Pledge campaigns to encourage charitable giving: a randomised controlled trial

Article (PDF Available) · January 2010 with 125 Reads
Abstract
This paper reports on a randomised controlled trial on the effects of pledging. The research was undertaken in Manchester, in partnership with the Community HEART charity. 11,812 households in two electoral wards were sent information about an upcoming charity campaign to develop school libraries in South Africa: they were told that in a few weeks they would be asked to donate a children's book. Households were randomly assigned to receive differently worded requests. The trial tested: firstly whether asking people to pledge makes it more likely that they will later donate to a charitable campaign and secondly whether people are more likely to pledge and later donate if they are told their involvement will be made public. There is a limited amount of research on pledging and from the available research it is difficult to know whether pledging works or not: pledges are usually invited as part of a wider publicity campaign, making it difficult to identify the effect of the pledge on its own. In this paper we review the available literature on pledging, describe the research design and methods and present some very early results.
Pledge campaigns to encourage charitable giving: a
randomised controlled trial
Sarah Cotterill, Peter John and Liz Richardson
Institute for Political and Economic Governance (IPEG)
School of Social Science
University of Manchester
sarah.cotterill@manchester.ac.uk
0161 275 0792
Political Studies Association Annual Conference, Edinburgh, March 29th-April 1st
2010, Public Administration Specialist Group Panel: Contemporary Challenges to
Public Administration - further international comparisons
Copyright PSA 2010
2
Abstract
This paper reports on a randomised controlled trial on the effects of pledging. The
research was undertaken in Manchester, in partnership with the Community HEART
charity. 11,812 households in two electoral wards were sent information about an
upcoming charity campaign to develop school libraries in South Africa: they were
told that in a few weeks they would be asked to donate a children’s book. Households
were randomly assigned to receive differently worded requests. The trial tested: firstly
whether asking people to pledge makes it more likely that they will later donate to a
charitable campaign and secondly whether people are more likely to pledge and later
donate if they are told their involvement will be made public. There is a limited
amount of research on pledging and from the available research it is difficult to know
whether pledging works or not: pledges are usually invited as part of a wider publicity
campaign, making it difficult to identify the effect of the pledge on its own. In this
paper we review the available literature on pledging, describe the research design and
methods and present some very early results.
Copyright PSA 2010
3
Introduction
Pledging offers something special for the person who wants to do something
positive for society. Partly the act of commitment feels good in itself. But there is a
bigger advantage. By committing to do the act the person may feel they are more
likely to carry it out. It appeals to the notion that people have good intentions to help
the wider society, but they sometimes fall short of their altruistic objectives, forgetting
to do something or feeling too busy as time passes. Pledging gets the person to where
they want to be, acting as a kind of credible commitment. And there is good reason to
think that this appeals to particular psychological processes (Bator and Cialdini 2000:
536). Compliance with the original commitment can be enduring, even if they are
called upon to act by a different person and some substantial time later (McKenzie-
Mohr & Smith 1999). So it is no surprise that charities and now public agencies
adopt this strategy to encourage people to do things, like give money or carry out a
civic act.
But of course it is hard to know whether such a tactic works or is just symbolic.
We know that many people who pledge also carry out what they promise. What we
do not know is whether citizens need the pledge to carry out the promised activity or
whether they would have done it anyway (see Burn and Oskamp, 1986; Ludwig et. al.,
2005; Mckenzie-Mohr and Smith, 2006), which makes a pledge campaign a waste of
time for the charities or government organisations who implement it. If so,
governments and organisations may as well spend their time asking people directly,
using persuasion or providing information, rather than waste energy seeking advance
commitments or pledges. People might be spared the guilt of being asked to pledge.
The claim of this paper is that only an experiment that randomises between
asking for help and asking plus pledging can evaluate the independent effect of a
pledge. To that end, we report such an experiment in Manchester, in partnership with
the Community HEART charity, testing whether asking for pledges makes it more
likely that people will later donate to charity. The paper first discusses the current
policy interest and research literature on pledging, making references to psychological
studies. It then moves on to outline the study design and discuss the lessons from the
implementation of the experiment in the field.
