmed invariance of the evaluation between the two
studies, supporting the selection and allowing for direct comparison Chapter 10
(see Table 3) presents the two studies together in order to raise generalizability.
(For a more detailed analysis of the two separate studies and the role of culture
and education, please see Frischlich & Rieger, (2013)).
Particularly, the role of recipient variables as potential influences lay in focal at-
tention. Against the theoretical background described in Chapter 5, demographic
variables as well as the additional explanative benefits of personality and attitu-
dinal factors were examined.
In order to allow detailed insights into processing and evaluation of propaganda,
Study III again reduced the stimulus material and amplified the methods used.
Four representative videos were selected. They served as stimuli in order to ana-
lyze the potential of video clips meant to shape the implicit valence association of
terrorism among German students. In order to increase readability and to distin-
guish between explicit and implicit approaches, Study III is described separately
in Chapter 11.
15 Conf irmatory multilevel-factoranalysis proved invariance of the factors assessed between the two
samples. Model fit was good on all relevant criteria.
44
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 45
9. Research Focus I: Description and Analysis of Right-Wing
Extremist and Islamic Extremist Propaganda Videos
Q1) What are the prototypical formats of right-wing and Islamic extre-
mist propaganda in Germany? Which formats and styles can be distin-
guished? Which messages are transported?
In order to answer the first research questions, we used a content analysis16. As
described in Chapter 4.2, although research has focused more on Islamic extre-
mist propaganda (Intel Center, 2005; Salem et al., 2008), it seems plausible
that comparable formats can also be identified in right-wing extremist material.
Our categorization system focused on typical formats of the two ideologies.
We conducted an Ethnographic Content Analysis (ECA, Altheide & Cheney,
2006) in which former categorizations of Islamic extremist propaganda served
as a starting point (see Chapter 4.2 and 6). Categories were refined in close inter-
action with the videos from the two ideologies. ECA regards the coder as well as
other recipients as part of the meaning of the message. As O’Shaughnessy and
Baines (2009) already noted, the specific meaning of the symbols used in Islamic
extremist propaganda has to be regarded within Muslim culture. Thus German as
well as Muslim coders were included in the coding team.
For the Islamic extremist material, a large database of 1.039 video clips was pro-
vided by the Joint Internet Surveillance Centre in Berlin (GIZ). As the study fo-
cused on the impact on young people living in Germany, only those 53 Islamic
extremist clip either without text, in German, or with German subtitles were ana-
lyzed in detail. As far as the right-wing extremist material is concerned, jugen-
dschutz.net provided us with a database of 60 video clips.
The resulting categorization system covered four main categories. They were si-
milar to those described by Baines and collegues. (2010, see Chapter 4.2): Videos
featuring “Talking heads”, journalistic “Reality clips”, produced “Movie clips”
and – in addition to their selection – “Extreme clips” were distinguished. As de-
picted in Figure 2, the main categories Talking heads and Reality clips spanned
three sub-categories each.
16 According to Berelson (cf. Bente & Krämer, 2004, p. 203), “content analysis is a research tech-
nique for the objective, systematic and quantitative description of the manifest content of com-
munication” (p. 18). Commonly in content analyses, a categorization system is developed, offer-
ing the possibility to classify the material at hand. It is inherent that categories should exclude each
other and cover the variability of the material. An additional “rest” category should be included.
The categorization system can be derived either theoretically or in interaction with the material, or
both at the same time. Different trained researchers should then categorize the same material
(Mayring & Gläser-Zikuda, 2008)
45
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 46
Figure 2. Formats of extremist propaganda videos either in German, with German subtitles, or without
text.
46
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 47
9.1 The Categorization System and Frequency of the Formats
Talking head (TH) formats were defined by one or more speakers directly addres-
sing the recipient, resembling Ideology/Statement clips. Direct addressing is
either accomplished by direct speech or by a camera angle simulating face-to-
face interaction. In these (stylistically simple) video clips, the camera focuses
on the speaker to bring the individual message into focal attention. Sometimes,
speeches are underlined by pictures in the background, but for the most part the
speaker fills the majority of the screen. Personal motives for entering Jihad (or
joining the National Socialists) are described from a first-person perspective. Re-
ferring to Hegghammer (2006), Ideologues (like Osama bin-Laden), Lifestyle
activists (like Abu-Thala17) and Martyrs (like Mohammed Atta Khan or Mo-
hammed Sidique18) were distinguished as sub-categories. The main communica-
tion goal of these clips is the legitimization of own deeds and recruiting new mem-
bers (see Chapter 4.2)
Reality clips (RC) resemble Propaganda/Promotion and Humiliation video clips
described in former categorizations (Finsnes, 2010; Intel Center, 2005, 2006).
They try to give the impression of a journalistic third-person perspective. Often,
usual news broadcasts are copied (Baines et al., 2006). Pictures taken from the
mass media are reused and partially accompanied by rhythmic music (Anasheeds
or right-wing songs). While for Talking head formats the speaker always lies in
focal attention, Reality clips focus on various people or groups. Thus, subcate-
gories, namely Group lifestyle,Public opinion and News broadcasts, were distin-
guished.
Group lifestyle videos display a cohesive, action-oriented and strong community.
They focus on the extremist ingroup and their (partially very violent) actions.
Usually handheld cameras are used to film the extremist’s daily life. Filmed inter-
views, in contrast, try to transport Public opinion by interviewing “normal” peo-
ple in the street. These videos aim for broad support of the transported extremist
worldview.
News broadcasts and the overall state of the world are the focus of the last sub-
genre. Although these News broadcasts are reported from the extremists’ perspec-
tive, similar to Public opinion, they simulate objectivity. Like Group lifestyle, the
other two formats are often filmed with a handheld camera. This lets them appear
to be “real” instead of “artificial”, action-oriented and agitated. As described in
17 Muslim from Bonn, Germany. In 2009, he was one of the most famous “frontline” Jihadists. He
alerted German security agencies with threats of attacks in a self-produced video clip but was
killed in the Afghanistan- Pakistani border region in January 2011.
Retrieved online http://www.spiegel.de/politik/deutschland/dschihadist-abu-talha-bonner-qaida-
kaempfer-soll-bei-gefecht-gestorben-sein-a-740140.html, 21/07/2012.
18 The perpetrators of 9/11 and of the bombings in London in 2005.
47
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 48
Chapter 4.2, the video clips address the potential ingroup and the global commu-
nity in order to legitimate and propagate their goals.
The Movie clips (Baines et al., 2006) or Produced videos (Finsnes, 2010; Intel
Center, 2005, 2006), in contrast, are produced similarly to feature films. Their de-
sign is highly professional, for instance, by using film music (sometimes even
from a well-known Hollywood movie), metaphoric pictures and showing incar-
nate destinies (children dying and mothers weeping). For right-wing as well as Is-
lamic extremist propaganda, music video clips also play an important role for pro-
pagation (see also Baines et al., 2010). In contrast to Talking head and Reality clip
formats, the content is transported both textually and audio-visually. The focus
lies on victims of the own ingroup while accusing outgroups of being guilty
(for their children dying). Videos of Islamic extremist material can last up to hours
but different scenes and foci can be distinguished. The clips primarily try to pro-
pagate but also legitimate own violence.
The last category covers Extreme clips. They can contain different elements of the
other three main categories. However, what singles them out is their norm-violat-
ing character. Besides dealing with the legitimization of martyr acts, they also
handle taboos such as pedophilia, hostage takings, and beheadings. Some even
give names and locations of outgroup members, demanding their death. In con-
trast to the other categories they contain a strong intimidation aspect.
The four identified categories allowed the description of Islamic as well as right-
wing extremist propaganda. However, some subcategories were only found for
either Islamic or right-wing extremist videos. A quantitative analysis of the videos
within each category is presented in Table 4. Videos which were unique or con-
sisted of many different genres without being thematically focused (such as Ex-
treme clips) were coded as Rest.
Table 4. Frequency of Videos within each Extremist Format Category
Category Right-wing
extremist videos
Islamic extremist
videos Total
TH Ideologue – 16 16
TH Lifestyle activist 9 5 14
TH Martyr – 4 4
RC Group lifestyle 21 13 34
RC Public opinion 3 – 3
RC News broadcast 2 2 4
Movie clip 19 1 20
Extreme clip 6 3 9
Rest 10 8 18
Total 60 53 112
Note. TH = Talking head, RC = Reality clip. The table shows the absolute number of videos
coded in one category.
48
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 49
As can be derived from Table 4, overall Reality clips were the most frequent for-
mat. In both ideologies, about a third of all clips fit this category (29 % of the Isla-
mic and 37 % of the right-wing extremist video clips). RC Public opinion however
appeared to be unique to right-wing extremist videos. In contrast, Group lifestyle
was frequently found among both Islamic and right-wing extremist propaganda.
Neither TH Ideologue nor TH Martyrs, were found for right-wing extremist pro-
paganda. Movie clips, in contrast, were more typical for right-wing (27 %) than for
Islamic extremist (2 %) material. Extreme clips were found to be rarest among
both ideologies (6 % of the Islamic and 9 % of the right-wing extremist videos).
Figure 3 shows the distribution.
49
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 50
Figure 3. The distribution of single formats within the two ideologies analyzed. TH = Talking head,
RC = Reality clip.
50
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 51
9.2 The Selected Videos for the Studies
Based on the developed categorization system, prototypical videos for each cate-
gory were selected to serve as stimulus material for the subsequent experiments.
Since not all categories were found for both ideologies, a total of 13 clips resulted
for Study I (one video more for Islamic extremist propaganda). Generally, it has to
be noted that Islamic and right-wing extremist videos differed regarding length:
Whereas right-wing extremist material took only a few minutes, the Islamic ex-
tremist material took up to three hours. Thus, representative scenes were cut
with a maximum duration of 06:04 minutes, in order to increase comparability.
TH Ideologue was found only for Islamic extremism. The selected scene origi-
nates from a speech of Osama bin Laden to the European Union about the Mo-
hammed caricatures19. In the video, a picture of bin Laden is dubbed in Arabic
with German subtitles. He is shown with a weapon in his hands and a lance in
the background (see Table 5). Bin Laden ascribes the killings of innocent Mus-
lims to the lack of morality in Western societies. He demands the protection of
innocent Muslim women in “their villages of mud” and threatens the West by say-
ing “(the) price ought to be paid”.
TH Lifestyle activist among the Islamic extremist material was selected from the
video “mother stay firm”. A young German Jihadist describes how badly the Um-
mah20 is treated by the West. His argumentation is underlined by background
images from Abu-Grahib and the story of a Muslim woman who was raped there.
He asks his mother while looking into the camera “How shall I sit still?” For the
right-wing extremist material, the video “German open your eyes” was chosen.
Three young men talk into the camera while walking through a German city.
The video givesthe impression that they talk with one voice while the camera iter-
ates. The topics addressed are the victimization and extinction of the Germans by
“the system”. The video ends in an appeal to join the movement. Both videos are
accompanied by music but no lyrics to underline the words of the speakers.
TH Martyr again was only found for the Islamic extremist clips. The selected
scene shows Mohammed Sidique talking into the camera, with German subtitles.
He justifies his bombing plan by referring to crimes committed against Muslims.
He pronounces the guilt of each elector in democratic countries and the derived
duty of each Muslim to take revenge.
Table5 shows screenshots of the four Talking head clips and summarizes informa-
tion about lengths and the topics addressed.
19 The publication of caricatures in a Danish newspaper displaying Mohammed raised worldwide
complaints.
20 Ummah: Community of the believers of Islam
51
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 52
Table 5. Prototypical Scenes for the Talking Head Formats
Subcategory Selected Clip Topics Length
Ideologue Muslim suffer-
ing, innocent
victims, Western
societies fail.
Cut: 00:05:01
Original:
00:05:13
Lifestyle activist Muslim suffer-
ing, innocent
victims, joining
the fight.
Cut: 00:02:46
Original:
00:30:26
Victims of the
system, joining
the fight.
Original:
00:02:04
Martyr Western injus-
tice, Muslim
suffering, Jihad
as duty.
Cut: 00:03:55
Original:
00:27:22
For RC Group lifestyle within the Islamic extremist clips, “Labbaik21”was cho-
sen. Short sequences show members of the armed Jihad while presenting their
weapons and scenes during combat. Throughout the video, the anasheed “Lab-
baik” accompanies pictures of the adventurous group lifestyle. The video does
not have further text. Within the right-wing extremist video clips “Security yes,
Surveillance no- preserve freedom of opinion” was chosen. The video displays
marching right-wing extremists all dressed in the same way. Similar to the Islamic
extremist video, the whole scene is accompanied with music. The sentence “Is
that freedom of speech? Total Surveillance! Is that democracy? Total Surveil-
21 Arabic: “Here I am, ready to obey your order”. Literally a response to a call.
52
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 53
lance!” is repeated over the course of the video. No further text or information is
given.
For RC Public opinion only right-wing extremist examples were found. The clip
“Anti-Globalization” presents interviews with pedestrians concerning the eco-
nomic situation in Germany and around the world. The interviewer is not seen
within the scene but his suggestive questions trigger the responses. The outsour-
cing of workplaces and the lack of integration among foreign workers is criti-
cized.
RC News broadcast for the Islamic extremist videos was represented by scenes
from “The crusade of the Western World”. It deals with a series of (broadcasted)
images about 9/11 along with various Western politicians.
A distorted (male) voice reports the number of Muslims who had already joined
Jihad to oppose the “Western crusade”. Finally he demands German soldiers to
leave Afghanistan. The pictures used and the segmentation of the screen is remi-
niscent of CNN. For the right-wing extremist material a scene from “Revolution
of the consciousness” was selected. In the video, sequences from German mass
media are presented, commented by blending in subtitles like “censorship” or
“stultification”. Media manipulation and the failure of the government are
some of the covered topics.
Table6 shows screenshots of the five clips representing the category Reality clip.
53
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 54
Table 6. Prototypical Scenes for the Reality Clip Formats
Subcategory Selected Clip Topics Length
Group lifestyle Jihadists, group,
trainings camp,
weapons, being
prepared.
Cut: 00:02:17
Original:
00:51:38
National socia-
lists, group,
demonstration,
surveillance,
security.
Original:
00:02:20
Public opinion Globalization,
difficult work
situation,
anti-migration.
Original:
00:01:42
News broadcast Crusade, 9/11,
Afghanistan
war, attendance
of Jihadists.
Cut:
00:03:05
Original:
00:03:59
Censorship,
media lie,
stultification.
Cut:
00:05:18
Original:
00:08:18
54
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 55
For the Islamic extremist Movie clip, a scene from “The Attack on our Siblings in
Iraq” was selected. This high-quality production presents civil victims of the Iraqi
war, accompanied by the “Lord of the Rings” original title theme. The question:
“Who is going to offer protection from the infidels?” is posed and underlined by
dramatic pictures of dead children and crying mothers. The Muslim Caliphate is
proposed as the only answer and is symbolized by “Hizbut al-Tahir22” members
around the world. In the meantime, the “Pirates of the Caribbean” theme is played.
The video is composed of pictures, written text (in German) and dramatic music.
In the right-wing extremist video “The Bombing of Siegen during WWII”, similar
topics are addressed. The video presents an aesthetic black and white paper cut
film about the bomb attack of the allied forces in 1944. The amount of victims
is presented and supported by images of crying women and dead children in ruins.
It focuses on the allies’ aggression, ending with two bombs named “democracy”
and “freedom”.
As Extreme clip for the Islamic extremist videos, a suicide attack from “Goodbye
of a Suicide Bomber”, was selected. The video shows a suicide bomber with his
car and a bomb planted inside. While driving the car to its destination, he is filmed
by a handheld camera and accompanied by chants about paradise (translated in
German). The video ends in an explosion on the horizon. For the right-wing ex-
tremist material, the clip “Tougher Penalties for Child Rapists” was chosen.
The clip promotes a campaign against child abuse. Headlines and photos of vic-
tims taken from German mass media are shown and accompanied by a song about
a young girl being raped. After a cut, national socialists (dressed in black) demon-
strate for higher penalties (again with the “Pirates of the Caribbean” soundtrack).
Table7 shows screenshots of the four clips representing the categories Movie and
Extreme clip.
22 An Islamic, neo-fundamentalist organization rooted in the Muslim brotherhood. The party is
banned in nearly all Arabic countries,Turkey and Germany.
55
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 56
Table 7. Prototypical Scenes for the Movie and Extreme Clip Formats
Subcategory Selected Clip Topics Length
Movie clip War in Iraq,
Muslim victims,
Western lies and
aggressions,
Muslim
Caliphate.
Cut: 00:06:04
Original:
00:23:42
War in Germany,
German victims,
American lies
and aggression.
Original:
00:02:35
Extreme clip Suicide attack,
becoming a
Martyr, paradise.
Cut: 00:02:02
Original:
00:22:57
Pedophilia, child
victims, harsher
penalties.
Original:
00:03:19
9.3 Procedure Study I and II
The selected clips were then presented to a large sample of participants. Study I
and II followed a similar procedure, by focusing on the explicit effects of (various)
extremist video clips (see also Chapter 8).
In order to ensure that participants would neither feel deceived nor get suspicious
about the study aims, recruiting flyers informed about the topic (propaganda) and
invited people with different cultural backgrounds to participate. The study took
place either in a laboratory at the University of Cologne or in classrooms of voca-
tional schools supporting the study. Participants took part in small groups of up to
six people per session. They received about A15 for one hour of their time. One
session took approximately two hours in Study I and one hour and a half in Study
II. After arriving in the room, participants were informed about their rights. This
included the right to quit the experiment at any given time without the risk of los-
ing the reward. Afterwards, participants picked a code at random from a box and
56
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 57
were seated in separate cubicles. Each workplace was equipped with a notebook,
headphones, finger sensors, and a handheld slider (SCR) (see Chapter 9.4.1.1).
Figure 4 displays the setting.
Figure 4. Experimental setting during video reception. The participant was equipped with headphones
and finger sensors on the left hand. He holds the slider in his right hand.
The experiment itself was run on a computer. It started automatically after parti-
cipants had entered their code. In a first step, a series of demographic, personality
and attitudinal questionnaires was presented (described in detail in Chapter 10).
Afterwards, participants were asked to put on the headphones, to rest the slider
(SCR) in their dominant hand and to start the video reception. They started either
with a series of the selected right-wing or with a series of Islamic extremist video
clips (see Chapter 9.2).
After each clip, participants were asked to return the slider to a neutral position
and then answered 20 questions assessing emotional and cognitive reactions to-
wards the single video (see Chapter 9.4.1.2). After each series of video clips
(of one ideology), participants were further forced to choose which one had im-
pressed them the most (“Forced choice”). “Please indicate which of the clips
you remember best? Select by means of the screenshots the clip which impressed
you the most”.
57
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 58
After the last questionnaire, participants were thanked and given a full debriefing
by a psychologist in groups of up to six people. Figure 5 depicts the course of a
session.
Figure 5. Procedure of Study I and Study II. The order of the two series (right-wing vs. Islamic ex-
tremist video clips) as well as the order of the single videos within, were randomized to prevent se-
quence effects (Hussy & Jain, 2002).
9.4 The Effects of Propaganda 2.0 – Immediate Evaluations
Former studies mostly found negative evaluations of propagandistic material and
extremist groups (see Chapter 6). This could, to some extent, be explained by the
methods used. For example, Baines and his colleagues (2006, 2010) asked parti-
cipants during group discussions how they had experienced such videos. How-
ever, propaganda might be evaluated rather negatively in general and participants
might have hesitated to show their agreement openly.
We therefore aimed at combining different measures in order to detect even slight
changes on different dimensions. As indicated in Chapter 9.3, the first two studies
58
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 59
combined so-called online measures during and questions after each video. The
impact a video has on the recipient might vary during exposure. It is a challenge
in communication research to deal with the continuous flow of audio-visual sti-
muli that is characteristic for media exposure. Different emotions, cognitions
and moment-to-moment evaluations are subject to change during reception
(Lang, 2000). Thus, if a video is (exclusively) evaluated afterwards, the judgment
is likely to predominantly reflect the last impression (Bolls, Lang, & Potter,2001).
By the applied combination of online (Chapter 9.4.1) and post-hoc measures
(Chapter 9.4.2) we depicted the process of explicit evaluation formation. The re-
sults for single video clips (Chapter 9.4.3) will thus describe the effects of propa-
ganda ranging from the process of online to post-hoc evaluations.
9.4.1 Online Measures
Online measurements reveal the participants’ reactions during reception by real
time measurement/recording. They comprise physiological arousal towards the
stimulus presented as well as the evaluation of its valence.
Arousal in general can be understood as an energizer of behavior; it prepares and
accompanies activities, experiences and emotions (Ravaja, 2004). One well re-
searched parameter for arousal is Galvanic Skin Response (GSR). It measures
the electrical activity of the skin that varies due to processes (neuronal activation)
in the autonomic nervous system. Due to physiological processes, GSR changes,
for example, when we are startled or frightened (Cantor, 2002). According to Ra-
vaja (2004) it is an excellent description of [the] “drive state or [the] non-specific
energizer of behavior, something that describes the intensity of an experience but
not its quality” (Duffy, 1962; Mandler, 1992). GSR is conceived as a sensitive and
reliable indicator especially for the emotional component of arousal (Grings &
Dawson, 1978; Ravaja, 2004).
The unspecificity of arousal also implies that very intense emotions (positive and
negative) can lead to similar psychophysiological reactions (e.g. tachycardia,
sweating). Thus, the subjective valence of these states is important for their inter-
pretation. Valence can be considered the second dimension of emotional re-
sponses, which also unfolds dynamically during stimulus presentation (Bente,
Aelker, & Fürtjes, 2009)
To measure these two dimensions of online responses towards extremist stimuli,
the “Integrated-Reception-Process analysis (IRP)” (Schmeisser, Bente & Isen-
bart, 2004) was employed. It has been frequently used in the context of advertising
research (Bente et al., 2009; Schmeisser et al., 2004). Arousal in terms of GSR
was recorded using finger sensors (Wild Devine IOM, 2006). Valence was assed
via continuous handheld slider ratings. The slider (scorer ratings = SCR) dis-
played a nine-point rating scale represented by red, yellow and green LEDs. It ran-
59
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 60
ged from –4 (red =very unpleasant) over 0 (yellow =neutral), to +4 (green =very
pleasant).
Data aggregation
GSR measurements were z-transformed on an individual level and controlled for
outliers following a z-criterion of +/–3 standard deviations (Field, 2009)23. The
transformed GSR averages were then calculated to represent the average arousal
for each video. The same procedure was applied to SCR.
To understand the average online effects of extremist propaganda, the data for
GSR and SCR were aggregated to a single mean for each participant and video
clip.
9.4.2 Post-hoc Measures
After each video clip, participants answered 20 items asking their cognitive and
emotional evaluations. These items were assessed on a 4-point Likert-scale ran-
ging from 1 (= I absolutely disagree) to 4 (= I absolutely agree). Five items
addressed the acceptance of message style (e.g. “I couldn’t take this video ser-
iously”), and five items the acceptance of content (e. g. “After this video I can un-
derstand the perspective of its makers much better”). Ten additional items asked
for different emotions as “interest, anger or boredom” on a 5-point Likert-scale
ranging from 1 (= not at all)to5(=very strong) (M-DAS, Renaud & Unz, 2006).
Data aggregation
The items were analyzed via multilevel factor analysis24. The factor analysis is
used to detect latent variables which cannot be measured directly25. In a first
step, an Exploratory Factor Analysis (EFA) was conducted to detect the underly-
ing structure. The term “multilevel” refers to the fact that each of the items had to
be answered multiple times (after each video) and by multiple participants.
23 Z-transformation standardizes the raw values to an extent that the mean value equals zero and the
other values are normally distributed around it. 68% of all values lie within –1 and +1 standard
deviations. Z-Values of +/– 3 have a probability of about 99.9 % to belong to the same distribution
of values (e.g. because they did result from video-induced arousal but not from movement arti-
facts).
24 A factor analysis contributes to a) an understanding of the structure underlying the questions
applied; and b) to reduce them to relevant dimensions (Field, 2009) via analyzing the associations
between the items. Items representing the same factor should be associated more strongly with
each other than with all other items. Generally, exploratory (EFA) and confirmatory (CFA) factor
analyses can be distinguished.
25 Conclusiveness of an argument, for example, is a latent variable but might be mirrored in the
answers to items as “more facts named”, a “better rhetorical style” and “more importance for the
receiver”. Factor analyses help detect whether single items represent the same underlying con-
struct.
60
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 61
It has to be kept in mind that different answers to one item could have occurred due
to differences between the video clips when evaluated by the same participant (le-
vel 1, the so-called within-level) as well as due to differences between the parti-
cipants evaluating the same video clip (level 2, the between-level). Further inter-
actions between the levels (cross-level) were possible.
In total, five latent variables or factors describing how the participants evaluated
extremist propaganda were detected. A total of ten out of the assessed twenty
items were identified to represent the post-hoc evaluation of propaganda. The
further analyses will thus focus on them.
The factors were found on both levels. They are thus able to display effects vary-
ing within single video clips as well as differences between participants. Factors
were proofed via CFA and remained stable within different educational levels as
well as within different cultural samples.
Three of the factors described the participants’ emotional reactions: Aversion,
Shame and Interest. While the first two are negatively connoted and express rejec-
tion of the material; the latter one can be understood as a first step of an approach
towards the message. The other two factors covered items referring to the cogni-
tive evaluation of the stimuli. The first one describes to what extent a video was
ascribed One-sidedness, the second the extent of its Persuasiveness. Again, ad-
verse and approaching effects were distinguished. Table 8 gives an overview of
the single items forming the five factors.
For further statistical analyses the items which represented the same factor were
aggregated to a single mean for each participant and video clip.
61
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 62
Table 8. Items and Factors of the Immediate Evaluation of Propaganda
Factor Items
After I had watched this video . . .
Emotional Evaluations
Aversion . . . I felt contempt
. . . I felt disgust
. . . I felt anger
Shame . . . I felt shame
. . . I felt guilt
Interest . . . I got interested
. . . I felt fascinated
Cognitive
Evaluations
One-sidedness . . . I found this video to be one-sided.
. . . I found this video to be sensational.
. . . I couldn’t take this video seriously.
Persuasiveness . . . I have now more compassion with the makers of this video.
. . . I can understand the makers’ perspective much better.
. . . I felt to be better informed about the background.
. . . the video was persuasive in my opinion.
Note. All emotional items were answered on a f ive-point Likert-scale from 1 (= not at all) to
5 (= very strong). Cognitive evaluations were given on a four-point Likert-scale ranging
from 1 (= absolutely disagree)to4(=absolutely agree). Appendix1 gives an overview of
the factor-loadings and statistical details.
9.4.3 Results for Single Videos
Our second research aim was to investigate on which dimensions the videos dif-
fered with regard to the effects they evoked in recipients (see Chapter 7).
Q2: Do distinct production formats and message styles differ with regard to
their potential to create interest or to trigger rejection?
For this purpose we analyzed online reactions (GSR and SCR) and post-hoc eva-
luations of N= 133 student participants in Study I. As described in Chapter 9.4.2,
the post-hoc factors can be distinguished into those rather evoking adverse evalua-
tions (Aversion,Shame,One-sidedness) and those suggesting an approach to-
wards extremist videos (Interest,Persuasiveness). During this first step, all 13 vi-
deos (see Chapter 9.2) were analyzed. Separate rANOVAs26 for the right-wing and
the Islamic extremist video clips were conducted. Table 9 reports the statistical
results for the within factor video clip (i. e., differences regarding the effects
due to differences between the single videos). For galvanic skin response
26 Analysis of variance (ANOVA) is able to answer the question if more than two means differ
significantly from each other (here, for example, the mean evaluations of the singlevideo clips).
Variance explained (e. g., resulting from the different videos)is compared to variance unexplained.
Repeated measurement ANOVAs (rANOVAs) were used as all participants had watched and
evaluated all video clips.
The chosen level of signif icance for the whole study was a= .05, meaning that with a probability of
95 % the values compared indeed varied from each other in the sample analyzed and would do so
in the population as well.
62
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 63
(GSR), the right-wing extremist videos did not differ among each other concern-
ing the evoked arousal, but the Islamic extremist videos did. SCR showed that the
videos of both ideologies differed in regard to the evoked evaluations of pleasant-
ness.
Table 9. rANOVA Results for the Online Measures for the within Factor Video Split by Ideology27
Ideology Online
measure F(df) p hp2
Right-wing extremist GSR F(5, 535) = 1.26 n.s. .01
SCR F(4.53, 571.27) = 32.35 < .001 .20
Islamic extremist GSR F(6, 636) = 5.27 < .001 .05
SCR F(5.08, 639.86) = 8.25 < .001 .06
Note. The table shows the main effects for the within factor “video clip”. Fis the statistical
parameter in ANOVAs measuring the ratio of the variance explained vs. the unexplainable
variance. As a rule of thumb bigger Fs indicate stronger effects. Df = degrees of freedom.
p= probability of the found differences to be random vs. significant. hp2= (partial) effect
size.
A closer look at single video clips (see also Figure 6) in order to interpret the re-
ported effects, revealed two main patterns:
(1) Arousal (GSR) did not change much depending on the video clip, except for
one Islamic extremist clip, RC Group lifestyle28. This clip contains high levels of
dramatic noise (exploding bombs, shootings), so one possible reason for the
higher arousal might be that participants are alerted. The corresponding negative
slider-ratings of the clip support this idea.
(2) Slider ratings (SCR) overall showed negative evaluations of propaganda video
clips. However, there were some videos which stood out due to either very nega-
tive or rather neutral SCRs.
For both ideologies, the TH Lifestyle activist and the Extreme clips were evaluated
most negatively while the Movie clips were evaluated as nearly neutral. The other
video clips showed distinct evaluations depending on ideology. The RC News
broadcast was rated neutrally when containing a right-wing extremist ideology
and very negatively for the Islamic extremist clip. Both categories that appeared
to be unique to one of the ideologies (TH Ideologue and Public opinion) were also
perceived to be rather neutral. Figure 6 displays the online measures for the right-
wing and Islamic extremist videos.
27 Single comparisons are depicted in Appendix 2.
28 Due to its low discriminative and explanatory power, GSR is no longer reported in the subsequent
chapters.
63
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 64
Figure 6. GSR and SCR measures for the different videos split by ideology. For displaying purposes
GSR and SCR values were z-transformed.
Besides online responses, we were also interested in the post-hoc evaluations after
each clip. In a first step, we analyzed the association between the slider ratings and
the post-hoc factors. The relations between variables (SCR and the post-hoc eva-
luations) were measured via correlational analyses29. It is important to note that
correlations do not allow causal interpretations.
29 Correlations indicate a) whether relations are positive, thus an increase, for example, in SCR is
associated with an increased post-hoc evaluation, or negative (an increase in SCR ratings asso-
ciated with a decreased post-hoc evaluation). They are expressed via r. The higher its value, the
stronger the relation. r= 1 is the highest possible correlation.
64
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 65
Table 10 shows that first impressions made during the reception of a video clip
already indicated tendencies for later evaluation. Videos which were rated as
very unpleasant (SCR) were also reported to be more adverse afterwards. Aver-
sion, Shame and One-sidedness correlated negatively with SCR during exposure.
Interest and Persuasiveness correlated positively with SCR. It can be concluded
that combining both online and post-hoc measures sufficed to depict an evalua-
tion formation process.
Table 10. Correlations between Online Measures and Post-hoc Factors
SCR Video Aversion Shame
One-
sided-
ness
Interest Persuasive-
ness
Right wing
extremist videos
TH Lifestyle acti-
vist –.48** –.09 –.34** –.00 .18*
RC News broadcast –.21*–.19*–.13 .11 .06
RC Group lifestyle –.32** –.09 –.27** .19*.34**
RC Public opinion –.40** –.26** –.36** .25** .36**
Movie clip –.18*–.04 –.13 –.01 .05
Extreme clip –.25** .08 –.32** .16 .28**
Islamic
extremist videos
TH Ideologue –.13 .19*–.18*.20*.27**
TH Lifestyle acti-
vist –.32** –.01 –.01 .08 .14
TH Martyr –.36** .05 –.08 –.02 .11
RC News broadcast –.21*.16 –.16 –.02 .13
RC Group lifestyle –.44** .06 –.06 .05 .05
Movie clip –.13 –.02 –.22*.15 .21*
Extreme clip –.26** .11 –.06 .02 .14
Note. *p< .05, ** p< .01 Significant correlations are marked in boldface. Correlations
r.10 can be regarded as low, r.30 as medium, r.50 as high (Field, 2009).
In the next step, the post-hoc factors were also analyzed via separate rANOVAs for
the right-wing and Islamic extremist videos. Table 11 depicts that again differ-
ences between the video clips elicited significant effects.
65
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 66
Table 11. rANOVA Results for the Post-hoc Measures within Factor Video Split by Ideology
Ideology Post-hoc Factor F(df) p hp2
Right wing
extremist
Aversion F(4.27, 546.42) = 77.89 < .001 .38
Shame F(4.03, 515.32) = 11.67 < .001 .08
One-sidedness F(4.86, 515.32) = 622.64 < .001 .17
Interest F(4.60, 588.10) = 23.55 < .001 .16
Persuasiveness F(3.72, 476.00) = 19.94 < .001 .14
Islamic
extremist
Aversion F(4.92, 630.24) = 24.27 < .001 .16
Shame F(4.42, 565.53) = 5.89 < .001 .04
One-sidedness F(5.52, 705.93) = 9.53 < .001 .07
Interest F(5.49, 702.69) = 16.12 < .001 .11
Persuasiveness F(4.26, 544.76) = 18.22 < .001 .13
Note. The table shows the main effects for the within factor “video clip”. Fis the statistical
parameter in ANOVAs measuring the ratio of the variance explained vs. the unexplainable
variance. As a rule of thumb, bigger Fs indicate stronger effects. Df = degrees of freedom. p
= probability of the found differences to be random vs. significant. hp2= (partial) effect
size.
Concerning the post-hoc evaluations indicating a rather adverse reaction towards
propaganda videos, Figure 7 shows that the most outstanding reaction among all
videos was Aversion. Especially the Extreme clips of both ideologies received
high ratings here.
This was not very surprising as these videos deal with controversial topics (see
Chapter 9.2), namely pedophilia and martyrdom. It might also be due to these to-
pics that participants also report high levels of Shame after watching them.
High Aversion was also found for both TH Lifestyle activist videos. These videos
share a commonality in that they try to recruit and motivate people to join the
movement. Furthermore, both try to justify extremist actions by emphasizing
the victimization of the “ingroup”. Moreover, they blame the audience for its ig-
norance and for not rising against the committed injustice. This might explain
why they – similar to the Extreme clips – also evoked high ratings in Shame.
Due to arguing from the “victims’” perspective only, it is not surprising that par-
ticipants also rated these videos to be rather one-sided. In line with this the Movie
clips also elicited feelings of shame. Although the videos are produced rather pro-
fessionally, it is reported from only one perspective. This might have triggered the
high one-sidedness ratings found.
Several videos of the category Reality clip – although seemingly presenting an
“objective” perspective– were rated high on one-sidedness. For example, the Is-
lamic extremist RC News broadcast deals with the extremists’ view of the actions
66
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 67
of 9/11 and their subsequent claims. Unsurprisingly, this famous terrorist action
was perceived to be one-sided because the video neglects the perspective of the
victims. As already implied in the online measures, the Islamic extremist RC
Group lifestyle also evoked very negative online ratings.
Figure 7. Adverse post-hocevaluations for the different videos split by ideology. All emotional items
were answered on a five-point Likert-scale from 1 (= not at all)to5(=very strong). Cognitive eva-
luations were given on a four-point Likert-scale ranging from 1 (= absolutely disagree) to 4 (= abso-
lutely agree).
67
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 68
Strikingly, those videos evoking negative reactions also elicited evaluations which
can be described in terms of a first approach (see Figure 8). For example, among
the right-wing extremist clips, the Extreme and Movie clips evoked high Aversion
and Shame and were evaluated as one-sided. Nonetheless, they also stimulated In-
terest. Similarly, the RC News broadcast received high ratings on Aversion and
One-sidedness as well as the highest ratings on evoked Interest. In general, the
Persuasiveness of the videos was always rated lower than the evoked interest.
Within this restriction, RC News broadcast and the Movie clips were rated to be
the most persuasive ones for the right-wing extremist material.
The same is true for the Islamic extremist video sample: TH Lifestyle activist, the
Movie and Extreme clip evoked the most Interest and were rated as aversive and as
evoking shame. TH Lifestyle activist and (as for the right-wing extremist material)
the Movie clip were also rated as the most persuasive ones.
68
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 69
Figure 8. Approaching post-hoc evaluations for the different videos split by ideology. All emotional
items were answered on a five-point Likert-scale from 1 (= not at all) to 5 (= very strong). Cognitive
evaluations were given on a four-point Likert-scale ranging from 1 (= absolutely disagree)to4
(= absolutely agree).
To draw a first conclusion, it seems that the videos either evoked some kind of
reaction (negative and positive) or they did not. It might be that videos not evoking
reactions might be evaluated as boring and thus are not able to trigger the first two
steps in the AIDA Model (Attention and Interest, (E.S. Lewis, 1903, see chap-
ter 1). Analyzing ratings on the item “boring” as a further control (see Chapter
9.3.1.2) supported this idea. Videos that were rated as boring did not show high
rankings on the post-hoc factors. For each video, the evaluation of the video as
boring was negatively correlated with Interest and Persuasiveness. Concerning
Interest, all correlation coefficients were significant at p< .01, r–.28, indicat-
69
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 70
ing that boredom and Interest in the same clip really contradict each other. With
regard to Persuasiveness, the majority of correlation coefficients were significant
at p< .05, r–.21, only for those Islamic extremist clips dealing with suicide
attacks (TH Martyr and Extreme clip) the correlation did not become significant.
Although these video clips were rated as interesting, they were not rated as very
persuasive.
Also, the participants’selection of the most impressive videos from each ideology
supported this idea. Generally, the memorization of certain videos within one
ideology can be seen as a further indicator of the impact of the presented stimuli.
As already stated in theories on advertising effects, material has to be recalled in
order to allow further processing (Moorman, 2003).
Figure 9 gives an overviewof the videos which were recalled best after presenta-
tion of all videos within one ideology. It can be concluded that videos evoking a
reaction (even the videos evoking negative reactions such as Aversion or Shame)
were also recalled better after reception.
70
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 71
Figure 9. Absolute number of recalls per video split by ideology.
In both ideologies, the clips TH Lifestyle activist,Movie and Extreme clip not only
stood out concerning their online and post-hoc ratings, but were also evaluated as
the most impressive ones. Crucial points might be the direct addressing of the
audience as well as their professional production style. Further, all six videos
combine animated pictures accompanied by music. The other formats (only) af-
fected the audience to a lesser extent.
As Study I also aimed at reducing and refining the stimulus material, for the sec-
ond study only the six videos evoking the strongest responses were included. We
made one exception; RC Group lifestyle videos were included in the subsequent
studies as well. One the one hand, the content analysis (see Chapter 9.1) has iden-
tified them as the most frequent format for both ideologies. On the other hand,
71
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 72
Baines et al. (2010) supposed that particularly such adventurous video clips might
stimulate to approach Jihad. Furthermore, with the selection of these four clips per
ideology, all four main categories of the content analysis were still realized. By
adding this category to the next study, the variability of formats (Talking heads,
Reality, Movie and Extreme clips) was preserved. Statistical analyses (see also
Chapter 8) confirmed the comparability of Study I and Study II regardless of vi-
deo reduction. Subsequent analyses thus present Study I and Study II together fo-
cusing on those eight clips found to have the largest effects on the audience.
Besides all communalities between Islamic and right-wing extremist propagation,
people can only be attracted (“recruited”) by one of them. Thus, we focused on
ideology as the main underlying factor, irrespective of the production style or for-
mat of the single video clips. We combined all video clips sharing the same ideo-
logical background by computing a single mean for each online and post-hoc fac-
tor for “right-wing” and “Islamic” extremist material. Including “Ideology” as a
factor allowed analyses of how participant characteristics influenced the evalua-
tions of either Islamic or right-wing extremist propaganda in general.
9.4.4 Summary: Research Focus I
R1: Description of right-wing extremist and Islamic extremist propaganda videos
and analysis of their effects.
Chapter 9 addressed our first research focus and aimed at characterizing modern
right-wing and Islamic extremist internet videos and giving first insights into their
effects. Four prototypes of extremist propagandawere found to characterize right-
wing as well as Islamic extremist audio-visuals linguistically understandable for a
German audience.
Q1: Which prototypical formats of right-wing and Islamic extremist propa-
ganda in Germany can be distinguished? Which messages are transported?
Analogous to the work by Baines and his colleagues (2006, 2010) as prototypes,
Talking head formats,Reality and Movie clips were distinguished. Furthermore,
Extreme clips as an additional category were included. According to the descrip-
tion of propaganda by Merten (2000), the structure of Islamic and right-wing ex-
tremist propaganda was similar. The suffering of the own ingroup, particularly of
innocent children and/or women, was opposed to the faults of an outgroup.
Furthermore, the extremist group was presented as a cohesive entity willing to ad-
dress the problems depicted before.
Q2: Do distinct production formats and message styles differ with regard to
their potential to create interest or to trigger rejection?
Indeed, differences were found in the potential of extremist propaganda to catch
the audience’s attention and trigger their response. Videos rated as boring could
not develop a possibly persuasive potential and were not sustainably recalled.
72
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 73
In contrast, videos “touching the recipient” in terms of playing with his emotions
(unpleasantness, aversion, shame) were also able to create slight reactions of in-
terest and persuasiveness.
Our video selection was derived from this analysis and those videos able to attract
attention were chosen. For the Talking head formats those clips displaying a Life-
style activist, were chosen. Further, the professional Movie clips and the Extreme
clips were perceived as potentially attracting. The Reality clips were identified as
the most frequent format available and were thus also included in the analyses.
However, this format overall had a smaller affecting potential compared to the
other three formats.
73
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 74
10. Research Focus II: Propaganda and Different Audiences
10.1 Characteristics of the Audience
The study focused on the potential ingroup of extremist propaganda. Wewere in-
terested in the reactions of “normal young adults” when they are confronted with
extremist videos.
A self-developed scale (Table 12) assessing the acceptance of terrorism justifica-
tions was thus applied to control for recipients already accepting terrorist means.
Table 12. Justifications of Terrorism Scale
Instrument Example items Rating scale
Justifications of Terrorism
(own scale)
“Terrorism can be morally justi-
fied if
. . . it helped the expulsion of an
occupying power”
“. . . it was about restoring social
justice”
“. . . it was about fighting for one’s
religious beliefs”
7-point Likert
(1 = totally disagree;
7=totally agree)
Note. Answers were aggregated to a single value expressing the justifications of terrorism
(Cronbachs a30 = .72 in Study I and a= .73 in Study II).
As extremist propaganda mainly focuses on young men (see Chapter 5), only male
participants took part in the study. We further assessed demographics as age,
media usage behavior, cultural background, educational level, and the major stu-
died.
Against the theoretical background depicted in Chapter 5, the characteristics sus-
pected to contribute to a recipient’s approach towards extremist content were con-
sidered (see Table 13). As one possible moderator, the self-reported German/cul-
ture of origin social identity self-concept (Maehler & Schmidt-Denter, n.d.) was
measured. The scales by Maehler (2011) measure distinct cultural identities with
the same six questions for each cultural identity of the participant (German and, if
existing, culture of origin). Questions concerning the culture of origin were only
presented when a migration background was reported.
To cover personality factors, the Big Five (Costa & McCrae, 1992) and a sensation
seeking scale were included. The big five (extraversion, neuroticism, conscien-
tiousness, openness and agreeableness) were assessed on a 10-item scale (Ramm-
30 Cronbachs ais a measure of the internal consistency of a scale and expression of its reliability.
a< .5 = unacceptable, < .6 = questionable, <.7 = acceptable, < .8 = good, < .9 = excellent.
74
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 75
stedt, 2007; Rammstedt & John, 2007). Sensation seeking was measured by
AISS-D (Roth & Hammelstein, 2003) which contains 20 items.
In order to address attitudinal patterns, lack of control, political attitude, author-
itarianism and acceptance of violence were included. The lack of control was
measured by means of 15 items by Agroskin and Jonas (2010) encompassing
items for political, economic and personal lack of control. The political attitude
was measured with the help of a single item, directly asking for the political atti-
tude on a five-point scale (Wasmer, Koch, Harkness, & Gabler, 1996). The assess-
ment of authoritarianism combined the two items by Frindte and Haußecker
(2010) as well as the eight item scale by Petzel,Wagner, Nicolai, and Van Dick
(1997).
As described in Chapter 5, acceptance of violence can be understood as a second
component in an approach towards extremism. We applied the concept by using
the six-item scale from Wagner, Christ and Kühnel (2002). See Table 13 for an
overview of all scales used. Variables not assessed via scales (gender, age, media
usage) are not included in the table.
75
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 76
Table 13. Personality and Attitudinal Variables Assessed
Category Instrument Example Items Rating scale
Self
concept
German/Non-German
Identity Self-concept
(Maehler & Schmidt-
Denter, n.d.)
“I feel German/Culture
of Origin”
5-point Likert
(1 = absolutely not ap-
plying;5= absolutely
applying)
Personality Factors
Sensation seeking
(Roth & Hammelstein,
2003)
“When I listen to music,
it should be loud” or
“On a general level, I
work better under pres-
sure”
4-point Likert
(1 = absolutely not ap-
plying, 4=absolutely ap-
plying)
Big-Five
(Rammstedt & John,
2007)
“I see myself as some-
one . . . who is relaxed,
who handles stress
well”, or “. . . who gets
nervous easily”
5-point Likert
(1 = absolutely not ap-
plying;5= absolutely
applying)
Attitudinal Patterns
Lack of Control (Agro-
skin & Jonas, 2010)
“It is hard for us to in-
fluence the actions of
politicians”
5-point Likert
(1 = absolutely not ap-
plying;5= absolutely
applying)
Authoritarianism
(Petzel et al., 1997)
“The most important
thing, children should
learn is obedience and
respect for parents and
superiors”
5-point Likert
(1 = totally disagree;
5=totally agree)
Authoritarianism
(Frindte & Haußecker,
2010)
“How important is it in
society to obey under
authorities?”