Pledging in the UK1
There are a number of pledge schemes in the UK. “A pledge scheme is an
invitation from an organisation to an individual to make a public commitment to a
behaviour change [in relation to climate change]” (Defra, 2008: 3). By far the greatest
volume of pledge schemes concern environmental issues, with many local authorities
and others now running some form of pledge scheme where individuals can commit
to one or more sustainable behaviours. Other individual pledge schemes include
protest campaigns, promises to be vegetarian or vegan, schemes aimed at young
people and a few local pledges. There are two national generic pledge sites, covering
a wide range of pledges. PledgeBank allows users to set up pledges and then
1 This section draws on a literature review on pledging, conducted by the authors for Communities and
Local Government (Cotterill and Richardson 2009)
Copyright PSA 2010
4
encourages other people to sign up to them (www.pledgebank.com). We Are What
We Do puts forward 130 actions on a website that individuals can sign up to do
(www.wearewhatwedo.org/).
Pledge schemes can be set up for a number of different purposes. Four
common objectives of pledge schemes are:
• Reinforcing the need to change by increasing awareness and changing
attitudes
• Changing behaviour
• Building a case to persuade others
• Collecting data on individuals
Source: Defra (2008)
There are few evaluation reports of pledge schemes, so it is hard to judge the
success of the schemes in attracting people to make pledges. The following points are
largely based on looking at the limited information available on pledging websites.
High profile campaigning and publicity can successfully promote the
opportunities to pledge. When Manchester is My Planet was launched in 2005,
£160,000 was spent on a high profile and intensive campaign over three months,
including website, branding, pledge cards, posters, campaign resource packs and
media coverage, designed to get 10,000 citizens to pledge on climate change. The
campaign culminated in a major event at Manchester Town Hall to celebrate the
achievement of the first 10,000 pledgers. The pledge campaign continued with a
lower budget and numbers continued to increase, but at a slower rate, gaining a
further 10,300 pledgers over the next two and a half years, to reach a total of 20,300.
The Energy Saving Trust attracted a total of 216,997 commitments to its Save
your 20% campaign and CRED’s Community Carbon Reduction Programme has
attracted 53,611 pledges. Both have a clear focus on environmental sustainability and
appear to be well-resourced and well-thought out campaigns with attractive websites.
The earliest pledge deadlines on Pledgebank are dated March 2005. In the four
years since then, 1,039 pledges have been created. 445 were successful in attracting
the number of pledgers required by the originator (including eight that are still open to
pledgers) and 575 failed. There are currently 19 “live” pledges that are actively
seeking signatures.
Table 1: Statistics on use of Pledgebank (25March 2009)
Number of pledges Deadline for signing
Current pledges that need
pledgers 19
28/03/09 – 01/11/11
Successful open pledges 8
30/04/09 – 09/02/11
Successful closed pledges 437
18/03/05 – 09/03/09
Failed pledges 575
14/08/05 – 22/03/09
Copyright PSA 2010
5
If the pledges have been created at a steady rate throughout the life of the site
– and we do not know if this is true – there have been on average 21 pledges posted a
month. The number of signatures requested for the pledges that are live varies from
10 to 1,000 and the mode is 20.
The We Are What We Do site has an action tracker so individuals can keep a
list of what they have pledged to do and go to the site to indicate each time they do
the action. It also lists where the individual is in a scale of activity on the site. For
example, for “Smile and Smile Back”: 6,468 individuals have signed up to do this and
it has been completed 101,243 times, the most prolific person has done it 25,624
times, and some people who signed up have not done it at all. For “Take public
transport whenever you can” 4,008 individuals have signed up to do this and it has
been completed 32,557 times, the most prolific person has done it 2,000 times, and
some people who signed up have not done it at all. The person who came in top at
2,000 times had also claimed to do a list of other actions at a similarly high rate.
These inflated figures suggest a need to be sceptical about self-reported claims
relating to completed pledges.
In addition to attracting pledgers, another objective of pledge schemes might
be to raise awareness in a wider population, not just among those who pledge. This
could be particularly important in sending out a positive message to those who have
not yet pledged; for example, “30 per cent of residents in your area have already
pledged to do x”. It is difficult to achieve a wide awareness this type of community
activity. A study of incentives found that there was very low awareness among local
residents of a Good Neighbour Scheme, where residents were asked to sign up to a
local charter, although those who had heard of it thought it was helpful in a minor
way. Almost 60 per cent of residents had never heard of it, and only 15 per cent had
heard a lot about it, with the remaining 25 per cent being unsure of what it was.