6-point Likert
(1 = totally unimportant,
6=totally important)
Political Attitude (Was-
mer et al., 1996)
“How would you clas-
sify your political atti-
tude?”
5-point Likert (1 = left-
wing, 5=right-wing)
Acceptance of Violence
(U. Wagner et al., 2002)
“Sometimes I have to
use violence to not draw
the short straw” or “The
usage of violence can
create clear conditions”
4-point Likert
(1 = absolutely not ap-
plying;4= absolutely
applying)
Data Preparation
Based on their demographic variables, participants were split into cultural and
educational groups. Three cultural groups were distinguished: 1) Participants
with both parents originating from Germany and a Christian or no religious af-
filiation were classified as “Germans”. 2) Participants with both parents originat-
ing from a Muslim country (e.g. UAE,Turkey, or Iran) and Muslim religious af-
filiation were referred to as “Muslims”. 3) Participants with parents originating
from a non-Muslim country (e.g. Latin America, Russia) and either Christian
76
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 77
or no religious affiliation were classified as “Others”. All other combinations
(e.g. Russian Jews) were excluded from further analyses in order to ensure that
groups were big enough to be compared statistically.
Concerning educational level, only students and pupils from vocational schools
were included in the analyses. If available, their major fields of study were manu-
ally coded as social-/humanist (e.g. psychology, philosophy) or science-oriented/
technical (e.g. physics, engineering) ones.
For the personality and attitudinal questionnaires, the items were recoded to
such an extent that higher values expressed higher manifestation of the accordant
trait. To enhance comparability, all items of the lack of control scale where re-
coded to the extent that higher values represented higher feelings of control in-
stead of lack of it. To avoid confusion, we thus refer to the accordant scale
from now on as “feeling of control”.
In accordance to literature, values for the single items were aggregated to a single
value for each scale.
10.2 Sample
A total of N31 = 339 participants took part in both Studies I and II. However, as
already mentioned in Chapter 10.1, the differentiation in cultural subgroups led
to the exclusion of some participants. A total of N= 39 participants could not
be taken into consideration for further analyses due to statistical reasons. Further
N= 23 had to be excluded due to their professional status (neither students nor
pupils). The remaining N= 277 were on average M= 24.14 (SD = 4.54) years old.
As far as the educational level is concerned, n= 160 participants were students and
n= 117 were pupils from vocational schools. Table 14 shows the absolute number
of participants with different cultural backgrounds within the educational sub-
groups.
Table 14. Total Number of Participants Split by Cultural Background and Educational Level
Educational Level
Students Pupils from
vocational schools
Cultural
Background
German 98 66
Muslim 24 27
Other 38 24
Note. The distribution of cultural background did not differ significantly32 among the
educational levels, x2(2) = 2.98, n. s.
31 N= Number of participants, M= Mean, SD = Standard Deviation.
32 Tested with a Chi-Square test. Chi-square tests analyze whether or not the frequencyof a variable
(e.g. cultural background) differs signif icantly between different groups (e.g. educational levels).
77
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 78
For a subgroup of n= 56 students in Study II, the major fields of study were ad-
ditionally assessed. Two major fields were differentiated: 39.3 % studied a huma-
nistic subject, 60.7 % pursued technical, scientific-oriented studies.
As the focus here lay on online propaganda addressing the digital natives, the
media usage was also assessed among students and pupils from vocational
schools. Similar to the results of the JIM Study in 2011, the vast majority of all
participants could be described as frequent online users. 97.9 % reported to
have their own notebook with internet access. Most of them (85.3 %) described
using the internet for a broad range of purposes (e. g., social media, emails, e-
shopping, video clips, etc.); 10 % reported using the internet frequently and
only 4.7 % stated that they only used it occasionally (e.g., only emails and infor-
mation search). Regarding the frequency of extremist online propaganda, it is thus
not surprising that 66.7 % of the students and 79.5 % of the pupils had already
found an extremist video clip online. Table 15 gives an overview of the clips found
before the participation in our study. Though participants were more familiar with
right-wing extremist videos, more than half of the students and a third of the pu-
pils also had pre-experiences with Islamic extremist material.
Table 15. Percentages of Videos Found, Split by Ideology
Have you already found a (n). . . .
Educational level Right-wing extremist video? Islamic extremist video?
Students yes 83.3 % 51.1 %
no 16.7 % 48.9 %
Pupils yes 85.7 % 39.6 %
no 14.3 % 60.4 %
Note. Chi-Square tests33 revealed that pre-experience with right-wing extremist material
did not differ among the samples, x2(1) = 0.46, n.s. However, students reported
significantly more pre-experience with Islamic extremist propaganda, x2(1) = 5.92,
p< .05, F= –.18.
Although a majority of the sample had already come into contact with extremist
video clips, only a few reported media to have a huge influence on them. The an-
swers reflect a typical third-person-effect (Davison, 1983; Duck, 1995). While the
impact of media on oneself is perceived as low, the others are thought to be easily
influenced (see Figure 10).
33 F=effect size measure.
78
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 79
Figure 10. Perceived influence of the mass media.
Chapter 10.1 already introduced the recipient variables we assessed in order to get
a more detailed description of our sample and to enable a more precise picture of
potential propaganda effects. In terms of justifications of terrorism, our sample
indeed could be classified as unsuspicious. On the applied scale between 1 (= to-
tally disagree) and 7 (= totally agree), our participants had a mean value of
M=2.36 (SD = 0.33). As far as the other recipient variables are concerned,Table
16 gives an overview of the average manifestations of the single aspects.
79
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 80
Table 16. Average Manifestation of Assessed Recipient Variables
Category Instrument Average manifestation Possible Range
(Minimum-
Maximum)
Self
concept
German Identity Self-Concept
(Maehler & Schmidt-Denter,
n.d.)
M= 3.45, SD = 0.82b1–5
Personality Factors
Sensation seeking
(Roth & Hammelstein, 2003)
M= 55.71, SD = 6.31c20 –80
Big-Five
(Rammstedt & John, 2007)
1–5
Extraversion M= 3.44, SD = 0.94a
Neuroticism M= 3.35, SD = 0.91a
Conscientiousness M= 3.31, SD = 0.87a
Openness M= 3.91, SD = 0.88a
Agreeableness M= 3.05, SD = 0.75a
Feeling of Control (Agroskin &
Jonas, 2010)
M= 2.77, SD = 0.65a1–5
Attitudinal
Patterns
Authoritarianism
(Frindte & Haußecker, 2010;
Petzel et al., 1997)
M= 28.71, SD = 7.17b10–52
Political Attitude M= 1.19, SD = 0.39a1–5
Acceptance of Violence (U.
Wagner et al., 2002)
M= 1.96, SD = 0.62a1–4
Note. Higher values express higher manifestations of the accordant characteristic. Only for
political attitude higher values express a more right-wing political attitude. Indices indicate
manifestation compared to scale mean when standard deviation is considered a) low, baver-
age, chigh. M= Mean, SD = Standard deviation.
80
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 81
10.3 Results for Ideology
10.3.1 Demographics and the Evaluation of Propaganda
Q3) Do recipients react differently to right-wing than to Islamic extremist
propaganda?
Q4) Does this reaction depend on the cultural background or other demo-
graphics of the participant?
As already described in Chapter 5, we planned to investigate whether these differ-
ences were due to potential ingroup relations between sender and recipient, inter-
acting with other demographics. Thus, particularly the cultural background and
the educational level of participants seemed to be of major interest. Hence, sepa-
rate 3 (Culture: German vs. Muslim vs. Other) × 2 (Educational level: students vs.
pupils) × 2 (Ideology: Islamic vs. Right-wing extremist clips) mixed34 ANOVAs,
with the last factor serving as a repeated factor, were conducted for each of the
dependent variables. Table 17 gives an overview of the significant interactions be-
tween the three factors35.
34 Mixed ANOVA because–in contrast to the rANOVAs described in Chapter 9.4.2–besides a within
factor (here: ideology) also between factors (culture and education) were included. Besides main
effects of each of the three factors, interactions between them were possible as well. An interaction
generally describes how the effect of one factor (e.g., cultural background) on the outcome varies
with differences in the other factor (e.g., educational level). Two-way interactions between two
factors and a three-way interaction between all three factors were possible. The latter is the “higher
one”. In case of significant interactions, the highest ones should always be interpreted as they
describe the full effect. For example, it might be that generally right-wing extremist videos are
evaluated as less unpleasant but that this effect is onlysignif icant among one of the cultural and/or
educational groups. Quality of these interactions is described later on. For further questions please
contact the authors.
35 Results for the pairwise comparisons are depicted in Appendix 3
81
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 82
Table 17. Results of the Mixed ANOVAS for SCR and the Post-hoc Measures
DV Interaction F(df)php2
SCR ideology × cultural
background × educational level
F(2, 183) = 5.021 < .01 .05
Aversion ideology × cultural background F(2, 266) = 3.538 < .05 .03
Shame ideology × cultural background F(2, 266) = 21.447 < .001 .14
One-sidedness ideology × cultural
background × educational level
F(2, 266) = 6.299 < .01 .05
Interest ideology × cultural
background × educational level
F(2, 266) = 24.308 < .001 .16
Persuasiveness ideology × cultural
background × educational level
F(2, 266) = 21.534 < .001 .14
Note. For each mixed ANOVA the highest significant interaction is reported. The inter-
pretation of these interaction effects can be derived from the main body of the text. Fis
the statistical parameter in ANOVAs measuring the ratio of the variance explained vs.
the unexplainable variance. As a rule of thumb, bigger F’s indicate stronger effects.
df = degrees of freedom. p= probability of the found differences to be random vs.
significant. hp2= effect size.
As already implied in Chapter 9.4.2, all propaganda videos were evaluated as un-
pleasant during reception and rather negatively afterwards (post-hoc). However,
different shades of this rejection were found in the six subsamples (student vs. pu-
pil × German vs. Muslim vs. other). The analysis of the online as well as the post-
hoc factors revealed a consistent pattern.
In general, Germans evaluated propaganda during reception more negatively
(SCR) than the other two cultural groups. However, this effect varied with the
ideology of the clips as well as the educational level of the participants.
During reception, Muslim participants reported less unpleasantness than the other
two groups. However, for the Islamic extremist material, Muslim students
reported significantly more unpleasantness than Muslim pupils. The two educa-
tional levels did not differ significantly regarding the right-wing extremist mate-
rial. As depicted in Figure 11, a trend is evident – though not significant- that
German and Muslim students rated propaganda of the own ingroup as more un-
pleasant than the other material and pupils showed the mirrored pattern; judging
the material of the respective other group as being more unpleasant.
82
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 83
Figure 11 Slider-ratings (SCR) in the subsamples split by ideology. Scale ranged from –4 (=very un-
pleasant) to +4 (=very pleasant).
The impression derived from the online ratings (SCR) was also found for the post-
hoc factors. Concerning the evoked aversion and shame, the evaluation only de-
pended on the ideology of the video clips and the participants’ cultural back-
ground. In general, the Islamic extremist ideology was reported to be more ad-
verse than the right-wing extremist one. However, this difference only reached
significance for the Muslim participants. They also evaluated all videos to evoke
more shame than other immigrants and Germans did. However, once again, this
difference was most pronounced for the Islamic extremist material. Germans,
in contrast, reported more shame after the right-wing extremist clips. Other immi-
grants – who might not feel addressed by either ideology – reported similar levels
of shame for both.
For the remaining three factors, One-sidedness,Interest and Persuasiveness, the
evaluation of propaganda differed between the ideologies, the cultural back-
grounds and the educational levels.
As for the slider-ratings, German students evaluated the videos overall as more
one-sided than the other two cultural groups. This effect varied with the ideology
of the clips. German participants evaluated particularly right-wing extremist pro-
paganda as more one-sided than the Islamic extremist material. With regard to the
Muslim participants, pupils judged the videos to be more one-sided than students.
This effect concerned particularly the right-wing extremist material, which was
evaluated as more one-sided than the Islamic extremist one. Other immigrants
83
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 84
did not differ concerning the evaluation of right-wing and Islamic extremist ma-
terial (see results for Aversion,Shame and One-sidedness in Figure 12).
Figure 12. Adverse post-hoc evaluations for the different subsamples split by ideology.
84
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 85
Both post-hoc factors expressing a cautious approach towards an ideology (Inter-
est and Persuasiveness) yielded a convergent pattern. While German students re-
ported less interest for right-wing extremist propaganda, Muslim students rated
the Islamic extremist material as less interesting. They also reported less persua-
siveness for the Islamic extremist material. Within the pupil subsample, as found
by the preceding measures, the pattern changed. German pupils reported more in-
terest and persuasiveness for the right-wing extremist material and Muslim pupils
for the Islamic extremist material. Further, pupils significantly differed from stu-
dents with the same cultural background when evaluating propaganda of the own
ingroup. Pupils evaluated this propaganda as more interesting.
Participants with another cultural background neither differentiated between
right-wing and Islamic extremist material nor did their evaluations differ in terms
of educational level. Figure 13 shows the pattern for the approaching post-hoc fac-
tors.
85
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 86
Figure 13. Approaching post-hoc evaluations for the different subsamples split by ideology
86
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 87
To sum up, a coherent picture emerged from the online and post-hoc measures
when focusing on participants’ cultural and educational background. The main
difference between both educational levels was that students tried to distance
themselves from the apparent “cultural ingroup” propaganda whereas pupils
from vocational schools reported to be in favor of said propaganda. Other immi-
grants showed no such systematic reactions towards the propaganda material.
It has to be taken into consideration the fact that propaganda in general has a ne-
gative connotation. When Germans watched right-wing extremist videos they
may have been reminded of the Nazi regime. Similarly, Muslims might have
felt negatively for being reminded of Islamic extremism. Their first reaction
thus might be a spontaneous distancing from the presented message in order to
defend themselves by denying the ingroup relation. This is in line with former re-
search on ingroup defense reactions. Belonging to a negative ingroup triggered
distancing from this group (Arndt et al., 2002; Tajfel & Turner, 1986).
This, however, does not explain why pupils from vocational schools did not show
such a defense reaction. On the one hand, it might be that students were more
aware of the own cultural guilt (Gunn & Wilson, 2011; Wohl, Branscombe, &
Klar, 2006). Thus social desirability might have prevented them from overt ex-
pression of approaching tendencies. Study III (see Chapter 11) addressed this
idea.
On the other hand, it might be a more common conventionalism to understand
other perspectives and to criticize closer shortcomings (such as the extremist
black-and-white worldview) within the student samples. Pupils from vocational
schools might not be that trained in rejecting or questioning such statements.
Against this background we further analyzed if the studied major influenced
how students reacted to propaganda of the ingroup. There are first hints (Gam-
betta & Hertog, 2007) that majors with a clear determined and structured focus
(technical or scientific-oriented) might lead to more interest in an extremist ideol-
ogy or reverse: Extremist people might be more prone to study technically and
scientifically-oriented majors. However, the inclusion of major field of study
did not change the pattern in the student sample. The rejection of ingroup propa-
ganda was similarly found among students with a humanistic as well as a technical
major.
This is in line with Gambetta and Hertog (2007) who reported a higher number of
engineers becoming Islamic extremists in countries with strict Muslim regimes
(Arab Peninsula) but not among Westerners. Potential explanations will be dis-
cussed in detail in Chapter 12.
Last but not least, it might be possible that although pupils in general reacted more
receptively to propaganda of the own ingroup, this would – in a second step – not
influence their evaluation of terrorism in general. We thus applied the self-devel-
oped justifications of the terrorism scale (see Chapter 10.2) in Study II not only as
87
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 88
a control variable before reception, but also after each series. By repeating the as-
sessment of justifications of terrorism, changes due to the videos could be dis-
played. Arguably, people might report terrorist actions to be more justifiable if
they were presented with arguments in favor of those actions (e.g., as in extremist
videos clips) before. To depict this short-term effect, the next analysis aimed at
investigating if the justifications of terrorism changed after being exposed to
either ideology. Notably, terrorism acceptance after the reception was only ac-
cessed in Study II. As a result, for the student participants, cultural background
could not be analyzed as nearly all of them were German. A 3 (Culture: German
vs. Muslim vs. Other) × 3 (Terror justification: before vs. after the right-wing vs.
after the Islamic extremist clips) mixed ANOVA36 with the last one as within fac-
tor was conducted only for the pupils from vocational schools.
The results indicated that German pupils did not change their justifications of ter-
rorism depending on the extremist ideology of the presented videos. Other mi-
grants showed increased justifications of terrorism after the series of right-wing
but not after the Islamic extremist videos. Muslim pupils reported more justifica-
tions of terrorism after the Islamic extremist videos but not significantly after the
right-wing series (see Figure 14).
36 Mixed ANOVAs include between (here: culture) and within (here: terrorism justification) factors.
Thus besides the main effects of each of the factors also interactions between them were possible.
88
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 89
Figure 14. Results for the justifications of terrorism among pupils split by cultural background and
ideology. Participants stated their agreement on 7-point Likert scales ranging from 1 (=absolutely not
agree)to7(=absolutely agree).
In general, the results so far indicated that propaganda evoked mainly negative
reactions. The strongest demographic factor influencing the shade of rejection
was found to be the cultural background. For all dependent measures, cultural
background shaped the evaluations. The results can be interpreted in terms of a
defense reaction against propaganda of the own cultural ingroup among students.
However, besides for Shame and Aversion which only depended on cultural back-
ground, the effect turned for pupils of vocational schools. Overall, they evaluated
propaganda of the own ingroup less negatively and more interest-evoking.
Nonetheless, Decker and his colleagues (Decker et al., 2010) showed that right-
wing attitudes were also found among the higher educated Germans (such as aca-
demics). Further, Sageman (2004) as well as Bakker (2006) reported a substantial
amount of Jihadist having finished college or university. Additionally, the broad
majority of students and pupils from vocational schools of all cultures certainly
reject extremism. Against this background, it seems implausible that cultural
background and educational level alone can explain extremist appeal. As already
discussed in Chapters 1 and 5, personality factors and attitudinal patterns were ex-
pected to contribute to the effects of propaganda.
89
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 90
10.3.2 The Role of Personality and Attitude
Q5) Are personality factors and pre-existing attitudinal patterns able to in-
fluence the evaluation of extremist propaganda?
Besides the influence of culture and educational level on the evaluation of extre-
mist propaganda, we were also interested in the effects personality and attitudinal
patterns may have. Correlational analyses showed that nearly all recipient vari-
ables correlated significantly with cultural background and/or with educational
level.
In order to identify the pure association between the recipient variables and the
evaluation measures, we thus conducted partial correlations37 controlling for
the influence of education and culture (dummy-coded38 with other migrants ser-
ving as the control group) on the recipient variables and the dependent variables.
Table 18 shows the correlation coefficients.
37 Partial correlations describe the association strength between a predictor (here: the recipient vari-
ables) and a criterion variableor outcome (here: the online and post-hoc measures). In contrast to a
normal correlation, the influence of other variables is controlled (Field, 2009), (here: culture,
educational level and the respective dependent measure). These correlation coefficients display
the additional association between the recipient variables and the evaluation of the propaganda
clips.
38 Dummy coding is a necessary step when grouping variables with more than two groups shall be
entered into a correlational analysis, as for example cultural background with its three manifesta-
tions (German, Muslim, Other).
90
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 91
Table 18. Partial Correlations for Personality Factors and Attitudinal Patterns
SCR Aversion Shame One-sidedness Interest Persuasiveness
RE RI RE RI RE RI RE RI RE RI RE RI
Cultural Background
Educational Level
Neuroticism .14 .13 –.05 .01 –.07 .02 –.05 .03 –.08 .04 –.02 –.15
Extraversion –.01 .04 .06 –.02 .02 –.02 –.01 .12 .12 .12 .14 .02
Conscientiousness –.08 –.01 –.08 .00 –.05 .01 .05 .00 –.16 –.11 –.13 –.09
Openness –.12 –.09 –.03 –.08 –.07 –.09 .07 .06 –.03 .07 –.17 –.04
Agreeableness –.20*–.25** .13 .11 .06 .00 –.11 –.25** .00 .10 –.09 –.08
German Identity .04 –.07 .03 .13*.09 .01 –.08 .06 .13*.00 .10 –.07
Non-German Identity .03 .30** –.03 –.16 –.07 .01 –.24*–.31** –.05 .10 .04 .01
Authoritarianism .38** .14*–.05 .05 .01 .08 –.35** –.25** .20** .04 .16** –.13*
Political Attitude .22** .08 –.10 –.07 –.14*–.06 –.11*.02 .10 –.11 .21** –.05
Acceptance of Violence .21** .15*.02 .12*–.04 .05 –.09 .07 .19** .09 .31** .13*
Feeling of Control –.09 –.03 .02 –.02 .01 –.06 .31** .31** –.09 –.11 .04 .06
Sensation Seeking .05 –.02 .09 .12*–.06 .01 .11 .14*.10 .10 .11 .05
Note. **p< .01 *p< .05 Significant correlations are marked in boldface. Correlations r> .10 can be regarded as low, r> .30 as medium, r> .50 as high.
91
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 92
For all dependent measures at least small associations were found. Based on these
partial correlations, the most promising predictor variables were chosen. We
aimed at identifying the most important predictors for the evaluation of propa-
ganda in terms of two main aspects: (1) Correlation coefficients which are con-
sidered as at least moderate (r > .30, see Field, 2009) and (2) variables which
affected a substantial part of the applied measures (at least 1/3 of our dependent
measures). Based on these two criteria, authoritarianism and acceptance of vio-
lence were considered the most promising predictors when trying to explain the
evaluations of online propaganda.
For deeper insights into the specific contribution of these variables we conducted
multivariate regression39 analyses for the online as well as post-hoc measures.
Since only authoritarianism and acceptance of violence emerged as promising ex-
planations, the other recipient variables were no longer included40.
Based on recommendations by Field (2009), a forced entry hierarchical approach
was chosen. Step one41 encompassed the demographic variables describing the
samples (culture and education, dummy-coded).
Since the mixed ANOVAs described in Chapter 10.2 showed an influence parti-
cularly for the interaction of cultural background and educational level, this inter-
action was entered in step 2 (dummy-coded). Authoritarianism was entered in step
three, acceptance of violence in step four. As the influence of culture and educa-
tion has already been described, this chapter will focus on the additional value of
authoritarianism and acceptance of violence.
First of all, it has to be noted that Aversion as well as Shame could not be explained
to a satisfying extent by the variables entered. After having seen either ideology,
neither reporting aversion nor shame depended on the participants’ degree of
authoritarianism or acceptance of violence. This is in line with results found in
Chapter 10.2; in these analyses only Aversion and Shame were not influenced
by the interaction of cultural background and educational level, but by culture
alone. This might be due to the basal nature of aversion and shame. Both can
39 Multiple regression predicts a criterion or dependent variable (here: our online or post-hoc mea-
sures) via multiple predictor variables. A linear relation means that more of a predictor leads to
either more (positive b or b-coefficient) or less (negative bor b-coefficient) manifestations on the
dependent measure. While the b-coefficients are unstandardized (thus not comparable between
different predictors); b-coefficients can be compared. The higher the coefficient is, the stronger
the effect on the criterion.
40 Including all predictors did not significantly increase the variance explained. The model fit was
equally good when all predictors correlating with at least three of the dependent measures were
included. Thus the more economic model with authoritarianism and violence acceptance as ad-
ditional predictors was chosen.
41 Blocks are entered one after another. Thus predictors in the second block are used to explain the
amount of variance in the data, not explained by those in the previous blocks. To control for the
effects of culture and education (see Chapter 10.2) they were thus included in block one.
92
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 93
be regarded as universal basic emotions (Zillmann, 2004) and might thus repre-
sent the most basal reaction when being confronted with propaganda. It is there-
fore not surprising that they are influenced more by characteristics of the stimuli
(see Chapter 9.4.3) than by specific characteristics of the recipient. Table 19
displays the results of hierarchical regressions for Aversion and Table 20 for
Shame.
93
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 94
Table 19. Hierarchical Regressions for both Ideologies on the Reported Aversion
Step 1 Step 2 Step 3 Step 4
Aversion b SE bbSE bbSE bbSE b
Right wing extremist
Student vs. Pupil –.04 .09 –.02 –.05 .17 –.03 –.08 .18 –.05 –.07 .18 –.04
German vs. Other .14 .11 .09 .10 .17 .06 .10 .17 .06 .11 .17 .07
Muslim vs. Other –.03 .15 –.01 .05 .21 .02 .06 .21 .03 .06 .21 .03
German Student .08 .22 .04 .06 .22 .03 .05 .22 .03
Muslim Student –.17 .30 –.05 –.14 .30 –.04 –.14 .30 –.04
Authoritarianism –.01 .01 –.05 –.01 .01 –.05
Acceptance of Violence .04 .08 .03
Model-fit R2= .01 DR2= .00 DR2= .00 DR2= .00
Islamic extremist
Student vs. Pupil –.19 .11 –.10 –.24 .19 –.13 –.19 .20 –.10 –.13 .20 –.07
German vs. Other –.07 .12 –.04 –.13 .19 –.07 –.13 .19 –.07 –.10 .19 –.05
Muslim vs. Other .05 .17 .02 .08 .24 .03 .06 .24 .03 .08 .24 .03
German Student .11 .25 .05 .13 .25 .06 .09 .25 .04
Muslim Student –.06 .34 –.02 –.12 .35 –.03 –.11 .35 –.03
Authoritarianism .01 .01 .07 .01 .01 .05
Acceptance of Violence .17 .09 .11
Model-fit R2= .01 DR2= .00 DR2= .00 DR2= .01
Note.R2= The amount of total variance explained by the model with a possible maximum of R2= 1. DR2indicates the changes for the subsequent step.
94
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 95
Table 20. Hierarchical Regressions for both Ideologies on the Reported Shame
Step 1 Step 2 Step 3 Step 4
Shame bSE bbSE bbSE bbSE b
Right wing extremist
Student vs. Pupil .00 .06 .00 –.09 .11 –.08 –.07 .11 –.07 –.09 .12 –.08
German vs. Other .10 .07 .09 .00 .11 .00 .00 .11 .00 –.01 .11 –.01
Muslim vs. Other .25 .09 .17* .24 .14 .16 .24 .14 .16 .23 .14 .16
German Student .17 .14 .15 .18 .14 .15 .19 .14 .16
Muslim Student –.02 .19 –.01 –.03 .20 –.01 –.03 .20 –.01
Authoritarianism .00 .01 .03 .00 .01 .04
Acceptance of Violence –.05 .05 –.06
Model-fit R2= .02 DR2= .01 DR2= .00 DR2= .00
Islamic extremist
Student vs. Pupil .00 .06 .00 –.10 .12 –.08 –.05 .12 –.04 –.04 .12 –.03
German vs. Other .05 .07 .04 –.05 .11 –.04 –.05 .11 –.04 –.05 .11 –.04
Muslim vs. Other .68 .10 .41** .63 .14 .38** .62 .14 .37** .62 .14 .37**
German Student .17 .15 .13 .19 .15 .15 .19 .15 .14
Muslim Student .09 .20 .04 .04 .21 .02 .04 .21 .02
Authoritarianism .01 .01 .10 .01 .01 .10
Acceptance of Violence .03 .05 .03
Model-fit R2= .15** DR2= .00 DR2= .01 DR2= .00
Note.R2= The amount of total variance explained by the model with a possiblemaximum of R2=1.*p<.05, **p< .01. DR2indicates the changes for the
subsequent step.
95
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 96
For all other dependent measures however, the pattern was more interesting. Con-
cerning the rather adverse measures, authoritarianism in particular emerged as an
important predictor.
For the slider ratings during the reception, the inclusion of authoritarianism sig-
nificantly increased the variance explained for both right-wing and Islamic extre-
mist material. In both cases, higher levels of authoritarianism predicted less
reported unpleasantness during the reception. Further including acceptance of
violence slightly reduced the effect of authoritarianism indicating a partial mod-
eration of the effects of authoritarianism via acceptance of violence. Nonetheless,
acceptance of violence only significantly contributed to the prediction of the sli-
der ratings for the right-wing extremist material. As expected, higher levels of ac-
ceptance of violence predicted lower Unpleasantness (SCR) ratings. Table 21 dis-
plays the results for the regression analysis on the slider ratings.
Concerning the post-hoc reported One-sidedness, again authoritarianism signifi-
cantly increased the prediction for the Islamic as well as for the right-wing extre-
mist material. In both cases, higher levels of authoritarianism predicted less one-
sidedness reported after the reception. Table 22 displays the results. Acceptance
of violence did not contribute significantly to this pattern.
96
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 97
Table 21. Hierarchical Regressions for both Ideologies on the Slider-ratings (SCR)
Step 1 Step 2 Step 3 Step 4
Slider ratings (SCR) bSE bbSE bbSE bbSE b
Right wing extremist
Student vs. Pupil .01 .14 .01 .22 .24 .11 .50 .23 .24* .60 .23 .29*
German vs. Other –.42 .15 –.21* –.11 .23 –.05 –.15 .22 –.08 –.09 .22 –.05
Muslim vs. Other .16 .21 .06 .01 .29 .00 –.07 .28 –.02 –.04 .27 –.02
German Student –.53 .31 –.24 –.35 .29 –.16 –.43 .29 –.19
Muslim Student .37 .41 .10 .06 .40 .02 .08 .39 .02
Authoritarianism .05 .01 .38** .05 .01 .34**
Acceptance of Violence .31 .11 .19**
Model-fit R2= .06** DR2= .03* DR2= .10** DR2= .03**
Islamic extremist
Student vs. Pupil .00 .12 .00 .02 .22 .01 .14 .22 .08 .21 .22 .11
German vs. Other .06 .14 .03 .06 .21 .03 .04 .21 .02 .08 .21 .04
Muslim vs. Other .66 .19 .26** .74 .27 .29** .70 .27 .28** .72 .27 .28**
German Student .01 .28 .00 .07 .28 .03 .02 .28 .01
Muslim Student –.16 .38 –.05 –.30 .39 –.09 –.30 .38 –.08
Authoritarianism .02 .01 .18* .02 .01 .15*
Acceptance of Violence .19 .10 .13
Model-fit R2= .06** DR2= .00 DR2= .02* DR2= .01
Note.R2= The amount of total variance explained by the model with a possible maximum of R2=1.*p< .05, **p< .01 DR2indicates the changes for the
subsequent step.
97
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 98
Table 22. Hierarchical Regressions for both Ideologies on the Reported One-sidedness
Step 1 Step 2 Step 3 Step 4
One-sidedness bSE bbSE bbSE bbSE b
Right wing extremist
Student vs. Pupil –.04 .07 –.04 –.16 .12 –.13 –.32 .12 –.26** –.34 .12 –.28**
German vs. Other .19 .07 .16* –.01 .12 .00 .01 .11 .01 .00 .11 .00
Muslim vs. Other –.36 .10 –.22** –.22 .15 –.13 –.17 .14 –.10 –.17 .14 –.10
German Student .34 .15 .25* .25 .15 .19 .27 .15 .20
Muslim Student –.34 .21 –.14 –.16 .20 –.07 –.16 .20 –.07
Authoritarianism –.03 .01 –.35** –.03 .01 –.34**
Acceptance of Violence –.05 .05 –.05
Model-fit R2= .10** DR2= .04** DR2= .09** DR2= .00
Islamic extremist
Student vs. Pupil –.18 .06 –.17** –.25 .10 –.24* –.36 .10 –.34** –.33 .10 –.31**
German vs. Other .11 .06 .11 .02 .10 .02 .03 .10 .03 .04 .10 .04
Muslim vs. Other –.38 .09 –.26** –.36 .13 –.25** –.33 .12 –.23** –.32 .12 –.22**
German Student .16 .13 .14 .11 .13 .09 .09 .13 .08
Muslim Student –.06 .18 –.03 .06 .18 .03 .06 .18 .03
Authoritarianism –.02 .00 –.26** –.02 .00 –.28**
Acceptance of Violence .09 .05 .10
Model-fit R2= .12** DR2= .01 DR2= .05** DR2= .01
Note.R2= The amount of total variance explained by the model with a possible maximum of R2= 1.* p<.05,**p< .01 DR2indicates the changes for the
subsequent step.
98
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 99
Regarding the rather positive post-hoc factors besides authoritarianism, particu-
larly acceptance of violence played a role, but was more pronounced for the right-
wing extremist material.
The post-hoc Interest ratings could only partially be explained (see Table 23). In-
terest in the Islamic extremist material was predicted by authoritarianism, where a
higher level of authoritarianism indicated a higher repor ted interest. However, this
significant prediction diminished when acceptance of violence was included.
Both, authoritarianism and acceptance of violence contributed to the interest
for right-wing extremist propaganda. Again, high levels of both authoritarianism
and acceptance of violence increased interest in the videos.
In total, the variance explained was highest for the Persuasiveness ratings. For
right-wing extremist material, authoritarianism explained a part; higher authori-
tarianism indicating higher ratings on Persuasiveness (see Table 24). Common for
both ideologies however was the contribution of acceptance of violence: Higher
levels of acceptance of violence significantly predicted higher levels of Persua-
siveness.
99
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 100
Table 23. Hierarchical Regressions for both Ideologies on the Reported Interest
Step 1 Step 2 Step 3 Step 4
Interest bSE bbSE bbSE bbSE b
Right wing extremist
Student vs. Pupil –.19 .09 –.12* –.17 .17 –.10 –.03 .17 –.02 .05 .17 .03
German vs. Other .19 .10 .12 .25 .16 .15 .24 .16 .15 .28 .16 .17
Muslim vs. Other .36 .14 .16* .30 .20 .13 .26 .20 .11 .28 .20 .12
German Student –.10 .21 –.06 –.03 .21 –.02 –.08 .21 –.05
Muslim Student .15 .29 .05 –.01 .29 .00 .00 .29 .00
Authoritarianism .03 .01 .23** .02 .01 .19**
Acceptance of Violence .23 .08 .17**
Model-fit R2= .04** DR2= .00 DR2= .04** DR2= .03**
Islamic extremist
Student vs. Pupil –.01 .10 .00 –.22 .16 –.13 –.14 .17 –.08 –.11 .17 –.06
German vs. Other .04 .11 .02 –.34 .16 –.20 –.35 .16 –.20* –.33 .16 –.19*
Muslim vs. Other .59 .15 .25** .89 .20 .38** .86 .20 .37** .87 .20 .37**
German Student .64 .21 .35** .68 .21 .37** .66 .21 .36**
Muslim Student –.69 .29 –.21* –.79 .29 –.24** –.78 .29 –.24**
Authoritarianism .02 .01 .13* .01 .01 .12
Acceptance of Violence .08 .08 .06
Model-fit R2= .06** DR2= .08** DR2= .01* DR2= .00
Note.R2= The amount of total variance explained by the model witha possible maximum of R2=1.*p<.05, **p< .01 DR2indicates the changes for the
subsequent step.
100
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 101
Table 24. Hierarchical Regressions for both Ideologies on the Reported Persuasiveness
Step 1 Step 2 Step 3 Step 4
Persuasiveness bSE bbSE bbSE bbSE b
Right wing extremist
Student vs. Pupil –.56 .05 –.52** –.54 .09 –.50** –.48 .09 –.45** –.40 .09 –.37**
German vs. Other .00 .06 .00 .05 .09 .05 .04 .09 .04 .09 .09 .08
Muslim vs. Other .15 .08 .10 .09 .11 .06 .07 .11 .05 .09 .11 .06
German Student –.09 .12 –.07 –.06 .12 –.05 –.11 .11 –.10
Muslim Student .14 .16 .07 .07 .16 .03 .07 .16 .04
Authoritarianism .01 .00 .15* .01 .00 .10
Acceptance of Violence .23 .04 .26**
Model-fit R2= .29*** DR2= .00 DR2= .02* DR2= .06**
Islamic extremist
Student vs. Pupil –.46 .05 –.45** –.57 .08 –.57** –.60 .09 –.60** –.56 .09 –.56**
German vs. Other –.09 .05 –.09 –.28 .08 –.28** –.28 .08 –.28** –.26 .08 –.26**
Muslim vs. Other .26 .08 .19** .40 .10 .29** .41 .10 .30** .42 .10 .31**
German Student .33 .11 .30** .31 .11 .29** .29 .11 .27**
Muslim Student –.34 .15 –.17* –.31 .15 –.16* –.31 .15 –.16*
Authoritarianism .00 .00 –.07 –.01 .00 –.09
Acceptance of Violence .09 .04 .12*
Model-fit R2= .28** DR2= .06** DR2= .00 DR2= .01*
Note.R2= The amount of total variance explained by the model with a possible maximum of R2=1.*p<.05, **p< .01 DR2indicates the changes for the
subsequent step.
101
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 102
10.4 Summary: Research Focus II
R2: Analysis of the immediate explicit effects of right-wing vs. Islamic extremist
propaganda on different audiences.
Chapter 10 aimed at addressing the research questions 3, 4 and 5 (see Chapter 7)
formulated for Research Focus II. In the first step, differences between the evalua-
tion of right-wing and Islamic extremist material, depending on demographic dif-
ferences between the participants, were addressed.
Q3) Do recipients react differently to right-wing than to Islamic extremist
propaganda?
Q4) Does this reaction depend on the cultural background or other demo-
graphics of the participant?
Results showed that participants indeed reacted differently to right-wing and Is-
lamic extremist propaganda. Further, responses were moderated by cultural back-
ground. Overall the Islamic extremist material was evaluated as more aversive and
shame evoking by Muslim participants. Contrastingly, Germans reported more
shame for the right-wing extremist material. Participants without shared cultural
background with the right-wing or Islamic extremist clips did not differentiate be-
tween the ideologies.
Besides cultural background, the educational level of the participants also played
a role. While German and Muslim students tended to evaluate propaganda of their
own ingroup rather negatively (one-sided, unpleasant, not interesting or persua-
sive), pupils showed the reverse pattern. They evaluated propaganda of their
own cultural ingroup less negatively.
In the second step (as depicted in Research Question 5), we were also interested in
the role of personality and attitudinal patterns in the effects of extremist propa-
ganda.
Q5: Are personality factors and pre-existing attitudinal patterns able to in-
fluence the evaluation of extremist propaganda?
Results showed that while general personality factors did not explain the evalua-
tions of right-wing and Islamic extremist propaganda, attitudinal patterns did.
However, from the assessed attitudinal patterns only authoritarism and violence
acceptance emerged as significant predictors beyond cultural background and
educational level. Higher authoritarism and acceptance of violence predicted
less negative and more positive evaluations of propaganda videos.
In conclusion, Aversion and Shame were overall found to be basic factors which
were evoked independently of personality factors or attitudinal patterns. They
were only influenced by the cultural background of the participants.
102
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 103
The other factors (SCR, One-sidedness, Interest, and Persuasiveness) could in
part be explained by the chosen predictors. Besides cultural background and edu-
cational level (see Chapter 10.2), authoritarianism and acceptance of violence
contributed significantly to an understanding of the evaluations of propaganda.
For the Unpleasantness during reception and post-hoc reported one-sidedness and
interest, authoritarianism emerged as a significant predictor. High levels of
authoritarianism led to lower experience of unpleasantness during the reception
and to less perceived one-sidedness and more interest in the propaganda material.
Further, for the right-wing extremist videos, high levels of acceptance of violence
reduced the experienced unpleasantness of the videos seen but increased the re-
ported interest. Persuasiveness was found to be predicted by acceptance of vio-
lence, where higher levels lead to more perceived persuasiveness of the propa-
ganda material.
The importance of the acceptance of violence component in understanding the at-
traction of right-wing extremism could be confirmed. Since the Islamic extremist
material overall consisted of both more violent language and cues, it seems strik-
ing that the explanative power was lower here. However, it has to be regarded that
overall acceptance of violence in our sample was low. Thus it might be that the
more extreme forms of violence depicted in the Islamic extremist clips exceeded
the participants’ level of violence tolerance. The lower predictive value accep-
tance of violence had for persuasiveness of the accordant material supports this
assumption.
In general, it has to be noted that the total amount of variance explained remained
rather small. Only for Persuasiveness more than 30% of variance could be ex-
plained42. In all other cases the included predictors could explain less than a quar-
ter of the total variance. Nonetheless, as a conclusion for Research Focus II,
including characteristics of the recipients significantly contributed to an under-
standing of the immediate effects of right-wing and Islamic extremist propa-
ganda.
42 The total variance explained can be depicted from the summation of all R2in the preceding tables.
103
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 104
11. Research Focus III: Implicit Associations and Propaganda
The preceding analyses indicated a rejection of extremist propaganda particularly
from the own cultural ingroup, especially among German students (see Chapter
10.2). However, as already outlined, this might be due to an increased awareness
of the social undesirability of approaching tendencies towards right-wing extre-
mist propaganda. Study III tested this idea by focusing on a sample of German
students and using implicit measures in order to avoid social desirable answers.
Implicit measures display the automatic, non-controlled component of sponta-
neous associations and evaluations. In contrast to explicit evaluations (e.g., the
(dis-) agreement to items in a questionnaire), they precede conscious awareness
(Schnabel, Banse, & Asendorpf, 2006). Particularly in the case of undesirable so-
cial attitudes, such as prejudice or extremism, implicit and explicit evaluations
might differ substantially (Greenwald, McGhee, & Schwartz, 1998) and should
be analyzed separately (Goodall, 2011). Even though in the current studies parti-
cipants were seated separately from each other during Study I and Study II, they
might still have hesitated to express overt agreement to propaganda. Therefore,
Study III focuses on the implicit associations of terrorism.
Implicit measures are able to identify subtle changes before explicit measures can
detect them (Goodall, 2011; Hefner, Rothmund, Klimmt, & Gollwitzer, 2011).
They are thus able to depict even unconscious short-term effects induced by single
video clips one might stumble over on the internet. Even though the process from
such spontaneous reactions to explicit agreement (far more actual behavior)
might be substantial, implicit measures can depict first unfiltered propaganda ef-
fects.
One widely used and established procedure among implicit measures is the ana-
lysis of associations via the recording of response latencies. The procedure is
based on the concept of human memory as a network (so called semantic or asso-
ciative network, see Collins & Loftus, 1975). This network consists of nodes and
links in between them.
Nodes can represent attributions, persons, groups or concepts (Greenwald et al.,
2002) and are created with every new concept a person acquires (e.g. terrorism is
a concept). Links between nodes are created when two concepts occur either si-
multaneously (Arnold, Eysenck, & Meili, 1997, pp. 158–162) or when they are
familiar (Greenwald et al., 2002), respectively when theyevoke common thoughts
(Wyer& Srull, 1994). The strength of these links can be understood as the poten-
tial of one node to activate adjacent others (spreading activation, Collins & Lof-
tus, 1975). Generally, nodes can be activated internally via links or through exter-
nal stimuli.
104
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 105
Besides concepts, valence is also stored in the form of a node. The more positive a
concept is evaluated, the stronger it is linked to the positive valence node. Simi-
larly, negatively evaluated concepts are linked to the negative valence node.
Therefore, the activation of a concept also activates its implicit valence. Figure
15 gives an example of how the associated network for terrorism as concepts
could look like.
Figure 15. An example of a possible associative network for terrorism-related concepts.
The implicit association test (IAT) uses this mechanism to measure valence asso-
ciations of concepts (e.g., Greenwald & Banaji, 1995). In a classical IAT, partici-
pants are confronted with four categories of stimuli, two representing the target
categories (e.g., insects and flowers as a classical example, see Greenwald et
al., 1998) and two representing the valence (positive and negative items).
Participants are asked to sort these categories by pressing certain keys (e.g., by
pressing an “i” for each flower and an “e” for each insect). The time span between
appearance and pressing the key (the response latencies) is measured.
During the target trials, two concepts always share the same key. First, flowers and
positive stimuli are both answered by pressing “e” and the other two by pressing
the “i” (congruent condition). In the second phase the combination is reversed
(incongruent condition).
The idea is that if the combination is congruent, pressing the key will be facilitated
and participants will respond more quickly (the response latencies will be lower).
105
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 106
Thus, comparing the response latencies for both target trials describes which va-
lence is experienced as more congruent.
Generally, the association strength between e.g., valence and a concept can be in-
fluenced also by preceding stimuli, so-called primes. Research has shown that re-
sponding to a category is faster when a prime that shares the same valence has
been presented before (Fazio, 2001; Fazio, Jackson, Dunton, & Williams, 1995;
Olson & Fazio, 2003). For example, people would react faster to negative asso-
ciated concepts if beforehand another negative stimulus (e.g., an insect in our ex-
ample) was presented (Hefner et al., 2011).
11.1 Sample, Procedure and Material Study III
A total of N43 = 111 male students participated in this study. Due to the language-
based methodology, only native Germans were included in the analyses. Based on
this, n= 32 participants had to be excluded. Random sampling left 14 participants
for each condition44. On average, the resulting N= 70 participants were
M= 24.31 (SD = 3.26) years old. All of them were studying at the moment of
data collection.
In the experiment, participants arrived in the laboratory and were then informed
about the procedure of the study and their rights (see also Chapter 9.3). Further,
they were offered the option to talk to a psychologist afterwards for debriefing,
further questions, or clarification. The experiment started by randomly assigning
participants to one of the experimental conditions in which they had to watch one
extremist video (either right-wing or Islamic extremist)45 or to a control condition
without video. Subsequent to the video exposure, they were asked to perform a so-
called single-category implicit association test (SC-IAT).
The procedure of the SC-IAT used in this study resembles the classic IAT de-
scribed in Chapter 11.1, except that only one concept is analyzed for its valence
association (Goodall, 2011; Karpinski & Steinman, 2006). The SC-IAT displays
differences in concept evaluations following media reception that are not detected
by explicit measures (Hefner et al., 2011). We used the SC-IAT to detect whether
priming via extremist propaganda could influence implicit associations of “ter-
rorism”.
43 N= Number of participants, M= Mean, SD = Standard Deviation.
44 Equal cell distribution had to be realized due to violations of the assumptions in ANOVA (Hussy &
Jain, 2002)
45 Participants in the control condition answered the SC-IAT without watching an extremist video
clip.