(Bastow et al 2007). A recent evaluation of Community Contracts found a similar low
level of awareness among local residents: people in Community Contract areas often
were not even aware of the obligations of residents under the Contract, let alone
changing their behaviour (IPEG 2010).
The Psychology of Pledging
Several theories from psychology suggest that in certain circumstances people
who pledge are likely to act on their good intention. Cognitive dissonance theory tells
us that people feel discomfort if there is an inconsistency or dissonance in their
understandings and they have strong motivations to avoid holding inconsistent views.
Inconsistency between an individual’s self view and their behaviour creates a
dissonance which people want to resolve (Festinger 1957; Aronson 1969).
Consistency is an important character trait, with people who behave inconsistently
being widely regarded as unreliable and untrustworthy. There is a strong internal
pressure on individuals to behave in a way that is consistent with how they see
themselves. Individuals who commit themselves to a particular behaviour can come to
see themselves in a way that is consistent with that behaviour, leading to long term
change in their attitudes and behaviour. The commitment can act as a catalyst,
providing the internal conviction for a new identity and leading to behaviour that
corresponds with that conviction, which can last well beyond the duration of the
Copyright PSA 2010
6
commitment. So, if an individual gives a commitment that they will volunteer, vote,
recycle or not drop litter, it increases the likelihood that they will later act in a way
that is consistent with those attitudes. “When individuals feel committed to a certain
type of behaviour, they will often adopt an identity that is consistent with that
behaviour, the result of which frequently is long-lasting behaviour change” (Bator and
Cialdini 2000: 536). Their compliance with the original commitment can be enduring,
even if they are called upon to act by a different person and some substantial time
later (McKenzie-Mohr & Smith 1999). The likelihood of a commitment leading to
long-lasting change will vary according to the nature of the pledge: change is more
likely if the commitment is voluntary, made in public and relates to an issue the
pledger is already concerned about. Of course the quest for consistency may also lead
people not to pledge or donate because they do not see this as consistent with their
usual behaviour.
Research on Pledging
A number of research studies have examined whether making a pledge or
commitment makes it more likely that the pledged action will be carried out. The
results of these studies are somewhat conflicting. The largest number of studies on
commitments have focussed on doorstep recycling. Securing pledges through direct
personal contact worked better than securing pledges through indirect contact or
educational information alone, but the studies did not compare pledge and non-pledge
methods (Reams and Ray 1993; Bryce et al. 1997). Asking for a pledge or
commitment works equally well in encouraging recycling as receiving a persuasive
leaflet (Burn and Oskamp 1986) or being offered a reward (Katzev and Pardini 1987).
A more recent report compared canvassing campaigns with and without pledges and
found that the pledge made no significant difference (Thomas 2006). Overall, the
message from recycling research is that asking people to pledge to recycle can raise
recycling rates if it is done through a personal approach on the doorstep, and it will
raise recycling at a similar rate to other alternative approaches, but it is not clear
whether it is the personal contact or the pledging that persuades households to recycle.
There are some studies which have found that pledging is successful, but the
pledge was part of a wider promotional campaign, making it difficult to assess the
particular contribution made by the pledge. A pledge campaign to encourage cyclists
to wear helmets was successful in raising the use of helmets, but participants were
provided with information and a voucher while being asked to pledge, so it is difficult
to separate out the different effects (Ludwig 2005). Similarly, use of car safety belts
rose among those who signed a pledge, but they were also provided with a card to
hang in their car as a reminder and entered into a prize draw, so, again, it is hard to
separate out the pledge effect (Geller 1989).
A US research paper examined the impact of pledging on voter registration
and voter turnout. In one experiment, students were contacted by telephone with
information about how to register to vote. Half were then randomly allocated to a
treatment group and were asked whether they were planning to register to vote. The
other half were allocated to a control group and not asked the additional question. The
proportion who did register was higher among the treatment group, who were asked
for a commitment, than in the control group, who received information about
registration but were not asked for a commitment. A further experiment where
students were asked to state whether they would turn out and vote had similar results:
Copyright PSA 2010
7
a higher proportion of the group who were asked for a commitment voted, compared
with a control group who received information about voting (Greenwald et al 1987).
This suggests that being asked for a commitment can have a positive effect on voter
registration and voter turnout, but it does not compare the commitment approach with
any other method of mobilisation.