106
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 107
Four videos were selected as primes based on the results of Study I (see Chapter
9.4.2). As cited before, “Propaganda which speaks the language of propaganda
fails to penetrate cognitive defences. The best propaganda is disguised as enter-
tainment” (O’Shaughnessy, 2012, p. 34).
Thus, propaganda which is too conspicuous in revealing its real purpose might fail
to affect implicit associations. As a conclusion, those videos scoring highest on
one-sidedness vs. those lowest on one-sidedness and highest on persuasiveness
were selected (see Table 25).
Table 25. Selection of the most One-sided and Persuasive Videos for Study III Split by Ideology
Ideology Highest one-sided video Highest persuasive video
Right-wing extremist TH Activist Extreme Video
Islamic extremist RC News broadcast TH Activist
To represent valence, ten positive and ten negative adjectives (pretested for fre-
quency and valence (Glock, 2010) were chosen. Additionally, words associated
with terrorism (such as bomb) were selected. All items were pretested by
N= 25 participants on a 7-point Likert-scale, ranging from 1 (= Not at all applying
to terrorism)to7(=Totally applying to terrorism). Refer to Appendix 4 for a list
of words selected. For both target trials, mean response latencies for words asso-
ciated with terrorism were calculated. To control for outliers, response latencies z
> +/–3 (Field, 2009)46 were excluded.
Two valuesresulted for the data analysis: terrorism-positive and terrorism-negative
response latencies. To describe the general valence of terrorism association, a dif-
ference of these twovalues was computed. The differencescore was created by sub-
tracting negative-terrorism response latencies from positive ones. The positive va-
lues of this difference score expressed a more negative association (people reacted
faster when terrorism and negative words shared a key) and the opposite occurred
with the negative values. After having completed the SC-IAT, participants filled
out a questionnaire about their demographics and were then thanked and rewarded.
46 Z-transformation standardizes the raw values (resulting M= 0, SD = 1). Z-values of +/– 3 thus have
a probability of less than 0.1 % to belong to the same distribution of values,for example, because
they resulted from the inattention of the participant.
107
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 108
11.2 Results for Implicit Effects
To measure the implicit association of terrorism and the impact of propaganda vi-
deos, a univariate ANOVA47 (Video prime: no-video vs. one-sided vs. persuasive
right-wing vs. one-sided vs. persuasive Islamic extremist video) was conducted
using the computed difference score as a dependent measure in order to identify
differences according to the presented video clip.
The main effect for video prime became only marginally48 significant,
F(4, 65) = 2.18, p= .08, hp2= .1249. The overall positive difference scores indi-
cated a negative association of terrorism in all conditions. However, the control
condition (without video prime) had a difference score close to zero, indicating
a rather neutral association of terrorism. Consequently, differences were found
between the control condition and nearly all video conditions (all p< .04).
Only participants who watched the persuasive right-wing extremist video did not
differ from the control group concerning their difference score (p= .11). Similar
to the condition without video prime, no pronounced negative association of ter-
rorism was given after having watched the persuasive right-wing extremist video.
To further investigate if the distinct associations of terrorism after the videos were
due to faster response latencies on the positive key share or due to slower response
latencies on the negative key share, a 5 (video prime: no video vs. one-sided vs.
persuasive right-wing vs. one-sided vs. persuasive Islamic extremist video) × 2
(valence: negative key share vs. positive key share) mixed ANOVA for the terror-
ism-positive and terrorism-negative response latencies was conducted.
The results showed a marginally significant interaction between video prime and
valence, F(4, 65) = 2.18, p= 08, hp2= .19. As already found for the explicit jus-
tifications of terrorism in Study I and Study II, the results overall indicated a ne-
gative association of terrorism. Terrorism-related words were categorized signif-
icantly faster when they shared one key with the negativewords. Figure 16 shows
the response latencies split by video seen.
47 As all participants saw only one video and difference scores were analysed between participants,
differences found are due to the different video clips.
48 The result is called marginally significant because the correspondent pvalue is between .05 and
.10, thus, in this case, the result is random with a probability of 8 %.
49 Fis the statistical parameter in ANOVAs measuring the ratio of the variance explained vs. the
unexplainable variance. As a rule of thumb, higher Fs indicate stronger effects. Df = degrees of
freedom. p= probability of the found differences to be random vs. systematic. hp2= (partial) effect
size.
108
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 109
Figure 16. Response latencies for the terrorism-related words when sharing a key with either the po-
sitive or the negative words.
However, a closer look at the distinct conditions revealed interesting differences.
In the control condition (thus without preceding video prime) the association of
terrorism was nearly neutral (difference close to zero), meaning that the negative
association of terrorism was weaker here. In all other conditions differences be-
came significant. Both analyses together clearly indicate that video primes led
to a more negative association with terrorism. This suggests that implicit propa-
ganda videos rather exacerbate negative evaluations of terrorism. However, the
effect was the smallest for the persuasive right-wing extremist clip.
11.3 Summary: Research Focus III
R3: Analysis of the implicit effects of selected propaganda videos.
Q 6) Do different propaganda videos have the potential to shape the impli-
cit associations of extremism in its most extreme form?
Concerning our sixth research question, we found that propaganda videos in gen-
eral influenced the implicit associations of terrorism. Terrorism was associated
more negatively after having been confronted with an extremist video clip. How-
ever, this effect was not found for the persuasive ingroup propaganda indicating
differences depending on style of the message and ideological background.
This could be interpreted in a way that right-wing extremist propaganda (if made
persuasively) would not activate the negative associative network more than it is
usually already activated. It is no surprise that terrorism is associated negatively
but this finding sheds first light on the role persuasive propaganda could play. It
has to be noted, however, that the result was found in a German student sample. It
is still questionable whether a broader sample would lead to the same findings.
109
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 110
Especially, the role of cultural background should be investigated in greater detail.
Our previous results suggested an explicit defense reaction in the student subsam-
ple. Implicitly, this defense reaction was not found for the persuasive ingroup pro-
paganda. More precisely, the right-wing extremist Extreme clip addressing pedo-
philia and child victims did not increase the negative association of terrorism. In
contrast, the TH Lifestyle activist directly addressing the recipient as well as the
Islamic extremist RC News broadcast all led to more negative associations of ter-
rorism.
This is in line with O’Shaughnessy’s (2012) assumption that propaganda dis-
guised as entertainment is severely rejected. While the Extreme clip focused on
the victims of pedophilia (the organization behind it is only presented in the
last few seconds), the other three videos portray the transported ideology more ob-
viously.
In both Islamic extremist clips, concrete references to Islam and Jihad are given.
The right-wing extremist TH Lifestyle activist video begins with the sentence
“German open your eyes, you are in war”.
Furthermore, as already discussed in Chapter 9.4.2, videos displaying children
suffering (the two Movie clips and the right-wing extremist Extreme clip) overall
reached rather low ratings on the adverse and high ratings on the approaching
post-hoc factors analyzed.
The results are thus interesting in two main aspects: (1) the wolf-in-sheep’s clothes
strategy used by right-wing extremists (see e. g. Chapter 3.2) generally seems to be
more dangerous than propaganda overtly expressing extremist thoughts. (2)
While German students seemed rather unsusceptible to extremist propaganda
when explicit measures were used, implicit measures found a less pronounced de-
fense reaction. The following chapter will conclusively summarize our findings
and discuss their implications.
110
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 111
12. Discussion
We introduced the book with a quote by Paul Watzlawick, stating that the “secret
of propaganda” lies in its potential to “totally saturate” the desired receiver
“without him even noticing”. Security agencies and researchers have recently
shown concern that modern propaganda videos are able to “saturate” even, so
far, non-obtrusive recipients and, “without them even noticing”, to contribute
to a distributed indoctrination or even radicalization. Particularly, the amount
of Islamic and right-wing extremist propaganda on the Internet reached an alarm-
ing potential in recent years (Europol, 2011; Precht, 2007). Strategists from both
ideologies also claimed internet propaganda as a powerful tool in their “war over
minds” (Weimann, 2006, p. 118).
However, as depicted in the famous AIDA formula, propaganda as a “product sell-
ing strategy” first has to get attention, raise interest and create desire before action
might follow (Lewis, 1903). So far, little research has tested whether a distributed
indoctrination is possible or whether the preceding steps (attention and interest)
can be accomplished via propaganda.
The large-scale study presented shed some light on this question and the related
assumptions. By referring to Lasswell’s (1948) formula of mass communication,
we addressed the question “Who spreads extremist propaganda to whom, in what
format and with what effect?” The answer can be summarized as: Right-wing and
Islamic extremists disseminate audio-visual propaganda with varying formats
over the internet. As discussed in Chapter 5, they particularly try to reach recipi-
ents forming their potential ingroup in terms of cultural background. In general,
recipients reacted to propaganda material with rejection. Nonetheless, effects var-
ied with the specific format of the video, the underlying ideology as well as with
factors on the side of the recipient.
In order to describe the formats of the video clips, a categorization system encom-
passing Islamic as well as right-wing extremist video clips was developed via Eth-
nographic Content Analysis (ECA). To our knowledge, this approach can be
considered as unique since it is the first attempt to adapt two ideologies into
one systematic categorization.
On that point, we found four prototypical formats of right-wing and Islamic extre-
mist propaganda in Germany (Q1, see Chapter 7),similar to those described by
Baines et al.(2010) (see Chapter 9.1). The same formats and styles could be dis-
tinguished for both ideologies (Q1, see Chapter7).
Videos featuring individual “Talking heads”, journalistic “Reality clips”, pro-
duced “Movie” and “Extreme clips” were found for both. Talking heads as
well as Reality clips were further differentiated into three subcategories each ac-
cording to the person or group focused on in the videos. Talking heads covered
111
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 112
videos featuring Ideologues,Lifestyle activists and Martyrs. Reality clips encom-
passed clips displaying the “Group lifestyle”, the “Public opinion” or mass media
styled “News broadcasts”. Differences between the two ideologies were only
found concerning the existence and the amount of these subcategories (e.g., Mar-
tyrs were not identified among the right-wing extremist material).
Overall, both senders tried to convey their messages via the same production
styles. Furthermore, the messages transported shared the same common strategies
(Merten, 2000) and served the same purposes: (1) Legitimization, (2) Propaga-
tion, and (3) Intimidation (Corman et al., 2006). Most of the videos aimed at le-
gitimization and propagation. Furthermore, similar messages were transported
(Q1, see Chapter 7).
Particularly, displaying the own group as innocent victims, and “the unprovoked
aggression of the others” laid in focal attention. Additionally, the depiction of a
“strong community” and the possibility of “own activities” played a major role
in Islamic as well as right-wing extremist propaganda.
Though ideological foundations (TH Ideologues) were more frequent within Isla-
mic extremist propaganda and fancy Movie clips among right-wing extremist vi-
deos, overall the distribution of the categories was similar.
Hence, though they fight on different ends and partly ideologically oppose each
other, a comparison between right-wing and Islamic extremism from a media psy-
chological perspective was found to be possible.
Out of the developed categories, we selected prototypical videos for each ideol-
ogy (see Chapter 9.2) and presented them as stimulus material in three consecu-
tive experiments. Study I and Study II followed the same procedure and focused
on the explicit evaluations of extremist propaganda. Study III additionally as-
sessed the implicit effects.
The explicit effects of propaganda were measured via online reported (Un)plea-
santness as well as immediate post-hoc evaluations. (Un)pleasantness was ac-
cessed via constant slider-ratings (SCR) during video reception. Post-hoc evalua-
tions were measured via questionnaires applied afterwards. Five post-hoc factors
were found to describe the effects of propaganda. Three of them indicated adverse
evaluations (Aversion, Shame and One-sidedness) and two can be interpreted in
terms of a first approach (Interest and Persuasiveness). First impressions made
during the reception of a video indicated tendencies for later evaluations. Com-
bining both online and post-hoc measurements depicted the evaluating process
ranging from immediate to short-term effects.
The analysis of the single video clips showed that distinct production formats and
message styles differed with regard to their potential to create interest or to trig-
ger rejection (Q2, see Chapter 7).
112
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 113
Summarizing, the overall effect found was a rejection of propaganda. In general,
young, normal recipients showed adverse reactions towards the videos. During, as
well as after, the reception they mainly stated that they experienced Unpleasant-
ness, Aversion,Shame and One-sidedness. However, some videos were able to
trigger Interest and Persuasiveness as well. Additionally, one could argue that
low ratings on the adverse factors might as well be interpreted in terms of a first
approach.
In general, the analyses showed that, on a broad level,videos which had an affect-
ing potential could be distinguished from those that had not. In line with
O’Shaughnessy (2012), propaganda disguised as entertainment was less rejected.
For Islamic as well as right-wing extremist videos TH lifestyle activist,Movie and
Extreme clips affected recipients online as well as post-hoc on the adverse as well
as the rather approaching factors.
Particularly, the individualized Talking heads and the norm-violating Extreme
clips were evaluated as being unpleasant during the reception and stimulated aver-
sion and shame afterwards. The Movie clips, with their high professional produc-
tion style, received less pronounced (Un)pleasantness ratings. However, post-
hoc, their victim-focused argumentation was also evaluated as being one-sided
and shame evoking. Strikingly, regardless of these rather adverse evaluations,
the same videos were also the best ones in evoking interest and persuasiveness.
The overall more frequent Reality clips (see Chapter 9.1) could not trigger similar
levels of Interest regardless of their “journalistic perspective”. In contrast to
Baines and his colleagues’ (2010) assumption, the adventurous clips were not bet-
ter in stimulating approaches towards the transported ideology.
It seems remarkable that videos evoked negative and rather positive evaluations at
the same time. However, results showed that the response to propaganda cannot be
conceptualized in simple positive/negative dimensions but has multiple facets.
Videos which were designed in a professional way or which directly addressed
the audience were more likely to affect. Instead of being regarded as boring
they were able to get attention and – as depicted in the AIDA formula (E.S. Lewis,
1903) – got the chance to be processed by the recipient. Furthermore, looking at
the messages and the pictures transported, first hints were found that, in particular,
videos displaying children suffering would touch the recipients. It is not surpris-
ing that these videos triggered aversion ratings as well as reactions of interest.
From a media psychological perspective, preventive approaches should particu-
larly address these “entertaining” videos. Videos rated as boring – even though
transporting ideological foundations or public opinions – do not touch to the
same extent. Concluding, on the level of different videos, more communalities
than differences between Islamic and right-wing extremist propaganda were
found.
113
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 114
However, it has to be noted that our categorization system encompasses broad for-
mats or “genre” distinctions. Further research should deepen this description. For
example, analyses of the way right-wing extremist and Islamic extremist senders
“sell” their ideology (e.g., via narration – or frame analyses) shed light on the de-
tails of the different formats.
As a first conclusion, on the side of the video clips, professionally made videos,
directly addressing the recipient and transporting personalized messages or indi-
vidual destinies were found to develop an affecting potential. Nevertheless, be-
sides all communalities, people are usually attracted (“recruited”) by only one
of them. Thus, in the next step, we focused on ideology as the main underlying
factor.
The results indeed showed that recipients reacted differently to right-wing than to
Islamic extremist propaganda (Q3, see Chapter 7). However, these differences
were shaped by the variance between the recipients. As depicted in research ques-
tion four (see Chapter 7), the reaction depended on the cultural background and
the educational level of the participants (see Chapter 10.2).While Germans can
be regarded as the potential ingroup addressed by right-wing extremist propa-
ganda, the same is true for Muslims and Islamic extremist audio-visuals. Still talk-
ing about the general rejection of propaganda, this ingroup relation was identif ied
as shaping the amount of rejection. The effect further varied with the educational
level of the participants.
German students rated right-wing extremist videos more negatively and less per-
suasive than Muslims did. Participants with a Muslim background showed the in-
verse pattern, that is, more negative and less positive responses to the Islamic ex-
tremist videos. Participants with another migration background (without ingroup
relation to either ideology) did not show pronounced reactions to either the Isla-
mic or the right-wing extremist material.
This pattern, however, was conversed when analyzing a different educational le-
vel. Muslim as well as German pupils reacted less negatively and more positively
to propaganda of their own cultural ingroup. Subsequent analyses showed an even
higher (though still low!) degree of acceptance of terrorism justifications for
Muslim pupils after watching Islamic extremist material.
Our data also show that the students and pupils did not differ concerning evoked
aversion and shame. In both samples a tendency was found that material of the
own ingroup evoked more aversionand shame. Nonetheless, pupils rated ingroup
propaganda as more interesting and persuasive and students rated it less so. Dif-
ferent explanations for this effect are possible. First of all, students might be more
trained in questioning propagandistic statements and be stricter in the avoidance
of these.
In line with this, the results of Study III implicitly showed a neutral evaluation of
terrorism in a student sample (see Chapter 11.2). Students who had not seen a pro-
114
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 115
pagandistic video (the control group) showed no pronounced association of ter-
rorism with neither positive nor negative concepts. The conducted single-cate-
gory implicit association test, however, revealed that after the reception of propa-
ganda videos terrorism was associated more negatively (Q6, Chapter 7). Never-
theless, in the case of a “wolf-in-sheep’s clothes” strategy, namely a persuasive
video of the own cultural ingroup, the rejection was less pronounced. The amount
of negative association did not differ from participants in the control group. Spec-
ulating, persuasive propaganda might slip through these defensive walls but once
under conscious awareness it is most likely to be rejected as well.
Regarding the higher general rejection of propaganda videos clearly identifiable
as such (see Chapter 9.4.2), it might be that the threshold value for students to re-
act may be lower. This can be explained when considering that people can be in-
oculated against persuasive attempts similarly to biological infections (Mcguire
& Papageorgis, 1961). Former confrontations with a persuasive topic and a strong
“immune system” trigger fast defensive reactions. It has to be remembered that
students reported pre-experience with Islamic extremist material significantly
more often (see Chapter 10.2), so that they may have overall reacted faster.
Two components of the inoculation process can be distinguished: threat and refu-
tation pre-emption (Pfau & Szabo, 2003). Threat has been identified as a more
important factor, motivating the defense reaction. In order for threat to work,
the recipient of the message must be warned that his attitude might be questioned
(Pfau & Szabo, 2003).
In the student sample, threat could have been induced because the evaluation of
propaganda as negative was a more pronounced value for students. Not only pro-
paganda in general but also extremism may have been evaluated as more socially
undesirable (King & Bruner, 2000).
In general, academics are traditionally seen as tolerant, left-wing oriented and
anti-propagandistic. In line with this, the political attitude in our student sample
was rather left skewed; no student stated aright-wing attitude. Thus, it seems plau-
sible that implied values such as rejection of extremist propaganda were held high.
Students may thus have felt greater threat than pupils from vocational schools,
when having been confronted with propagandistic material. Future studies should
therefore assess individual values in order to account for this interpretation.
Threat could further have been induced by being addressed as a potential ingroup
member of an extremist ingroup. The conveyed ideologies are referring to a spe-
cific cultural background which is perceived as either ingroup or outgroup. Ac-
cording to the Social Identity Theory (Tajfel & Turner, 1986), people are moti-
vated to see their ingroups in a positive way. In the case of being confronted
with reminders of collective guilt (Wohl et al., 2006), this positive interpretation
becomes more difficult. As known from literature on stereotype threat (Spencer,
Steele, & Quinn, 1999), people fear to confirm negative stereotypes about their
ingroup. Right-wing extremist Germans and radical Islamists surely represent
115
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 116
the dark side of one’s corresponding cultural ingroup. Therefore, it seems plausi-
ble that in the case of being addressed by an extremist group as a potential recruit,
people try to distance themselves (Arndt et al., 2002). Future research should clar-
ify why this effects is most pronounced for the student sample.
This can also be regarded as a possible starting point for prevention. The most im-
portant aspect for a specific behavior is the underlying intention to behave that
way. However, this intention is not only influenced by one’s attitudes but also
by the expectancies concerning the results of these attitudes and the behavior.
For instance, social norms influence the expectancies concerning a certain goal
(Ajzen & Fishbein, 1970). This implies that the fostering of anti-extremist or
anti-propagandistic norms in non-student samples could hamper the potential
of extremist messages.
Last but not least, former research has discussed professional frustration as poten-
tial influence (Gambetta & Hertog, 2007; Heitmeyer, 1992; Munton et al., 2011).
Since academics are usually highly respected within German society, we cannot
exclude that more positive expectancies for the future might have played a role. In
line with this, a representative survey conducted by Continental found the major-
ity of students to have rather positive expectations concerning their own future50.
Contrastingly, in another survey, more than 50% of pupils from vocational
schools reported “mixed feelings” (Bacher & Prosch, 2002). Thus, future studies
should also include professional satisfaction as a possible influence on the effects
of extremist propaganda. The actual shell study (Albert, Hurrelmann, Quenzel, &
Infratest Sozialforschung, 2010) complements the pattern by pronouncing that,
particularly, the certainty in finding or keeping a job moderates future expectan-
cies.
Besides these demographic variables, personality factors and pre-existing attitu-
dinal patterns were partly able to influence the evaluation of extremist propa-
ganda (Q5, see Chapter 7). However, broad personality factors, such as the Big
Five or sensation seeking, could not contribute to an understanding.
This is in line with other research that could not identify a typical “terrorist per-
sonality” (Hudson et al., 1999; Kemmesies, 2006). Nevertheless, including attitu-
dinal patterns as additional explanations identified authoritarianism and accep-
tance of violence as influential factors. Though additional explained variance
remained rather low, particularly authoritarianism showed a consistent pattern.
Participants, who scored higher on authoritarianism, overall evaluated propa-
ganda more positively than those with lower manifestations. Regarding the con-
ceptualization of authoritarianism as “flight into security” (Oesterreich, 2005),
the idea that uncertainty (e.g., concerning the own future) might play a role in
50 Representative opinion poll by Continental 2006. Retrieved online http://www.wiwi-treff.de/
home/index.php?mainkatid=1&ukatid=1&sid=9&artikelid=2764&pagenr=0 09/02/2012, 15:00
116
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 117
the attraction of extremist propaganda could be supported. Authoritarianism re-
flects the personal search for order and structure by looking to authorities or
powerful groups. It is not surprising that extremists try to address this desire
and, vice versa, recipients respond to this proposing.
Research until now has not sufficiently answered the question about the roots of
authoritarianism and acceptance of violence. Difficult familiar situations
(Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950), social learning ap-
proaches (Altemeyer, 1988) and cultural norms and values (Rubinstein, 1996)
have been discussed. However, modern approaches particularly pronounce the
role of perceived insecurity (Duckitt, 2011) and authoritarianism as motivated at-
tachment to ingroups. In line with this, the identification with powerful others has
been found to increase under uncertainty (Hogg, 2000).
In a similar vein, the acceptance of violence committed by members of the own
group was found to increase when people were faced with existential anxieties
(Pyszczynski et al., 2006). Nonetheless, it was also found that increased self-es-
teem and security might serve as a resilience factor when it comes to such situa-
tions (Harmon-Jones et al., 1997; Rosenberg & Pearlin, 1978). Further research
on potential prevention should thus analyze self-esteem and personal security
as potential resilience factors.
Authoritarianism and acceptance of violence furthermore share the demand for
punishment for those who threaten their social order. Extremist groups offer
this punishment. Diffuse, infrahuman “others” as the “system” or “the Western
world” are scapegoated and threatened by the extremist groups. A supplementary
assessment of demand for punishment could analyze this aspect in more detail.
To sum up, patterns related to security, structure and conservatism were found to
predict the appeal of extremist messages. The explanative power of authoritarian-
ism and acceptance of violence for the case of right-wing extremism could be ex-
tended to Islamic extremism. Surprisingly, the acceptance of violencewas a more
important predictor for right-wing extremist propaganda. We already noted that
our sample overall reported low acceptance of violence. The higher exposure of
violence in the Islamic extremist material would have assumed that acceptance
of violence predicted more variance of the evaluation of this ideology. Neverthe-
less, the data suggests that the question posed by Heitmeyer and his colleagues
(2010) of whether ideology or acceptance of violence precede each other in radi-
calization might also be interesting in the context of Islamic extremism.
Our results showed that the chosen media and social psychological approach
could contribute to a deeper understanding of susceptibility to and effects of ex-
tremist internet propaganda. However, it has to be stated that radicalization is a
multi-factorial process in which media effects might play only a minor role.
The overall small amount of variance explained (see Chapter 10.3.2) underlines
this.
117
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 118
Extremist careers cannot be understood without reflecting the social context and
the communication networks in which individuals are embedded (Kemmesies,
2007; Munton et al., 2011). The phenomenon of extremism has to be understood
as triggered by factors on cultural, societal, social and individualist levels (Kem-
mesies, 2006, p. 17).
Besides this general constraint, the study also had some more specif ic limitations,
restricting the generalizability of our results. These restrictions also might point
towards future research directions.
First of all, the sample analyzed has to be mentioned. As only male participants
took part, potential gender influences could not be addressed. Although male re-
cipients are the primary target group of extremist propaganda, future studies
should also include female samples. Baines and his colleagues could show differ-
ences between female and male reactions towards propaganda (2010). Addition-
ally the focus on Germans with occidental and Muslims with oriental cultural
background allows no inferences on converted Germans. The actual discussion
about salafist extremists in Germany, however, indicates that they might particu-
larly be at risk51.
Furthermore, the focus on students and pupils from vocational schools excluded
participants who had not graduated or were unemployed. The discussed potential
influence of negative life-expectancies makes an extension to such groups desir-
able or even necessary.
Our results suggest that there are general factors inherent in propaganda irrespec-
tive of ideology. These interact with factors on the side of the recipient and might
increase interest in propaganda material. It seems interesting to generalize these
findings to other extremist groups, as for example left-wing extremists or sects.
Since they are less present on the internet (in terms of audio-visuals), future stu-
dies should maybe include also other types of propaganda (homepages, forums,
etc.).
Second, our study focused on the potential ingroup of extremist propaganda, not
on already radicalized individuals as confirmed by the low pre-existing accep-
tance of justifications of terrorism (see Chapter 10.2). The potential influence ex-
tremist propaganda might have in the confirmation of accordant pre-existing at-
titudes or on already radicalized recipients cannot be deduced from our studies.
Particularly the influence of real social groups when confronted with such mate-
rial can be important. In line with this, social influences are discussed as a main
factor in the radicalization process (National Coordinator for Counterterrorism,
2009a). Research has shown that, in interpersonal situations, phenomena such
51 Report in the ARD morning magazine from 07/06 2012. Retrieved online
http://www.einslive.de/medien/html/1live/2012/07/16/ard-morgenmagazin-salafisten.xml
02/09/2012.
118
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 119
as conformity (Asch, 1956), informative and normative social influence (Cialdini
& Goldstein, 2004), de-individuation (Postmes, Spears, & Lea, 1998) or polari-
zation of attitudes towards more risk acceptance (Blascovich, Ginsburg, & Veach,
1975) play a substantial role in forming attitudes and influencing behavior.
Analyzing the evaluation of extremist content in groups with positive compared to
negative attitudes towards the content thus seems promising. However, as already
discussed in Chapter 5, finding such real-life groups comes along with a set of
difficulties for researchers. Besides getting in contact with an accordant group,
people willing to take part may substantially differ from those who do not. Holtz
et al. (2012) pointed out some of these difficulties when analyzing only forums of
extremist groups instead of real-life situations.
As a third limitation, the focus lay on short-term effects. Short- and long-term ef-
fects may differ. Generally, research on persuasion has shown that the effect of
mediated information on attitudes decreased over time (Petty & Wegener,
1998). However, under certain circumstances, persuasive appeals were found to
be stronger after a while.
One the one hand, a so-called sleeper effect (Hovland & Weiss, 1951) cannot be
excluded. Studies showed that factors reducing the persuasive power of a mes-
sage, for instance low credibility of the source or adverse cues (e.g., extremism,
violence), were remembered less after a while, while content memorization re-
mained stable (Kumkale & Albarracín, 2004). As a consequence, general persua-
siveness increased. However, the effect was mainly found under laboratory con-
ditions (Pratkanis, Greenwald, Leippe, & Baumgardner, 1988).
On the other hand, over time, a mere exposure effect might develop (Zajonc,
1968). Persons, situations or objects are perceived as being more positive when
they are seen again, due to the apparent higher familiarity. However, the mere-ex-
posure effect only suggests a better evaluation of propaganda when watching it for
the second or third time and only if it was not perceived as negative during first
contact (Faullant, 2007).
To address these possibilities, further research should consider long-term designs
with repeated presentation of extremist propaganda. Ethical concerns will have to
be regarded and carefully addressed.
In conclusion, despite some limitations, the presented findings can be regarded as
meaningful, especially against the background of the applied methodology. To
our knowledge, this is the first study to experimentally analyze the effects of
right-wing and Islamic extremist propaganda on a non-obtrusive sample. The
main reaction of all our recipients was a rejection of the message and the ideology
behind it. The responses were more pronounced when propaganda of the own cul-
tural ingroup was evaluated. Nonetheless, factors increasing susceptibility (pro-
fessional style of the clips, low educational level, high authoritarianism and ac-
119
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 120
ceptance of violence) as well as starting points for increasing resilience towards
extremist propaganda were identified.
To sum up, the actual project was a first exploratory access, trying to understand
the effects of extremist propaganda 2.0. We hope that it will stimulate further in-
vestigations in this field assessing the new arising questions which can be pulled
out of our results.
120
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 121
References
Aaron, D. (2008). In their own Words-Voices of Jihad. Santa Monica, Arlington,
Pittsburg: RAND Corporation.
Adorno,T. W., Frenkel-Brunswik, E., Levinson, D.J., & Sanford, R.N. (1950). The
Authoritarian Personality. Oxford, England: Harpers.
Agroskin, D., & Jonas, E. (2010). Out of Control: How and why does perceived
lack of control lead to ethnocentrism. Review of Psychology,17 (2), 79–90.
Ajzen, I., & Fishbein, M. (1970). The prediction of behavior from attitudinal and
normative variables. Journal of Experimental Psychology: Learning, Mem-
ory, and Cognition,6, 466–487.
Albert, M., Hurrelmann, K., Quenzel, G., & Infratest Sozialforschung. (2010). Ju-
gend 2010- Eine pragmatische Generation behauptet sich. (Shell Deutsch-
land Holding, Ed.). Frankfurt am Main, Germany: Fisher Taschenbuch Ver-
lag.
Allen, M. (1991). Meta-analysis comparing the persuasiveness of one-sided and
two-sided messages. Western Journal of Speech Communication,55 (4),
390–404. doi:10.1080/10570319109374395
Altemeyer, B. (1988). Enemies of Freedom. San Francisco, California: Jossey-
Bass Inc.
Altheide, D.L., & Cheney, D. (2006). Terrorism and the politics of fear. Cultural
Studies <=> Critical Methodologies,6(4), 415–439. doi:10.1177/
1532708605285733
Arndt, J., Greenberg, J., Schimel, J., Pyszczynski, T., & Solomon, S. (2002). To
belong or not to belong, that is the question: Terror management and iden-
tification with gender and ethnicity. Journal of Personality and Social Psy-
chology,83 (1), 26–43. doi:10.1037//0022–3514.83.1.26
Arnold,W., Eysenck, H. J., & Meili, R. (1997). Lexikon der Psychologie. Freiburg:
Herder.
Aronson, E. (1992). The return of the repressed-Dissonance Theory makes a
comeback. Psychological Inquiry,3(4), 303–311.
Asch, S.E. (1956). Studies of independence and conformity: Minority of one
against an unanimous majority. Psychological Monographs: General and
Applied,70 (9), 1–70.
Bacher, J., & Prosch, B. (2002). Lebensbedingungen und Lebensstile von Auszu-
bildenden Ergebnisse der Leipziger Berufsschulbefragung 2000. Nürnberg:
Universität Erlangen.
121
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 122
Baines, P. R., O’Shaughnessy, Nicholas, J. ., Moloney, K., Richards, B., Butler, S.,
& Gill, M. (2006). Muslim voices: The British Muslim response to Islamic
video-polemic: An exploratory study. Cranfield University School of Ma-
nagement Working Paper Series,3(06), 1–28.
Baines, P. R., O’Shaughnessy, N.J., Moloney, K., Richards, B., Butler, S., & Gill,
M. (2010). The dark side of political marketing: Islamist propaganda, Re-
versal Theory and British Muslims. European Journal of Marketing,44
(3/4), 478–495. doi:10.1108/03090561011020543
Bakker, E. (2006). Jihadi terrorists in Europe. Den Haag: Netherlands Institute of
International Relations.
Becker, R. (2009). Erlebniswelt Rechtsextremismus. 11. wissenschaftliche Fach-
tagung der kriminologischen Gesellschaft, Uni Gießen . Gießen: Universität
Gießen.
Bem, D.J. (1972). Self-Perception Theory. In L. Berkowitz (Ed.), Advances in Ex-
perimental Social Psychology (6th ed., pp. 2–57). New York: Academic
Press Inc.
Bente, G., Aelker, L., & Fürtjes, M. (2009). Tracing ad experience: Continuous
response measurement in advertising research. In J. Maier, M. Maier, M.
Maurer, C. Reinemann, & V. Mayer (Eds.), Real-Time Response Measure-
ment in the Social Sciences (pp. 117–136). Frankfurt am Main: Peter Lang.
Bente, G., & Krämer, N.C. (2004). Inhaltsanalyse medialer Angebote. In R. Man-
gold, P. Vorderer, & G. Bente (Eds.), Lehrbuch Medienpsychologie
(pp. 201–228). Göttingen: Hogrefe Verlag für Psychologie.
Bernays, E.L. (1942). The marketing of national policies: A study of war propa-
ganda. Journal of Marketing,6(3), 236–244.
Blascovich, J., Ginsburg, G. P., & Veach,T.L. (1975). A pluralistic explanation of
choice shifts on the risk dimension. Journal of Personality and Social Psy-
chology,31 (3), 422–429. doi:10.1037/h0076479
Bockstette, C. (2008). Jihadist Terrorist use of Strategic Communication Ma-
nagement Techniques.George C. Marshall. Garmisch-Partenkirchen, Ger-
many.
Bolls, P. D., Lang, A., & Potter, R.F. (2001). The effects of message valence and
listener arousal on attention, memory, and facial muscular responses to
radio advertisements. Communication Research,28 (5), 627–651.
doi:10.1177/009365001028005003
Bueno de Mesquita, E., & Dickson, E. S. (2007). The propaganda of the deed: Ter-
rorism, counterterrorism, and mobilization. American Journal of Political
Science,51 (2), 364–381. doi:10.1111/j.1540–5907.2007.00256.x
122
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 123
Bundesamt für Verfassungsschutz. (2008). Verfassungsschutz gegen Rechtsextre-
mismus. Cologne.
Bundesamt für Verfassungsschutz. (2009). “Autonome Nationalisten”– Rechts-
extremistische Militanz. Cologne.
Bundesministerium des Inneren. (2004). Extremismus in Deutschland: Erschei-
nungsformen und aktuelle Bestandsaufnahme. Berlin: Bundesministerium
des Inneren.
Burke, J. (2004). Al’Qaeda. The True Story of Radical Islam. London: Penguin
Bokks, Ltd.
Busch, C. (2005). Rechtsradikale Vernetzung im Internet. WeltTrends,13 (48),
67–68.
Busch, C. (2008). Rechtsradikale im Web 2.0 – “.. . take up Positions on ‘Main-
stream’ Group.” In U. Dittler & M. Hoyer (Eds.), Aufwachsen in virtuellen
Medienwelten – Chancen und Gefahren digitaler Medien aus medienpsy-
chologischer und medienpädagogischer Perspektive (pp. 1–17). München:
kopaed.
Canetti-Nisim, D. (2004). The effect of religiosity on endorsement of democratic
values: The mediating influence of authoritarianism. Political Behavior,26
(4), 377–398.
Cantor, J. (2002). Fright reactions to mass media. In J. Bryant & D. Zillmann
(Eds.), Media Effects Advances in Theory and Research (pp. 287–306).
Mahwah, N.J.: Lawrence Erlbaum Associates.
Chen, S., Duckworth, K., & Chaiken, S. (1999). Motivated heuristic and systema-
tic processing. Psychological Inquiry,10 (1), 44–19.
Cialdini, R.B., & Goldstein, N.J. (2004). Social influence: compliance and con-
formity. Annual review of psychology,55 (1974), 591–621. doi:10.1146/an-
nurev.psych.55.090902.142015
Claus, R., Lehnert, E., & Müller,Y. (2010). Was ein rechter Mann ist – Männlich-
keit im Rechtsextremismus. (Rosa-Luxemburg Stiftung, Ed.). Berlin: Dietz.
Collins, A.M., & Loftus, E.F. (1975). A spreading-activation theory of semantic
processing. Psychological Review,82 (6), 407–428. doi:10.1037//0033–
295X.82.6.407
Combating Terrorism Center at WestPoint. (2006). The Islamic Imagery Project
Visual Motifs in Jihadi Internet Propaganda. West Point.
Corman, S.R., Schiefelbein, J.S., Acheson, K., Goodall, B., Mcdonald, K., &
Trethewey, A. (2006). Communication and Media Strategy in the Jihadi
War of Ideas. Consortium for Strategic Communication.
123
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 124
Costa, P. T., & McCrae, R.R. (1992). Revised NEO Personality Inventory (NEO-
PI-R) and NEO Five-Factor Inventory (NEO-FFI) Professional Manual.
Odessa, FL: Psychological Assessment Resources.
Davison,W. P. (1983). The Third-Person effect in communication. Public Opinion
Quarterly,47 (1), 1–15. doi:10.1086/268763
Decker, O.,Weißmann, M., Kiess, J., & Brähler, E. (2010). Die Mitte in der Krise
(pp. 1–180). Berlin: Friedrich-Ebert Stiftung.
Denso, C. (2011, March 10). Dschihad im Netz. Zeit online, pp. 1–4. www.zeit.de
Duck, J.M. (1995). The perceived impact of the mass media: reconsidering the
third person effect. European Journal of Social Psychology,25, 77–93.
doi:10.1002/ejsp.2420250107
Duckitt, J. (2011). Authoritarianism and group identification: A new view of an
old construct. Political Psychology,10 (1), 63–84.
Duckitt, J., & Sibley, C.G. (2010). Personality, ideology, prejudice, and politics: a
dual-process motivational model. Journal of Personality,78 (6), 1861–
1893. doi:10.1111/j.1467–6494.2010.00672.x
Duffy, E. (1962). Activation and Behavior (p. xvii, 384). Oxford, England: Wiley.
Ebert, L., Feierabend, S., Karg, U., & Rathgeb, T. (Eds.). (2011). JIM 2011-Ju-
gend, Information, (Multi-)Media. Stuttgart: Medienpädagogischer For-
schungsverband Südwest.
Egerton, F. (2011). Jihad in the West: The Rise of Militant Salafism. Cambridge:
Cambridge University Press.
Ekehammar, B., Akrami, N., Gylje, M., & Zakrisson, I. (2004). What matters
most to prejudice: Big Five personality, social dominance orientation, or
right-wing authoritarianism? European Journal of Personality,18 (6),
463–482. doi:10.1002/per.526
Engelmann, K., Günther, F., Heise, N., Hohmann, F., Irrgang, U., & Schmidt, S.
(2010). Muslimische Weblogs. Der Islam im deutschsprachigen Internet.
Berlin: Frank & Timme.
English, K., Sweetser, K.D., & Ancu, M. (2011). YouTube-ification of Political
Talk: An Examination of Persuasion Appeals in Viral Video. American Be-
havioral Scientist,55 (6), 733–748. doi:10.1177/0002764211398090
Europol. (2011). EU Terrorism Situation and Trend Report. European Police Of-
fice. doi:10.2813/14705. Den Haag.
Faullant, R. (2007). Psychologische Determinanten der Kundenzufriedenheit:
Der Einfluss von Emotionen und Persönlichkeit. Wiesbaden: Deutscher
Universitätsverlag.
124
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 125
Fazio, R.H. (2001). On the automatic activation of associated evaluations: An
overview. Cognition and Emotion,15, 115–141. doi:10.1037/0022–
3514.50.2.229
Fazio, R.H., Jackson, J.R., Dunton, B.C., & Williams, C.J. (1995). Variability in
automatic activation as an unobtrusive measure of racial attitudes: A bona
fide pipeline? Journal of Personality and Social Psychology,69, 1013–
1027.
Feather, N. (1990). Bridging the gap between values and actions: Recent applica-
tions of the expectancy-value model. In I.E. Higgins & R. Sorrentino (Eds.),
Handbook of Motivation and Cognition: Foundations of Social Behavior
(2nd ed., pp. 151–192). New York: Guilford Press.
Ferguson, C.J. (2007). Media violence effects and violent crime: Good science or
moral panic? In C.J. Ferguson (Ed.),Violent Crime: Clinical and Social Im-
plications (pp. 37–56). Thousand Oaks, CA: SAGE Publications, Inc.
Field, A. (2009). Discovering Statistics Using SPSS (3rd ed.). Los Angeles, Lon-
don, New Delhi, Singapore, Washington D.C.: Sage.
Finch, L. (2000). Psychological propaganda: The war of ideas on ideas during the
first half of the twentieth century. Armed Forces & Society,26 (3), 367–386.
doi:10.1177/0095327X0002600302
Finsnes, C. (2010). What is audio-visual jihadi propaganda. Norwegian Defense
Research Establishment (FFI). Retrieved from http://www.ffi.no/no/Rap-
porter/10–00960.pdf. Kjeller, Norway.
Fischer, P., Greitemeyer,T., Kastenmüller, A., Frey, D., & Oßwald, S. (2007). Ter-
ror salience and punishment: Does terror salience induce threat to social or-
der? Journal of Experimental Social Psychology,43 (6), 964–971.
doi:10.1016/j.jesp.2006.10.004
Frindte,W., Boehnke, K., Kreikenbronn, H., & Wagner,W. (2012). Lebenswelten
junger Muslime in Deutschland. Berlin: Bundesministerium des Inneren.
Frindte, W., & Haußecker, N. (2010). Inszenierter Terrorismus. Wiesbaden: Ver-
lag für Sozialwissenschaften Springer Fachmedien.
Frischlich, L. & Rieger, D. (2013). Dealing with the dark side: Negative ingroups
and the effects of right-wing and islamic extremist propaganda videos. Pa-
per presented at the 63rd Annual International Communication Association
Conference, London, UK.
Fuchs, M. (2003). Rechtsextremismus von Jugendlichen: Zur Erklärungskraft
verschiedener theoretischer Konzepte. Kölner Zeitschrift für Soziologie
und Sozialpsychologie,55 (4), 654–678.
125
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 126
Gambetta, D., & Hertog, S. (2007). Engineers of Jihad (pp. 1–90). Oxford: Ox-
ford University Press.
Glaser, S. (2011). Rechtsextremismus online Aktuelle Entwicklungen und Strate-
gien zur Bekämpfung. Medien + Erziehung,55 (5), 1–10.
Glock, S. (2010). Rauchen ist uncool!: Die Repräsentation der verschiedenen As-
pekte des Rauchens im Selbstkonzept und ihre Bedeutung für die Verände-
rung des Verhaltens. Hamburg: Dr. Kovack.
Goodall, C.E. (2011). An overview of implicit measures of attitudes: Methods,
mechanisms, strengths, and limitations. Communication Methods & Mea-
sures,5(3), 203–222.
Greenberg, J., Pyszczynski,T., Solomon, S., Rosenblatt, A.,Veeder, M., & Kirk-
land, S. (1990). Evidence for Terror Management Theory II: The effects of
mortality salience on reactions to those who threaten or bolster the cultural
worldview. Journal of Personality and Social Psychology,58 (2), 308–318.
Greenwald, A.G., & Banaji, M.R. (1995). Implicit social cognition: attitudes,
self-esteem, and stereotypes. Psychological Review,102 (1), 4–27.
Greenwald, A.G., McGhee, D.E., & Schwartz, J.L. K. (1998). Measuring indivi-
dual differences in implicit cognition: The Implicit Association Test. Jour-
nal of Personality and Social Psychology,74 (6), 1464–1480. doi:10.1037//
0022–3514.74.6.1464
Greenwald, Banaji, M.R., Rudman, L.A., Farnham, S.D., Nosek, B.A., & Mel-
lott, D.S. (2002). A unified theory of implicit attitudes, stereotypes, self-es-
teem, and self-concept. Psychological Review,109, 3–25.
Grings, W., & Dawson, M.E. (1978). Emotions and bodily responses: A psycho-
physiological approach. New York: Academic Press.
Gunn, G.R., & Wilson, A.E. (2011). Acknowledging the skeletons in our closet:
the effect of group affirmation on collective guilt, collective shame, and re-
paratory attitudes. Personality & Social Psychology Bulletin,37 (11),
1474–1487. doi:10.1177/0146167211413607
Harmon-Jones, E., Simon, L., Greenberg, J., Pyszczynski, T., Solomon, S., &
McGregor, H. (1997). Terror management theory and self-esteem: evidence
that increased self-esteem reduces mortality salience effects. Journal of
Personality & Social Psychology,72 (1), 24–36.
Hefner, D., Rothmund,T., Klimmt, C., & Gollwitzer, M. (2011). Implicit measures
and media effects research: Challenges and opportunities. Communication
Methods and Research,5(3), 181–202.
Hegghammer,T. (2006). Terrorist Recruitment and Radicalization in Saudi Ara-
bia. Middle East Policy,13 (4), 39–60.
126
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 127
Heitmeyer,W. (1992). Die Bielefelder Rechtsextremismus-Studie: Erste Langzeit-
untersuchung zur politischen Sozialisation männlicher Jugendlicher. Wein-
heim, München: Juventa.
Heitmeyer, W. (2002). Deutsche Zustände (1st ed.). Frankfurt am Main: Suhr-
kamp.
Heitmeyer,W., Borstel, D., Grau, A., & Marth, J. (2010). Rechtsextreme Struktu-
ren, Gruppenbezogene Menschenfeindlichkeit und bürgerschaftliches En-
gagement in der Landeshauptstadt Dresden. Bielefeld: Landesprävention
Sachsen.
Hodson, G., & Busseri, M.a. (2012). Bright minds and dark attitudes: lower cog-
nitive ability predicts greater prejudice through right-wing ideology and
low intergroup contact. Psychological Science,23 (2), 187–195.
doi:10.1177/0956797611421206
Hoffmann, B. (2006). Terrorismus- der unerklärte Krieg. Frankfurt: Fischer.