A US campaign to encourage voter turnout asked young people at rock
festivals to complete one of two postcards, “I will rock the system by exercising my
right to vote” or “I will vote because….”. The two differently designed postcards
were used at different times, so there was no random allocation of the two groups.
The cards were posted back to the young person a week before the presidential
election. People who had entered their own pledge were more likely to see it through
and turn out to vote than those who had completed the generic pledge (Burgess et al
2000). This does not say whether pledging works, but does indicate that people are
more likely to carry out the action if they have been allowed to personalise the pledge.
Research with 142 smokers who all completed a written pledge to abstain
from smoking for one hour a day over a month found that whether people were heavy,
moderate or light smokers made no difference to them keeping to the pledge. There
was no control group. People were more likely to successfully keep their pledge if
they had already expressed a desire to quit smoking or reduce their level of
consumption. There was no difference between males and females, but younger
smokers (under 21) found it harder to stick to the pledge (Hallaq 1976). This suggests
that people are more likely to keep to a pledge if it is something they were already
thinking about before they were asked to pledge.
Action for Sustainable Living (AfSL), a Manchester based charity which uses
pledges as part of their work to encourage sustainable behaviour, conducted phone
and email interviews with 104 of the 2,400 people who had previously pledged to
sustainable behaviour as part of a face-to-face meeting with an AfSL staff member or
volunteer. 95 per cent of people said that as a result of their contact with AfSL they
were now doing more than before to reduce their environmental impact, and 30 per
cent were doing much more (Boyd 2008). The pledging was undertaken as part of a
one to one meeting providing information and advice about sustainability, so it is hard
to separate out the particular effect of the pledge. People found it easier to stick to
their pledge if it was easier to implement – e.g. 79 per cent of pledgers managed to
take all their unwanted clothes to charity shops – but on the most challenging pledges
like “generate my own energy” none of the pledgers had achieved it. People are more
likely to choose pledges that are less challenging. Less challenging pledges are more
likely to be implemented than pledges that are more challenging.
Developing a successful pledge campaign
Community-based social marketing (CBSM) is an approach, which brings
together psychological theory with theories of marketing to develop practical tools to
foster sustainable behaviour (McKenzie-Mohr and Smith 1999 and see review of
CBSM in Jackson, 2005). Community-based social marketing has been adopted in a
number of projects relating to sustainable behaviours including water use, recycling,
composting and energy use. CBSM has found that people are more likely to stick to
Copyright PSA 2010
8
their commitments if they are written down and made in public. Group commitments
can be effective in well-established and cohesive groups where individuals care how
they are viewed by others. Actively involving the person in the issue helps to increase
commitment. Using community “block leaders” – local people who already engage in
the behaviour – is an effective and cheap way of seeking commitment; asking people
who have already committed to approach their neighbours can be effective in
changing the behaviour of both. Commitments should be voluntary, and only sought
for behaviours in which people express an interest. Within the Community-Based
Social Marketing perspective, commitment approaches work best when combined
with other tools to change behaviour, and commitment on its own is unlikely to work.
The other tools that encompass the community-based social marketing approach are
prompts, social norms, good quality communication, incentives and removing
external barriers.
The design and evaluation of any behaviour change scheme are crucial: the
lessons from community-based social marketing are that the following four steps
should be taken before introducing a new pledge campaign:
1) Selecting behaviours and identifying barriers. It is important to: identify the
target population and understand their background attitudes and behaviour;
identify the behaviours to be targeted by the scheme and prioritise which
behaviours to focus on; and identify the potential barriers to behaviour change.
2) Designing the programme. Select which other behaviour change tools might
be most useful to use alongside the pledge: prompts or retrieval cues such as
stickers, lapel badges, window posters, fridge magnets; building social norms
(e.g. discussion forums and street parties); communication (e.g. through local
radio, leaflets and websites) to satisfy the need for well-placed positive
messages from a credible source; incentives (e.g. prizes and rewards);
removing any external barriers where possible. It is important to check that the
right behaviours are being targeted and the messages are appropriate to the
population through focus groups or surveys and utilising existing data or local
knowledge; and to link the message to personal experience as far as possible
to appeal to people in a way that evokes their emotion, triggers their
imagination and is immediate to them.
3) Pre-test/pilot the pledge campaign and compare with a control group.
4) Implementation and evaluation.