Hogg, M.A. (2000). Subjective uncertainty reduction through self-categoriza-
tion: A motivational theory of social identity processes. European Review
of Social Psychology,11 (1), 223–255. doi:10.1080/14792772043000040
Holtmann, P. (2009). Martyrdom, not suicide: The legality of Hamas’ bombing in
the mid-1990s in modern Islamic jurisprudence. Norderstedt: GRIN Verlag.
Holtz, P., Kronberger, N., & Wagner, W. (2012). Analyzing internet forums: A
practical guide. Journal of Media Psychology: Theories, Methods, and Ap-
plications,24 (2), 55–66.
Hovland, C. I., & Weiss,W. (1951). The influence of source credibility on commu-
nication effectiveness. Public Opinion Quarterly,15 (4), 635–650.
Hudson, R.A., Majeska, M., & Metz, H. C. (1999). The Sociology and Psychology
of Terrorism: Who Becomes a Terrorist and Why? Washington: Library of
Congress.
Hussy,W., & Jain, A. (2002). Experimentelle Hypothesenprüfung in der Psycho-
logie. Göttingen: Hogrefe Verlag für Psychologie.
Inan, A. (2007). Islam goes Internet. Websites islamischer Organisationen im
World Wide Web. Marburg: Tectum Verlag.
Intel Center. (2005). Evolution of Jihadi Video. Alexandria, USA: IntelCenter/
Temperest Publishing.
Intel Center. (2006). Jihadi Master Video Guide. Alexandria, USA: IntelCenter/
Temperest Publishing.
127
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 128
Jenkins, B.M. (2010). Would-Be Warriors Incidents of Jihadist Terrorist Radica-
lization in the US Since September 11, 2001. Santa Monica, Arlington,
Pittsburg: Cooperation, RAND.
Jowett, G. S., & O’Donnell, V. (2012). Propaganda & Persuasion. Los Angeles,
London, New Delhi, Singapore, Washington D.C.: SAGE.
Karpinski, A., & Steinman, R. B. (2006). The single category implicit association
test as a measure of implicit social cognition. Journal of Personality and So-
cial Psychology,91 (1), 16–32.
Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion
Quarterly,24 (2, Special Issue: Attitude Change), 163. doi:10.1086/266945
Kemmesies, U. (2006). Terrorismus und Extremismus – der Zukunft auf der Spur.
Forschungsstand zum Phänomenfeld des islamischen Extremismus und Ter-
rorismus. München: Luchterhand.
Kemmesies, U. (2007). Forschungsstand zum Phänomenfeld des islamischen Ex-
tremismus und Terrorismus. (U. Kemmesies, Ed.)Terrorismus und Extre-
mismus – der Zukunft auf der Spur (pp. 1–61). Wiesbaden: Bundeskriminal-
amt.
Khorchide, M., & Schenk, A. (2012, May 24). Gott ist kein Diktator. die Zeit,
p. 66.
Kimmage, D. (2008). The Al’Qaeda Media Nexus. Washington: Radio Free
Europe/Radio Liberty.
Kimmage, D., & Ridolfo, K. (2007). Iraqi insurgent media: The war of images and
ideas. Journal of International & Security Studies,1(2), 7–89.
King, M.F., & Bruner, G.C. (2000). Social Desirability Bias: A neglected aspect
of validity testing. Psychology & Marketing,17 (2), 79–103.
Kumkale, G.T., & Albarracín, D. (2004). The sleeper effect in persuasion: A
meta-analytic review. Psychological Bulletin,130 (1), 143–172.
Landesamt für Verfassungschutz. (2003). Ideologie und Mythen online. Cologne:
Landesamt für Verfassungsschutz NRW.
Lang, A. (2000). The limited capacity model of mediated message processing.
Journal of Communication,50 (1), 46–70. doi:10.1093/joc/50.1.46
Langenbacher, N., & Schellenberg, B. (2011). Is Europe on the “Right” Path?
Bonn: Friedrich-Ebert-Stiftung.
Larsen, K.S., & Schwendiman, G. (1969). Authoritarism, self-esteem and inse-
curity. Psychological Reports,25, 229–230.
128
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 129
Lasswell, H.D. (1948). The Communication of Ideas. New York: Harper and
Brothers.
Lasswell, H.D. (1927). Propaganda Techniques in the World War. (T. & co Kegan
Paul, Trench, Ed.). London.
Lewis, B. (2003). The Crisis of Islam. New York: Random House Inc.
Lewis, E.S. (1903). Catch-line and argument. The Book-Keeper,15, 124 ff.
Lia, B. (2007a). Al-Suri ’s doctrines for decentralized Jihadi training – part 1. Ter-
rorism Monitor,5(1), 1–11.
Lia, B. (2007b). Architect of global jihad- The Life of al-Qaida Strategist Abu
Musáb al-Suri. London: HURST Publishers.
Liu, J.E., & Smeesters, D. (2010). Have you seen the news today? The effect of
death-related media contexts on brand preferences. Journal of Marketing
Research,47 (2), 251–262. doi:10.1509/jmkr.47.2.251
Loza,W. (2007). The psychology of extremism and terrorism: A Middle-Eastern
perspective. Aggression and Violent Behavior,12 (2), 141–155.
doi:10.1016/j.avb.2006.09.001
Lützinger, S. (2010). Die Sicht der Anderen: Eine qualitative Studie zu extremis-
tischen Biographien. München: Bundeskriminalamt/Hermann Luchter-
hand Verlag.
Maehler, D., & Schmidt-Denter, U. (n.d.). Handbuch Migrationsforschung: Leit-
faden und Messverfahren zur Erfassung psychologischer Konstrukte bei
Personen mit Migrationshintergrund in Deutschland.
Mandler, G. (1992). Memory, arousal, and mood: A theoretical integration. In S.-
Å. Christianson (Ed.), The Handbook of Emotion and Memory: Research
and Theory (pp. 93–110). Hillsdale, NJ, England: Lawrence Erlbaum As-
sociates.
Marshall, R.D., & Galea, S. (2004). Science for the community: Assessing mental
health after 9/11. The Journal of Clinical Psychiatry,65 (1), 37–43.
Mayring, P., & Gläser-Zikuda, M. (2008). Die Praxis der Qualitativen Inhaltsana-
lyse (2nd. ed.). Weinheim & Basel: Beltz.
McGuire, W. J. (1968). Attitudes and attitude change. In A. G. Greenwald, T.C.
Brock, & T.M. Ostrom (Eds.), Psychological Foundations of Attitudes
(pp. 171–196). New York: Academic Press.
McGuire,W.J., & Papageorgis, D. (1961). The relative efficacy of various types of
prior belief-defense in producing immunity against persuasion. Journal of
Abnormal and Social Psychology,62 (2), 327–37.
129
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 130
McKenna, K.Y.A., & Bargh, J.A. (1998). Coming out in the age of internet. Jour-
nal of Personality and Social Psychology,75 (3), 681–694.
Merten, K. (2000). Struktur und Funktion von Propaganda. Publizistik,45 (2),
143–162.
Moorman, M. (2003). Content considered. The Relationship between media en-
vironment and advertising Effects. Amsterdam: Universiteit van Amster-
dam.
Mudde, C. (1995). Right-wing extremism analyzed. A comparative analysis of the
ideologies of three alleged Right- Wing Extremist Parties (NPD, NDP, CP ’
86). European Journal of Political Research,27, 203–224.
Mudde, C. (2000). The Ideology of the Extreme Right. Manchester and New York:
Manchester University Press.
Munton, T., Martin, A., Lorenc, T., Marrero-Guillamon, I., Jamal, F., Lehmann,
A., Cooper, C., et al. (2011). Understanding Vulnerability and Resilience
in Individuals to the Influence of al Qaida Violent Extremism. Croydon:
Great Britain Home Office.
Musharabash,Y. (2006, July). The London bombings: One year after a new breed
of terror was born. Spiegel Online. Retrieved from http://www.spiegel.de/
international/the-london-bombings-one-year-after-a-new-breed-of-terror-
was-born-a-425504.html [06/02/13].
Müller, E., & Seiler, O. (2010). Virtuelle Selbstinszenierung einer rechtsradikalen
Jugendsubkultur am Beispiel media-pro-patria.net. In C. Busch (Ed.),
Rechtsradikalismus im Internet (pp. 158–177). Siegen: Universitätsverlag
Siegen.
Münker, S. (2009). Emergenz digitaler Öffentlichkeiten: Die sozialen Medien im
Web 2.0. Frankfurt am Main: Suhrkamp.
Münkler, H. (2002). Die neuen Kriege [The New Wars]. (Rowohlt, Ed.). Reinbek:
Germany.
National Coordinator for Counterterrorism. (2009a). Ideology and Strategy of
Jihadism. BZ: The Hague.
National Coordinator for Counterterrorism. (2009b). Jihadists and the Internet.
BZ The Hague.
Neugebauer, G. (2010). Einfach war gestern. Zur Strukturierung der politischen
Realität in einer modernen Gesellschaft. Politik und Zeitgeschichte, Beilage
zur Wochenzeitung das Parlament,44, 3–9.
Oesterreich, D. (2005). Flight into security: A new approach and measure of the
authoritarian personality. Political Psychology,26 (2), 275–297.
130
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 131
Olson, M.A., & Fazio, R.H. (2003). Relations between implicit measures of pre-
judice: what are we measuring? Psychological Science,14 (6), 636–9.
O’Shaughnessy, N. (2012). The death and life of propaganda. Journal of Public
Affairs,12 (1), 29–38. doi:10.1002/pa
O’Shaughnessy, N.J., & Baines, P. R. (2009). Selling terror: The symbolization
and positioning of Jihad. Marketing Theory,9(2), 227–241. doi:10.1177/
1470593109103069
Payne, K. (2009). Winning the battle of ideas: Propaganda, ideology, and terror.
Studies in Conflict & Terrorism,32 (2), 109–128. doi:10.1080/
10576100802627738
Petty, R.E., & Wegener, D.T. (1998). Attitude change: Multiple roles for persua-
sion variables. In D.T. Gilbert, S.T. Fiske, & G. Lindzey (Eds.),The Hand-
book of Social Psychology (pp. 1–78). Oxford: Oxford University Press.
Petzel,T.,Wagner, U., Nicolai, K., & Van Dick, R. (1997). Ein kurzes Instrument
zur Messung der Autoritarismus-Neigung. Gruppendynamik,28, 251–258.
Pfau, M., & Szabo, E.A. (2003). Inoculation and resistance to persuasion. In R.
Gass & J. Seiter (Eds.), Perspectives on Persuasion, Social Influence, and
Compliance Gaining. Boston: Allyn & Bacon.
Pfeiffer, T. (2003). “Das Internet ist billig, schnell und sauber. Wir lieben es”,
Rechtsextremisten entdecken den Computer. Retrieved from www.im.nrw.
de/sch/35.htm
Pfister, H. (2012). Extremisten erobern das Internet. Stuttgarter Nachrichten.
Retrieved from http://www.stuttgarter-nachrichten.de/inhalt.internetkrimi-
nalitaet-extremisten-erobern-das-internet.dce7978b-1f02–4b76-a22e-
24723b1839c3.html [18/03/12].
Postmes, T., Spears, R., & Lea, M. (1998). Breaching or building social bound-
aries: Side effects of computer-mediated communication. Communication
Research,25 (6), 689–715.
Pratkanis, A.R., Greenwald, A.G., Leippe, M.R., & Baumgardner, M.H. (1988).
In search of reliable persuasion effects III – The sleeper effect is dead. Long
live the sleeper effect. Journal of Personality and Social Psychology,54 (2),
203–218.
Precht, B.T. (2007). Home grown Terrorism and Islamist Radicalisation in
Europe: From Conversion to Terrorism. Copenhagen: Ministry of Justice.
Propper, R.E., Stickgold, R., Keeley, R., & Christman, S. D. (2007). Is television
traumatic? Psychological Science,18 (4), 334–341.
131
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 132
Puschnerat,T. (2006). Zur Bedeutung ideologischer und sozialer Faktoren in isla-
mistischen Radikalisierungsprozessen – eine Skizze. Möglichkeiten und
Grenzen einer prognostischen Empirie. In U. Kemmesies (Ed.),Terrorismus
und Extremismus – der Zukunft auf der Spur. Beiträge zur Entwicklungsdy-
namik von Terrorismus und Extremismus (pp. 217–235). München: Wolters
Kluwer Deutschland GmbH.
Pyszczynski,T., Abdollahi, A., Solomon, S., Greenberg, J., Cohen, F., & Weise, D.
(2006). Mortality salience, martyrdom, and military might: the great satan
versus the axis of evil. Personality & Social Psychology Bulletin,32 (4),
525–37. doi:10.1177/0146167205282157.
Qutb Shaheed, S. (1964). Milestones. Retrieved from http://www.izharudeen.-
com/uploads/4/1/2/2/4122615/milestones_www.izharudeen.com.pdf [06/
02/13].
Rammstedt, B. (2007). The 10-Item Big Five Inventory. European Journal of Psy-
chological Assessment,23 (3), 193–201. doi:10.1027/1015–5759.23.3.193
Rammstedt, B., & John, O.P. (2007). Measuring personality in one minute or less:
A 10-item short version of the Big Five Inventory in English and German.
Journal of Research in Personality,41 (1), 203–212. doi:10.1016/
j.jrp.2006.02.001
Ramírez, C.Z., & Verkuyten, M. (2011). Values, media framing, and political tol-
erance for extremist groups. Journal of Applied Social Psychology,41 (7),
1583–1602.
Ravaja, N. (2004). Contributions of psychophysiology to media research. Media
Psychology,6, 193–235.
Renaud, D., & Unz, D. (2006). Die M-DAS – eine modifizierte Version der Dif-
ferentiellen Affekt Skala zur Erfassung von Emotionen bei der Mediennut-
zung. Zeitschrift für Medienpsychologie,18 (2), 70–75. doi:10.1026/1617–
6383.18.2.70
Rippl, S., & Seipl, C. (2002). Ökonomische Lage, Bildungsniveau und Fremden-
feindlichkeit. In K. Boehnke, D. Fuß, & J. Hagan (Eds.), Jugendgewalt und
Rechtsextremismus (pp. 79–91). München: Weinheim.
Roessing,T., & Siebert, S. (2006). Perception and Assessment of Left- and Right-
Wing Extremism by Public Opinion. An Experimental Study. 59th WAPOR
Annual Conference, (pp. 1–11). Montreal, Canada: WAPOR.
Rogan, H. (2007). Al’Qaeda’s online media strategies: From Abu Reuter to Ir-
habi. Norwegian Defense Research Establishment (FFI).
132
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 133
Rogers, M.B., Loewenthal, K.M., Lewis, C.A., Amlôt, R., Cinnirella, M., & An-
sari, H. (2007). The role of religious fundamentalism in terrorist violence: a
social psychological analysis. International Review of Psychiatry,19 (3),
253–62. doi:10.1080/09540260701349399
Rosenberg, M., & Pearlin, L.I. (1978). Social class and self-esteem among chil-
dren and adults. American Journal of Sociology,84 (1), 53–77.
Roth, M., & Hammelstein, P. (2003). Sensation Seeking-Konzeption, Diagnostik
und Anwendung. Göttingen, Ber n, Toronto, Seattle: Hogrefe.
Rubinstein, G. (1996). Two peoples in one land: A validation study of Altemeyer’s
right-wing authoritarianism scale in the Palestinian and Jewish societies in
Israel. Journal of Cross-Cultural Psychology,27 (2), 216–230. doi:10.1177/
0022022196272005
Ruhe, M. (2010). Die neue Rechte – Eine attraktive Jugendkultur? Technische
Universität Dortmund.
Sageman. (2004). Understanding Terror Networks. Philadelphia, Pennsylvania:
University of Pennsylvania.
Sageman. (2008). Leaderless Jihad – Terror Networks in the 21st Century. Phila-
delphia, Pennsylvania: University of Pennsylvania.
Salem, A., Reid, E., & Chen, H. (2008). Multimedia content coding and analysis:
Unraveling the content of Jihadi extremist groups’ videos. Studies in Con-
flict & Terrorism,31 (7), 605–626. doi:10.1080/10576100802144072
Sarangi, S., & Canter, D.(2009). The rhetorical foundation of militant jihad. In D.
Canter (Ed.), The Faces of Terrorism. Multidisciplinary Perspectives
(pp. 35–61). Chichester: John Wiley & Sons Ltd.
Schedler, J. (2009). Übernahme von Ästhetik und Aktionsformen der radikalen
Linken – Zur Verortung der “Autonomen Nationalisten” im extrem rechten
Strategiespektrum. In S. Braun, A. Geisler, & M. Gerster (Eds.), Strategien
der extremen Rechten. Hintergründe – Analysen – Antworten (pp. 332–
357). Wiesbaden: VS Verlag für Sozialwissenschaften/GWV Fachverlage
GmbH.
Schedler, J. (2011). Style matters: Inszenierungspraxen ‘Autonomer Nationalis-
ten’. In Jan Schedler & A. Häusler (Eds.), Autonome Nationalisten- Neona-
zis in Bewegung,Wiesbaden (pp. 67–89). Wiesbaden: VS Verlag für Sozial-
wissenschaften. doi:10.1007/978–3-531–9
Schmeisser, D.R., Bente, G., & Isenbart, J. (2004). Am Puls des Geschehens – Die
Integrierte Rezeptionsprozessanalyse. Zum Mehrwert rezeptionsbegleiten-
der Untersuchungsmethoden in der Werbewirkungsforschung. Planung und
Analyse,1, 28–34.
133
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 134
Schmidt, J.H. (2009). Das neue Netz: Merkmale, Praktiken und Folgen des Web
2.0. Konstanz: UVK.
Schmidt, J.H., Lampert, C., & Schwinge, C. (2010). Nutzungspraktiken im Social
Web: Impulse für die medienpädagogische Diskussion. In B. Herzig, D.M.
Meister, & H. Niesyto (Eds.), In Jahrbuch Medienpädagogik 8: Medien-
kompetenz und Web 2.0. (pp. 255–270). Wiesbaden: Springer.
Schnabel, K., Banse, R., & Asendorpf, J.B. (2006). Assessment of implicit per-
sonality self-concept using the Implicit Association Test (IAT): Concurrent
assessment of anxiousness and angriness. British Journal of Social Psychol-
ogy,45, 373–396.
Schneckener, U. (2006). Transnationaler Terrorismus. (Suhrkamp, Ed.). Frank-
furt am Main.
Schubert, K., & Klein, M. (1997). Das Politiklexikon. Bonn: Bundeszentrale für
Politische Bildung.
Seib, P., & Janbek, D.M. (2011). Global Terrorism and New Media. New York:
Routledge.
Spencer, S.J., Steele, C.M., & Quinn, D.M. (1999). Stereotype threat and wo-
men’s math performance. Journal of Experimental Social Psychology,35
(1), 4–28. doi:10.1006/jesp.1998.1373
Tajfel, H. (1982). Social Psychology of Intergroup Relations. Annual Review of
Psychology,33 (1), 1–39.
Tajfel, H., & Turner, J.C. (1986). The social identity theory of intergroup beha-
vior. In S. Worchel & W.G. Austin (Eds.), Psychology of Intergroup Rela-
tions (pp. 7– 24). Chicago, IL: Nelson-Hall.
Taylor, P. M. (2003). Munitions of the Mind – A History of Propaganda from the
Ancient World to the Present Day (3rd. ed.). Manchester and New York:
Manchester University Press.
Terwey, M. (2000). Ethnozentrismus in Deutschland: Seine weltanschaulichen
Konnotationen im sozialen Kontext. In R. Alba, P. Schmidt, & M.H. Was-
mer (Eds.), Deutsche und Ausländer: Fremde, Freunde oder Feinde?
(pp. 401–438). Wiesbaden: Westdeutscher Verlag.
Thackrah, J.R. (2004). Dictionary of Terrorism. London, New York: Routledge.
Thamm, B.G. (2005). Al-Quaida Das Netzwerk des Terrors (pp. 1–247).
München: Heinrich Hugendubel Verlag.
Torres-Soriano, M.R. (2009). Spain as an Object of Jihadist Propaganda.
Studies in Conflict & Terrorism,32 (11), 933–952. doi:10.1080/
10576100903259977
134
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 135
Verfassungsschutz. (2004). White Pride Worldwide. Cologne: Landesamt für Ver-
fassungsschutz NRW.
Wagner, S. (2008, April 24). Zehn Wahrheiten über den Ku-klux-klan. Focus On-
line. Retrieved from www.focus.de
Wagner, U., Christ, O., & Kühnel, S.M. (2002). Diskriminierendes Verhalten: Es
beginnt mit Abwertungen. In Suhrkamp (Ed.), Deutsche Zustände. Frank-
furt a.M.: Germany.
Wasmer, M., Koch, A., Harkness, J., & Gabler, S.(1996). Konzeption und Durch-
führung der “Allgemeinen Bevölkerungsumfrage der Sozialwissenschaf-
ten” (ALLBUS) 1996, (Mannheim).
Watson, S., Appiah, O., & Thornton, C.G. (2011). The Effect of name on pre-in-
terview impressions and occupational stereotypes: The case of black sales
job applicants. Journal of Applied Social Psychology,41 (10), 2405–2420.
doi:10.1111/j.1559–1816.2011.00822.x
Wegner, P. (2009). Die neuen Krieger – die Propaganda der Al-Qaida als Mobi-
lisierungswerkzeug. Saarbrücken: Verlag Dr. Müller.
Weichert, S. A. (2007). Die Propaganda der Tat – Zur Kommunikationsstrategie
des modernen Aufmerksamkeitsterrorismus. In S. Glaab (Ed.), Medien
und Terrorismus. Auf den Spuren einer Symbiotischen Beziehung (pp. 83–
98). Berlin: Berliner Wissenschaftsverlag.
Weimann, G. (2006). Terroron the Internet. Washington, D.C.: United States In-
stitute of Peace.
Wilner, A., & Dubouloz, C.-J. (2009). Homegrown terrorism and transformative
learning: An interdisciplinary approach to understanding radicalization.
Global Change, Peace & Security,22 (1), 33–51.
Winn, A. & Zakem, V.L. (2009). Jihad.com 2.0. The New Social Media and the
Changing Dynamics of Mass Persuasion. In J.J.F. Forst (Ed.), Influence
Warfare. How Terrorists and Governments Fight to Shape Perceptions in
a War of Ideas (pp. 27–48). Westport, Connecticut & London: Praeger Se-
curity International.
Wohl, M.J.a., Branscombe, N.R., & Klar,Y. (2006). Collective guilt: Emotional
reactions when one’s group has done wrong or been wronged. European Re-
view of Social Psychology,17 (1), 1–37. doi:10.1080/10463280600574815
Wortmann, T. (2007, August 8). Karrierekiller Internet – Vorsicht, der Personaler
liest mit! Focus Online. Retrieved from http://www.focus.de/finanzen/
karriere/bewerbung/internet/tid-7051/bewerbung_aid_69068.html [06/02/
13].
135
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 136
Wyer, R. S., & Srull,T.K. (1994). Handbook of Social Cognition: Basic Processes.