(adapted from McKenzie-Mohr and Smith 1999 and Bator and Cialdini 2000)
Research Design and Methodology
Our approach was to design a trial that would evaluate a pledge, and also
create some of the extra conditions that that have been highlighted in the literature, in
particular that a pledge will be more successful if it is public. We addressed this issue
by varying the treatment groups accordingly. A sample population of individuals were
randomly assigned to one of three groups. Members of all three groups were sent
Copyright PSA 2010
9
information advertising an upcoming charity campaign to develop school libraries in
South Africa: they were told that in a few weeks they would be asked to donate a
second hand children’s book. Treatment group one was asked to return a pledge card
promising to make a donation. Treatment group two was asked to return a pledge card
and advised that a list of donors will be publicly advertised. The control group simply
received a campaign letter without being asked to pledge. Some weeks later all
participants were asked to donate a book. The outcome measure was whether a book
donation was received. The project is illustrated in Appendix 1.
Research objectives
We were interested to discover whether making a pledge encourages people to
carry out a civic activity: are those who make a pledge to do something more likely to
later carry out the activity, compared to people who were not asked to pledge? We
expect that households who are invited to make a pledge are more likely to later
donate a book, because they will feel they have made a commitment or promise and
want to see it through. We set this out more explicitly below:
H1 The pledge group will donate more books than the control group.
H2 The pledge group and the pledge&publicity group (combined) will donate
more books than the control group.
Households who are advised their donation will be made public will be
encouraged to donate because their generosity will be advertised to their peers, (but a
minority might be discouraged by the publicity).
H3 The pledge&publicity group will donate more books than the pledge group.
Community HEART
The research was undertaken in partnership with the Community HEART
charity. Community HEART is a UK registered charity which supports local self-help
initiatives in South Africa (registered charity number 1052817). They collect
children’s books in the UK and transport them to South Africa, where they are used to
set up school libraries. Further details can be found at http://www.community-
heart.org.uk/projects/books/books.htm
The Intervention
All households were sent two letters about the book collection. The initial
letter was sent in late January 2010, addressed to “The Residents” and sent on
University of Manchester letterhead, including the logo of Community HEART. It
contained some information that was identical across all three groups: details of
Children’s Book Week, to be held Saturday 27th February – Saturday 6th March 2010;
a request to donate a book for schools in South Africa; information about Community
Heart; a statement saying the resident would be contacted again with details of local
drop-off points. The initial letter to the Control group contained no additional
information. The initial letter to the Pledge group contained all the above information
and asked the residents to pledge to donate a book. They were given information on
how to pledge, by phone, email or postcard, and were provided with a pledge card.
Copyright PSA 2010
10
The initial letter to the Pledge plus publicity group included all the same information
as the Pledge group, plus an additional statement, “A list of everyone who donates a
book will be displayed locally”.
We sent a second letter to each household four weeks later, in mid-February
2010. All households in the control group plus households who were in either of the
pledge groups but had not pledged were sent a letter reminding them of the children’s
book collection and advising them of the local drop-off points for their book. The
letter was addressed to The Residents unless they had been in touch for some reason
to give their name. All those who had pledged were sent a letter thanking them for
their pledge, reminding them of the children’s book collection and advising them of
the local drop-off points for their book. The letter was addressed to them by name if
they had given their name when pledging. Where people had pledged by email or
pledge card, their pledge was returned to them as a reminder. All letters to the
Pledge&Publicity Group additionally included the statement “A list of everyone who
donates a book will be displayed at the drop-off points afterwards, to say thank you”.
All households were encouraged to put their book in a bookbag which
accompanied the second letter. Each bookbag was labelled with the dates of the
children’s book week, the local drop-off points, a unique identifier for each household
and a space for the resident to write their name and address. The second letter asked
households who wanted to donate more than one book, to put them in a carrier bag,
and place the bookbag inside their own bag. Spare bookbags were available at the
drop-off points.
Three local drop-off points were available in each electoral ward (six in total).
They were chosen for their geographic spread across the ward and to offer variety to
appeal to different people. They were a library, school and café in one ward and a
library, community centre and children’s centre in another. Posters were displayed in
each drop-off point during the week prior to the book collection to reinforce the
message and encourage book donations.
Outcome Measurement
There are two outcome measures of interest: pledging and book donation.
Pledging was measured by whether a pledgecard, email or phone call was received
from a household promising to donate a book. If a pledgecard was returned with a
different address than it had been sent to, it was listed as a pledge from the original
address. We assume that either the householder has moved elsewhere and used a card
from their previous address, or that the householder has passed their card onto a friend
or relative who can donate.
All book donations that were received in a bookbag were attributed to the
household, using the unique identifier on the bookbag. Any book donations that
included an address were similarly attributed to the appropriate household. Donations
with addresses outside our study area were assumed to be from passers-by or people
who had heard about the scheme (e.g. staff or users of the drop-off points). Donations
without any identification could potentially be from a) householders in our study who
chose to remain anonymous; b) households in our study who did not understand the
instructions or forgot; c) households in our study who had already filled our bookbag
Copyright PSA 2010
11
and were donating additional books; d) people who were not part of the study but had
heard about it, saw the collection box, staff or users of the drop-off points).
Households were not informed that they were taking part in a research
study. All three groups were blind to the fact that their book donations or pledges
were being monitored. No personal data was collected during the study other than
those people who chose to provide their names or contact details. All data was kept
securely, destroyed once the mailings had been completed and not used for any other
purpose. Staff at the drop-off points and any resident who asked about the university’s
involvement was honestly informed that this was a research study looking at the
effectiveness of different messages in encouraging charitable giving. Staff were asked
to encourage people to use the bookbags, so we could identify where donations came
from. The data entry of the book donations was done by a researcher who was blind to
the study in the sense that she had no earlier involvement in the project and was
unaware which group households were in or whether they had made an earlier pledge.
Pilot study
A pilot study was undertaken in one of the two electoral wards in September
2009. The purpose of the pilot was to test out the letters, pledgecards and drop-off
arrangements; to see if a larger experiment was viable. It included 163 households in
8 randomly selected postcodes. The households were randomly allocated to one of
three groups: a control group who were sent 2 letters asking for a donated book or
mobile phone; a pledge group who were sent similar letters with a pledgecard; a
second pledge group who had postage paid on their pledgecard. As a result of the pilot,
some changes were made to the final study: the working of the letters and pledgecards
were refined; the opportunity for phone and email pledges were added, as well as
pledgecards; larger book bags were provided; postage was not paid on the pledgecards;
residents were asked for one book, and also advised of how to donate more than one
book if preferred; donations of mobile phones were not requested, because of concern
at drop-off points about security and the growing number of adverts for recycling
phones elsewhere for cash. The numbers were too small to make any estimates of
response rates for the final study.
Sample population and randomisation
We were given a list of all the postcodes and associated lower super output
area codes in two electoral wards of Manchester. We used an address finder to
identify all the addresses associated with these postcodes, excluding all business
addresses. This resulted in 11,812 households, 5851 in one ward and 5961 in the other.
We asked two direct marketing companies what response we might get if we
asked people on their lists to give a book for charity (i.e. without any pledge): Listlab
guessed at a response rate between 1% and 5%. Cameo Lifestyles guessed at 0.5% –
3%. In the pilot 3% of the pledge group donated books and 0% of the control group
(but the numbers were too small to make any assumptions based on this). We
undertook power calculations using the DSS Calculator
(http://www.dssresearch.com/toolkit/spcalc/power_p2.asp). We estimated that in one
of the wards, with approximately 1900 in each group, we would have 99% of
statistical power to detect a 2% difference between a control group of 1.5% and a
treatment group of 3.5%. If the difference was closer, with the same group sizes, we
Copyright PSA 2010
12
would have 71.1% of statistical power to detect a 1% difference between a control
group of 1.5% and a treatment group of 2.5%. We eventually undertook the research
in two wards, so the number of households is double those our estimates were based
on.
The two datasets were separately randomised by Professor David Torgerson
at York Trials Unit into three treatment groups - divided equally using the SPSS
random selection function. We checked that within each group the randomisation had
been executed properly and the groups were more or less equal according to lower
super output area.
Table 2. Allocation of Households to Group
Group Ward D
Ward W
Control (0) 1950
1987
Pledge (1) 1950
1987
Pledge & Publicity (2) 1951
1987
TOTAL 5851
5961
Data and Variables Used in the Analysis
Household Level Variables: Group (Control, Pledge or Pledge & Publicity);
Pledge (No pledge made, Pledged); Pledgetype (No pledge made, Pledgecard, Phone,
Email - create separate dummies for each type, with 1 for the group 0 for all else);
Other levels of variable: Ward, Super Output area, Drop off point
The outcome variables are Book (No donation, Book donated) and Pledge.