Mahwah, New Jersey: Lawrence Erlbaum Associates.
Zajonc, R.B. (1968). Attitudinal effects of mere exposure. Journal of Personality
and Social Psychology,9(2), 1–27.
Zhu, B., Chen, C.C., Loftus, E.F., He, Q., Lei, X., Lin, C., & Dong, Q. (2012).
Brief exposure to misinformation can lead to long-term false memories.
Applied Cognitive Psychology,26 (2), 301–307. doi:10.1002/acp.1825
Zillmann, D. (2004). Emotionspsychologische Grundlagen. In Roland. Mangold,
P. Vorderer, & G. Bente (Eds.), Lehrbuch der Medienpsychologie (pp. 102–
128). Göttingen: Hogrefe Verlag für Psychologie.
Zimbardo, P. G., & Leippe, M.R. (1991). The Psychology of Attitude Change and
Social Influence. New York: McGraw-Hill.
Zuckermann, M. (1979). Sensation seeking: Beyond the Optimal Level of Arou-
sal. Hillsdale, N.J.: Lawrence Erlbaum Associates.
136
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 137
Appendix – Supplementary Statistical Material
The supplementary statistical material addresses mainly those already used to in-
terpreting statistical results. Therefore, values used are not explained.
A1. Factor Loadings and Multilevel Factor Analysis
All analyses were conducted using the statistical software Mplus version 6.11
(Byrne, 2012). The corresponding 2-level EFAwith a robust maximum likelihood
estimator and a geomin rotation method revealed five factors based on 14 items
with significant factor loadings and the same factor structure on both levels.
Ten items were dropped in the subsequent CFA due to cross-loadings indicating
lack of construct sensitivity. The final 2-level CFA denoted an excellent model
fit with x2(123) = 222.94, p < .01, RMSEA = .02, CFI = .98, TLI = .97,
SRMRwithin = .02, and SRMRbetween = .07.
137
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 138
Table A1. Factor Loadings of Items on the Cognitive Post-hoc Factors: Within Level
Cognitive Evaluations Stimuli-Differences (Within-Level)
Cognitive Evaluations Emotional Evaluations
Item* ICC One-sidedness Persuasiveness Interest Aversion Shame R2
I have now more sympathy with the makers of the video. .28 .76 .58
After this video I can understand the perspective of the
makers much better. .27 .76 .58
Through this video, I feel myself better informed about
the background. .31 .68 .46
On my opinion, the video was persuasive. .30 .66 .18 .63
I found this video one-sided. .36 .84 .29 .38
I found this video sensational. .42 .44 .50 .25 .12
I couldn’t take this video seriously. .26 .59 .35
Note. ICC = Intra-class correlation. All factor loadings were significant, p< .05. Main loadings are marked in boldface. R2= The amount of total
variance by an item explained by the underlying factor.
* Items were translated for publication purposes.
138
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 139
Table A2. Factor Loadings of Items on the Cognitive Post-hoc Factors: Between Level
Cognitive Evaluations Person-Differences (Between-Level)
Cognitive Evaluations Emotional Evaluations
Item* ICC One-sidedness Persuasiveness Interest Aversion Shame R2
I have now more sympathy with the makers of the video. .28 .95 .89
After this video I can understand the perspective of the makers
much better. .27 .94 .88
Through this video, I feel myself better informed about the
background. .31 .87 .76
In my opinion, the video was persuasive. .30 .91 .82
I found this video one-sided. .36 .94 .88
I found this Video sensational. .42 .78 .32 .61
I couldn’t take this video seriously. .26 .76 .58
Note. ICC = Intraclass correlation. All factor loadings were signif icant, p< .05. Main loadings are marked in bold face. R2= The amount of total variance
of an item explained by the underlying factor.
* Items were translated for publication purposes.
139
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 140
Table A3. Factor Loadings of Items on the Emotional Post-hoc Factors: Within Level
Emotional Evaluations Stimuli-Differences (Within-Level)
Cognitive Evaluations Emotional Evaluations
Item* ICC One-sidedness Persuasiveness Interest Aversion Shame R2
Through this video, I got interested. .29 .86 .73
Through this video, I felt fascination. .40 .68 .46
Through this video, I felt contempt. .30 .83 .69
Through this video, I felt disgust. .31 –.15 .72 .57
Through this video, I felt anger. .32 –.24 .71 .62
Through this video, I felt shame. .35 .87 .75
Through this video, I felt guilt. .33 –.22 .43 .24
Note. ICC = Intra-class correlation. All factor loadings were significant, p< .05. Main loadings are marked in boldface. R2= The amount of total variance
of an item explained by the underlying factor.
* Items were translated for publication purposes.
140
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 141
Table A4. Factor Loadings of Items on the Emotional Post-hoc Factors: Between Level
Emotional Evaluations Person-Differences (Between-Level)
Cognitive Evaluations Emotional Evaluations
Item* ICC One-sidedness Persuasiveness Interest Aversion Shame R2
Through this video, I got interested. .29 .85 .72
Through this video, I felt fascination. .40 .85 .72
Through this video, I felt contempt. .30 .79 .62
Through this video, I felt disgust. .31 .79 .63
Through this video, I felt anger. .32 .93 .86
Through this video, I felt shame. .35 .82 .68
Through this video, I felt guilt. .33 –.25 .73 .59
Note. ICC = Intra-class correlation. All factor loadings were significant, p< .05. Main loadings are marked in boldface. R2= The amount of total variance
by an item explained by the underlying factor.
* Items were translated for publication purposes.
141
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 142
A2. Pairwise Comparisons for the Effect of Single Videos
Table A5. Pairwise comparisons refer to the 6 (Right-wing extremist video clips) respectively 7 (Islamic extremist video clip) rANOVA displayed in Chapter 9.4.3.
Adverse evaluations SCR Aversion Shame One-sidedness
Right-wing Extremist Videos M SE M SE M SE M SE
TH Lifestyle activist -1.42a0.13 2.71a0.10 1.60a0.06 2.23a0.05
RC News broadcast -0.24c0.10 1.85d0.07 1.24c0.05 1.69d0.05
RC Group lifestyle -0.88b0.11 2.01ad 0.08 1.26c0.04 1.99b0.05
RC Public opinion -0.51d0.10 1.98d0.08 1.36b0.05 1.87c0.05
Movie clip -0.89b0.10 2.10b0.09 1.39b0.06 2.03b0.06
Extreme clip -1.28a0.11 3.43c0.10 1.53a0.07 1.79c0.06
Note. Different indices in a column indicate significant differences in the pairwise comparisons within this column. p < .05. Values in lines cannot be
compared as scales differed from each other. SCR = Slider rating from –4 (= very unpleasant) to +4 (= very pleasant). Aversion and shame were evaluated
on a five point Likert scale ranging from 1 (= not at all) to 5 (= very strong). One-sidedness was evaluated on a four point Likert scale ranging from
1 (= absolutely disagree)to4(=absolutely agree).
142
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 143
Table A6. Pairwise Comparisons for Online and Post-hoc Ratings: Right-wing Extremist Videos
Adverse evaluations SCR Aversion Shame One-sidedness
Right-wing Extremist Videos M SE M SE M SE M SE
TH Lifestyle activist -1.42a0.13 2.71a0.10 1.60a0.06 2.23a0.05
RC News broadcast -0.24c0.10 1.85d0.07 1.24c0.05 1.69d0.05
RC Group lifestyle -0.88b0.11 2.01ad 0.08 1.26c0.04 1.99b0.05
RC Public opinion -0.51d0.10 1.98d0.08 1.36b0.05 1.87c0.05
Movie clip -0.89b0.10 2.10b0.09 1.39b0.06 2.03b0.06
Extreme clip -1.28a0.11 3.43c0.10 1.53a0.07 1.79c0.06
Note. Different indices in a column indicate significant differences in the pairwise comparisons within this column. p < .05. Values in lines cannot be
compared as scales differed from each other. SCR = Slider rating from –4 (= very unpleasant) to +4 (= very pleasant). Aversion and shame were evaluated
on a five point Likert scale ranging from 1 (= not at all) to 5 (= very strong). One-sidedness was evaluated on a four point Likert scale ranging from 1
(= absolutely disagree)to4(=absolutely agree).
143
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 144
Table A7. Pairwise Comparisons for Online and Post-hoc Ratings: Islamic Extremist Videos
Adverse evaluations SCR Aversion Shame One-sidedness
Islamic Extremist Videos M SE M SE M SE M SE
TH Ideologue -0.72a1.18 1.80a0.08 1.30a0.06 1.83a0.04
TH Lifestyle activist -1.06b1.30 2.36b0.09 1.50b0.07 1.76a0.04
TH Martyr -0.93a1.06 2.05c0.09 1.27a0.05 1.80a0.04
RC News broadcast -1.31c1.31 2.44b0.10 1.30a0.05 2.07b0.05
RC Group lifestyle -1.11b1.15 2.46b0.10 1.32a0.06 1.97c0.04
Movie clip -0.77a0.95 2.34b0.09 1.52b0.07 1.89ac 0.05
Extreme clip -1.18bc 1.13 2.91d0.10 1.47b0.07 1.82a0.04
Note. Different indices in a column indicate significant differences in the pairwise comparisons within this column. p< .05. Values in lines cannot be
compared as scales differed from each other. SCR = Slider rating from –4 (= very unpleasant) to +4 (= very pleasant). Aversion and Shame were
evaluated on a five point Likert scale ranging from 1 (= not at all)to5(=very strong). One-sidedness was evaluated on a four point Likert scale ranging
from 1 (= absolutely disagree)to4(=absolutely agree).
144
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 145
Table A8. Pairwise Comparisons for Post-hoc Ratings: Right-wing Extremist Videos
Approaching evaluations Interest Persuasiveness
Right-wing Extremist Videos M SE M SE
TH Lifestyle activist 2.15a0.09 1.17a0.02
RC News broadcast 2.63c0.10 1.36c0.04
RC Group lifestyle 1.84d0.08 1.14a0.03
RC Public opinion 1.81d0.07 1.17a0.03
Movie clip 2.38b0.10 1.27b0.04
Extreme clip 2.37b0.09 1.44c0.05
Note. Different indices in a column indicate significant differences in the pairwise com-
parisons within this column. p< .05. Values in lines cannot be compared asscales differed
from each other. Interest was evaluated on a five point Likert scale ranging from 1 (= not at
all)to5(=very strong). Persuasiveness was evaluated on a four point Likert scale ranging
from 1 (= absolutely disagree)to4(=absolutely agree).
Table A9. Pairwise Comparisons for Post-hoc Ratings: Islamic Extremist Videos
Approaching evaluations Interest Persuasiveness
Islamic Extremist Videos M SE M SE
TH Ideologue 1.80a0.08 1.21a0.03
TH Lifestyle activist 2.36b0.09 1.39b0.05
TH Martyr 1.86ac 0.08 1.17a0.03
RC News broadcast 1.98c0.08 1.14bc 0.03
RC Group lifestyle 2.20b0.10 1.09c0.02
Movie clip 2.41b0.09 1.34b0.04
Extreme clip 2.31b0.09 1.14b0.03
Note. Different indices in a column indicate significant differences in the pairwise com-
parisons within this column. p< .05. Values in lines cannot be compared asscales differed
from each other. Interest was evaluated on a five point Likert scale ranging from 1 (= not at
all) to 5 (= very strong). Persuasiveness was evaluated on a four point Likert scale ranging
from 1 (= absolutely disagree) to 4 (= absolutely agree).
145
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 146
A3. Pairwise Comparisons for the Interactions between Cultural
Background, Educational Level and Ideology
Results refer to the 3 (Culture: German vs. Muslim vs. Other) × 2 (Educational
level: Students vs. Pupils from vocational schools) × 2 (Ideology: Right-wing
vs. Islamic extremist material) depicted in Chapter 10.3.1. Only for significant in-
teractions pairwise comparisons are displayed.
Table A10. Pairwise Comparisons for the Three-way Interaction: SCR
SCR Educational level
Pupil Student
Cultural Background M SE M SE
German Right-wing extre-
mist -1.03a0.19 -1.31a0.11
Islamic extremist -1.31a0.17 -1.22a0.10
Muslim Right-wing extre-
mist -0.99ac 0.26 -0.47c0.22
Islamic extremist -0.45b0.24 -0.76bc 0.20
Other Right-wing extre-
mist -1.20acd 0.31 -0.75bc 0.16
Islamic extremist -1.58ad 0.28 -1.18acd 0.15
Note. Different indices in a column or a line indicate significant differences in the pairwise
comparisons. p< .05. SCR = Slider rating from –4 (= very unpleasant) to +4 (= very
pleasant).
146
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 147
Table A11. Pairwise Comparisons for the Two-way Interaction: Aversion
Cultural Background Aversion
M SE
German Right-wing extremist 2.74a0.07
Islamic extremist 2.73a0.08
Muslim Right-wing extremist 2.60ab 0.12
Islamic extremist 2.89ac 0.13
Other Right-wing extremist 2.67a0.11
Islamic extremist 2.86a0.12
Note. Different indices in a column indicate significant differences in the pairwise com-
parisons. p< .05. Aversion was evaluated on a scale ranging from 1 (= not at all)to5(=very
strong).
Table A12. Pairwise Comparisons for the Two-way Interaction: Shame
Cultural Background Shame
M SE
German Right-wing extremist 1.47a0.04
Islamic extremist 1.35b0.04
Muslim Right-wing extremist 1.65c0.07
Islamic extremist 2.05d0.08
Other Right-wing extremist 1.39ab 0.07
Islamic extremist 1.29ab 0.07
Note. Different indices in a column indicate significant differences in the pairwise com-
parisons. p< .05. Shame was evaluated on a scale ranging from 1 (= not at all)to5(=very
strong).
147
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 148
Table A13. Pairwise Comparisons for the Three-way Interaction: One-sidedness
One-sidedness Educational level
Pupil Student
Cultural Background M SE M SE
German Right-wing extremist 2.38a0.07 2.54a0.06
Islamic extremist 2.41ae 0.06 2.29e0.05
Muslim Right-wing extremist 2.17ab 0.11 1.64c0.12
Islamic extremist 2.03b0.09 1.68c0.10
Other Right-wing extremist 2.42ad 0.12 2.21d0.09
Islamic extremist 2.41a0.10 2.10d0.08
Note. Different indices in a column or a line indicate significant differences in the pairwise
comparisons. p< .05. One-sidedness was evaluated on a four point Likert scale ranging
from 1 (= absolutely disagree) to 4 (= absolutely agree).
148
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 149
Table A14. Pairwise Comparisons for the Three-way Interaction: Interest
Interest Educational level
Pupil Student
Cultural Background MSEMSE
German Right-wing extremist 2.39a0.10 2.14b0.08
Islamic extremist 2.01b0.10 2.40a0.08
Muslim Right-wing extremist 2.47a0.15 2.44ab 0.16
Islamic extremist 3.21c0.15 2.30a0.16
Other Right-wing extremist 2.07a0.16 2.09ab 0.13
Islamic extremist 2.14ab 0.16 2.10abd 0.13
Note. Different indices in a column or a line indicate significant differences in the pairwise
comparisons. p< .05. Interest was evaluated on a four point Likert scale ranging from 1
(= absolutely disagree)to4(=absolutely agree).
149
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 150
Table A15. Pairwise Comparisons for the Three-way Interaction: Persuasiveness
Persuasiveness Educational level
Pupil Student
Cultural Background M SE M SE
German Right-wing extremist 1.97a0.06 1.31b0.05
Islamic extremist 1.67c0.05 1.39bd 0.04
Muslim Right-wing extremist 2.01a0.09 1.57c0.09
Islamic extremist 2.37b0.08 1.42d0.08
Other Right-wing extremist 1.87a0.09 1.40bd 0.07
Islamic extremist 1.88ad 0.08 1.37bd 0.07
Note. Different indices in a column or a line indicate significant differences in the pairwise comparisons. p< .05. Persuasiveness was evaluated on a four
point Likert scale ranging from 1 (= absolutely disagree) to 4 (= absolutely agree).
150
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 151
A4. Words Used in the Single Category Implicit Association Test
The table shows the translated stimuli used in the SC-IAT (see Chapter 11) as well as the amount to which they were associated with
terrorism in the conducted pre-test.
Table A16. Stimuli for the SC-IAT
Terrorism Positive Adjectives Negative Adjectives
Item* M(SD) Item* M(SD) Item* M(SD)
Annihilation 6.3 -0.89 Kind 1 -0.2 Fussy 1.7 -1.17
Explosive charge 6.3 -0.99 Funny 1.1 -0.28 Dependent 1.9 -1.36
Extremism 6.4 -1.04 Sensitive 1.1 -0.44 Pompous 2 -1.58
Bomb 6.4 -0.87 Candid 1.2 -0.37 Fawning 2.2 -1.38
Radicalism 6.4 -0.82 Cheerful 1.2 -0.8 Distorted 2.2 -1.55
Devastation 6.4 -1.23 Charming 1.2 -0.83 Apathetic 2.2 -1.66
Hatred 6.6 -0.77 Caring 1.3 -0.74 Smug 2.2 -1.68
Perpetrator 6.6 -0.64 Content 1.3 -0.99 Unfaithful 2.2 -1.85
Assault 6.7 -0.46 Tender 1.3 -1.25 Phlegmatic 2.2 -1.61
Terrorist 6.9 -0.28 Obliging 1.4 -0.7 Mincing 2.3 -1.74
Note. * Items were translated for publication purposes. Items were rated on a 7-point-Likert ranging from 1 (= absolutely not applying to terrorism), to 7
(= absolutely applying to terrorism).
151
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 152
Tables
Page
Table 1. Categorization Approaches of Islamic Extremist Material ....... 23
Table 2. Structures of Islamic and Right-wing Extremist Propaganda ..... 26
Table 3. The Research Foci and their Realization during the Project ..... 43
Table 4. Frequency of Videos within each Extremist Format Category .... 48
Table 5. Prototypical Scenes for the Talking Head Formats ............. 52
Table 6. Prototypical Scenes for the Reality Clip Formats .............. 54
Table 7. Prototypical Scenes for the Movie and Extreme Clip Formats .... 56
Table 8. Items and Factors of the Immediate Evaluation of Propaganda . . . 62
Table 9. rANOVA Results for the Online Measures for the within Factor Video
Split by Ideology.............................................. 63
Table 10. Correlations between Online Measures and Post-hoc Factors . . . 65
Table 11. rANOVA Results for the Post-hoc Measures within Factor Video
Split by Ideology.............................................. 66
Table 12. Justifications of Terrorism Scale.......................... 74
Table 13. Personality and Attitudinal Variables Assessed............... 76
Table 14. Total Number of Participants Split by Cultural Background and
Educational Level............................................. 77
Table 15. Percentages of Videos Found, Split by Ideology .............. 78
Table 16. Average Manifestation of Assessed Recipient Variables ........ 80
Table 17. Results of the Mixed ANOVAS for SCR and the Post-hoc Measures 82
Table 18. Partial Correlations for Personality Factors and Attitudinal
Patterns .................................................... 91
Table 19. Hierarchical Regressions for both Ideologies on the Reported
Aversion .................................................... 94
Table 20. Hierarchical Regressions for both Ideologies on the Reported
Shame ...................................................... 95
Table 21. Hierarchical Regressions for both Ideologies on the Slider-ratings
(SCR) ...................................................... 97
Table 22. Hierarchical Regressions for both Ideologies on the Reported One-
sidedness ................................................... 98
Table 23. Hierarchical Regressions for both Ideologies on the Reported
Interest ..................................................... 100
Table 24. Hierarchical Regressions for both Ideologies on the Reported
Persuasiveness ............................................... 101
Table 25. Selection of the most One-sided and Persuasive Videos for Study III
Split by Ideology.............................................. 107
Table A1. Factor Loadings of Items on the Cognitive Post-hoc Factors:
Within Level ................................................. 138
Table A2. Factor Loadings of Items on the Cognitive Post-hoc Factors:
Between Level................................................ 139
152
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 153
Table A3. Factor Loadings of Items on the Emotional Post-hoc Factors:
Within Level ................................................. 140
Table A4. Factor Loadings of Items on the Emotional Post-hoc Factors:
Between Level................................................ 141
Table A5. Pairwise comparisons refer to the 6 (Right-wing extremist video
clips) respectively 7 (Islamic extremist video clip) rANOVA displayed in
Chapter 9.4.3................................................. 142
Table A6. Pairwise Comparisons for Online and Post-hoc Ratings: Right-
wing Extremist Videos ......................................... 143
Table A7. Pairwise Comparisons for Online and Post-hoc Ratings: Islamic
Extremist Videos.............................................. 144
Table A8. Pairwise Comparisons for Post-hoc Ratings: Right-wing Extre-
mist Videos .................................................. 145
Table A9. Pairwise Comparisons for Post-hoc Ratings: Islamic Extremist
Videos...................................................... 145
Table A10. Pairwise Comparisons for the Three-way Interaction: SCR. . . . 146
Table A11. Pairwise Comparisons for the Two-way Interaction: Aversion. . 147
Table A12. Pairwise Comparisons for the Two-way Interaction: Shame . . . 147
Table A13. Pairwise Comparisons for the Three-way Interaction: One-
sidedness ................................................... 148
Table A14. Pairwise Comparisons for the Three-way Interaction: Interest . 149
Table A15. Pairwise Comparisons for the Three-way Interaction: Persua-
siveness..................................................... 150
Table A16. Stimuli for the SC-IAT ................................ 151
153
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
{luchterh_neu}20130001_BKA_Propaganda/kap01.3D 1 08.05.2013 S. 154
Figures
Page
Figure 1. Assumed factors influencing the immediate evaluations of
extremist propaganda .......................................... 41
Figure 2. Formats of extremist propaganda videos either in German,
with German subtitles, or without text . . . . . . . . . . . . . . ............... 46
Figure 3. The distribution of single formats within the two ideologies
analyzed.................................................... 50
Figure 4. Experimental setting during video reception. . ............... 57
Figure 5. Procedure of Study I and Study II. . . . . . . . . . ............... 58
Figure 6. GSR and SCR measures for the different videos split by
ideology .................................................... 64
Figure 7. Adverse post-hoc evaluations for the different videos split by
ideology .................................................... 67
Figure 8. Approaching post-hoc evaluations for the different videos split
by ideology. . . . . . . . .......................................... 69
Figure 9. Absolute number of recalls per video split by ideology ........ 71
Figure 10. Perceived influence of the mass media . . . . . ............... 79
Figure 11. Slider-ratings (SCR) in the subsamples split by ideology . . . . . . 83
Figure 12. Adverse post-hoc evaluations for the different subsamples split
by ideology. . . . . . . . .......................................... 84
Figure 13. Approaching post-hoc evaluations for the different subsamples
split by ideology. . . . .......................................... 86
Figure 14. Results for the justifications of terrorism among pupils split
by cultural background and ideology . . . . . . . . . . . . . . . ............... 89
Figure 15. An example of a possible associative network for terrorism-
related concepts . . . . .......................................... 105
Figure 16. Response latencies for the terrorism-related words when sharing
a key with either the positive or the negative words . . . . ............... 109
154
Satz-Offizin Hümmer BKA, Propaganda 2.0 Daten 2. Revision