We will undertake some preliminary descriptive statistics including
frequencies, cross tabs and correlations, plus independent samples T-Tests:
Comparison Group 1 Comparison Group 2
T-Test 1 (tests H1) Control Pledge
T-Test 2 (tests H2) Control Pledge + Pledge&Publicity
T-Test 3 (tests H3) Pledge Pledge&Publicity
We will run these tests on the whole dataset and also separately for each of the
two electoral wards. We think it is appropriate to use a one-tailed test for tests 1 and
2 because we are confident that we do not expect the pledge to reduce participation.
We are less confident about test 3 and may therefore use a two-tailed test, which is
appropriate when our hypothesis is that there could be a difference in any direction.
We will undertake Use Multilevel Logistic Regression in Stata, using the
combined dataset.
Dependent variable - Book (binary)
Independent variables –
Household level: Group (need to create dummys – 3 groups)
Copyright PSA 2010
13
Level 1 (SOA): % on benefits (or similar)
Level 2 (Ward): Ward
NB Level 1 and 2 are contiguous. The level 1 and level 2 variables are control
variables.
Intention to treat analysis was adopted: all of the households who were
included in the initial randomisation were included in the analysis, regardless of
whether or not they received the intervention. We know that 81 letters were returned
to us by Royal Mail because the property was empty, 51 letters were returned that
were wrongly addressed, and 4 were sent back by residents who were not interested.
These addresses are spread across all three groups and there is no pattern to them that
can adversely affect the experiment.
Preliminary Results
The book collection week is underway at the time of writing, so we can only
report interim results. Further findings will be presented at the conference.
1. Pledges received
Overall the numbers pledging are higher than we anticipated. So far, we have
received 374 pledges (4.8% of those asked to pledge). There is a higher response
among those who have been promised publicity (199 pledges – 5.1% of those offered
publicity) than those who were not (175 pledges – 4.5% of those asked to pledge). We
are still receiving pledges.
2. Enquiries about Book Collections
Eight people have contacted us who want to arrange their own book
collections: 2 primary schools, a church, a parent wants to organise a collection in
daughter’s school, a parents’ support group in a Surestart centre, a nursery, staff in a
mental health trust and Body Shop staff. We have put all these people in contact with
Community HEART.
Acknowledgements
The research study is part of the Rediscovering the Civic and Achieving Better
Outcomes in Public Policy project (see www.civicbehaviour.org.uk), funded by the
Economic and Social Research Council, Department of Communities and Local
Government and the North West Improvement Network, grant reference: RES-177-
025-0002. The authors wish to thank Community HEART, staff at the drop-off points,
volunteers and colleagues at the University of Manchester who helped with the book
collection.
Copyright PSA 2010
14
References
Aronson, E. (1969) “The Theory of Cognitive Dissonance: A Current Perspective.” In
Advances in Experimental Social Psychology, ed. L. Berkowitz, vol. 4, pp. 1–
34. San Diego: Academic Press.
Bastow, S., Beck, H., Dunleavy, P. and Richardson, L (2007) ‘Incentives Schemes
and Civil Renewal’, in Brannan, T, John, P and Stoker, G. (eds) Re-Energising
Citizenship, Strategies for Civil Renewal, Basingstoke: Palgrave Macmillan.
Bator, R. J. and Cialdini, R.B. (2000) ‘The Application of Persuasion Theory to the
Development of Effective Proenvironmental Public Service Announcements’,
Journal of Social Issues 56, 3 527 – 541, p536.
Boyd, A. (2008) Action for Sustainable Living (Manchester) evaluation report to
Defra.
Burgess, D. Haney, B, Snyder, M., Sullivan, J. L. & Transue, J. E. (2000) ‘Rocking
the Vote: Using Personalised Messages to Motivate Voting Among Young
Adults’, Public Opinion Quarterly, 64 29–52.
Bryce, W. J. et al. (1997) ‘Commitment Approach to Motivating Recycling: New
Zealand Curbside Trial’, The Journal of Consumer Affairs, 31, 1 27–52.
Burn, S. M. and Oskamp, S (1986) Increasing Community Recycling with Persuasive
Communication and Public Commitment, Journal of Applied Social
Psychology, 16, 1, 29-41.
Cotterill, S. and Richardson, L. (2009) Can Pledging Increase Civic Activity: A
Literature Review on Developing Community Pledgebanks. London: CLG
http://www.communities.gov.uk/documents/communities/pdf/1328160.pdf
Department for Environment, Food and Rural Affairs (Defra) (2008) Best Practice
Guide to Designing and Operating Climate Change Pledge Schemes.
Festinger, L. (1957) A Theory of Cognitive Dissonance. Palo Alto, CA: Stanford
University Press.
Geller, E.S. et al. (1989) ‘Promoting Safety Belt Use on a University Campus: An
Integration of Commitment and Incentive Strategies’, Journal of Applied
Social Psychology, 19, 1 3–19.
Greenwald et al. (1987) ‘Increasing Voting Behavior by Asking People if The Expect
to Vote’, Journal of Applied Psychology, 72, 2 315–318.
Hallaq, J. H. (1976) ‘The pledge as an instrument of behavioural change’, The Journal
of Social Psychology, 98, 147–148.
IPEG (2010) An Evaluation of the Community Contracts Pilot Programme, report to
the Department of Communities and Local Government, London: CLG.
http://www.communities.gov.uk/publications/communities/evaluationcommun
itycontracts
Copyright PSA 2010
15
Jackson, T. (2005) Motivating Sustainable Consumption. Centre for Environmental
Strategy, University of Surrey.
Katzev, R.D. and Pardini, A.U. (1987) ‘The comparative effectiveness of reward and
commitment approaches in motivating community recycling’, Journal of
Environmental Systems, 17, 2, 93–114.
Ludwig, T. D. et. al. (2005) Using Social Marketing to Increase the Use of Helmets
Among Bicyclists. Journal of American College Health. 54, 1 51-58
McKenzie-Mohr, D. & Smith, W. (1999) Fostering Sustainable Behavior: An
Introduction to Community-Based Social Marketing. Gabriola Island, Canada:
New Society Publishers.
Policy Studies Institute (2008) London Student Pledge Evaluation
Reams, M. A. & Ray, B. H. (1993) ‘The Effects of 3 Prompting Methods on
Recycling Participation Rates – a Field Study’, Journal of Environmental
Systems, 22, 4 371–379
Thomas, C. Open University (2006) Recycle for Hampshire – Campaign Evaluation
Report.
Copyright PSA 2010
16
Appendix 1 Implementation of the Book collection
POPULATION
CONTROL GROUP
Letter: “Man Uni is
collecting children’s
books and phones for
worthwhile cause.
Please donate. Please
set aside and we will
contact you again in x
wee
ks how to donate”
PLEDGE GROUP
Same letter plus:
“Please return the
enclosed postcard
(or email/web?) to
pledge your
donation”
CONTROL
After x weeks:
send a reminder
letter with
enclosed return
envelope.
NON-PLEDGERS
Same as control PLEDGERS
Send a letter
reminding them
that they pledged
and enclosing
return envelope(s)
Receive books
Send thank you
and summary info
Receive books
Send thank you
and summary info
Receive books
Send thank you
and summary info
PLEDGE &
PUBLICITY GROUP
Same letter plus:
“Please return the
enclosed postcard (or
email/web?) to pledge
your donation” plus
advised of publicity
NON-PLEDGERS
Same as control PLEDGERS
Send a letter
reminding them
that they pledged
and enclosing
return envelope(s)
Receive books
Receive books
Send thank you
and summary info.
Publicise
involvement
Send thank you
and summary info.
Publicise
involvement
Copyright PSA 2010
Show more

Recommended publications

Discover more publications, questions and projects in Campaigns
Article
The Impact of a Pledge Request and the Promise of Publicity: A Randomized Controlled Trial of Charit...
March 2013 · Social Science Quarterly · Impact Factor: 0.99
Read more
Article
The Impacts of a Pledge Campaign and the Promise of Publicity: A Randomized Controlled Trial of Char...
May 2011 · SSRN Electronic Journal
Read more
Article
The influence of population characteristics on household response to a charity book collection based...
January 2016 · International Journal of Nonprofit and Voluntary Sector Marketing
Read more
Article
Expanding the Use of Experiments on Civic Behavior: Experiments with Local Government as a Research...
February 2010 · The Annals of the American Academy of Political and Social Science · Impact Factor: 1.01
Read more
Discover more
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.
This publication is from a journal that may support self archiving.
Learn more