Does social connection turn good deeds into good feelings? On the value of putting the 'social' in prosocial spending

Abstract

When are the emotional benefits of generous behaviour most likely to emerge? In three studies, we demonstrate that the hedonic benefits of generous spending are most likely when spending promotes positive social connection. Study 1 shows that people feel happier after giving more to charity, but only when they give to someone connected with the cause. Studies 2 and 3 show that the emotional rewards associated with giving to friends or acquaintances are greatest in situations that facilitate social connection. Thus, social connection may be important for turning good deeds into good feelings, and maximising connectedness between givers and recipients may enhance the emotional payoff of charitable initiatives.

Full text

Int. J. Happiness and Development, Vol. 1, No. 2, 2013 155
Copyright © 2013 Inderscience Enterprises Ltd.
Does social connection turn good deeds into good
feelings?: On the value of putting the ‘social’ in
prosocial spending
Lara B. Aknin*
Department of Psychology,
Simon Fraser University,
8888 University Drive, Burnaby,
British Columbia, V5A 1S6, Canada
E-mail: laknin@sfu.ca
*Corresponding author
Elizabeth W. Dunn and Gillian M. Sandstrom
Psychology Department,
University of British Columbia,
2136 West Mall, Vancouver,
British Columbia, V6T 1Z4, Canada
E-mail: edunn@psych.ubc.ca
E-mail: gsandstrom@psych.ubc.ca
Michael I. Norton
Harvard Business School,
Marketing Unit,
Soldiers Field Road, Boston,
Massachusetts, 02163, USA
E-mail: mnorton@hbs.edu
Abstract: When are the emotional benefits of generous behaviour most likely
to emerge? In three studies, we demonstrate that the hedonic benefits of
generous spending are most likely when spending promotes positive social
connection. Study 1 shows that people feel happier after giving more to charity,
but only when they give to someone connected with the cause. Studies 2 and 3
show that the emotional rewards associated with giving to friends or
acquaintances are greatest in situations that facilitate social connection. Thus,
social connection may be important for turning good deeds into good feelings,
and maximising connectedness between givers and recipients may enhance the
emotional payoff of charitable initiatives.
Keywords: happiness; money; prosocial spending; social connection;
well-being; donations; charitable giving; warm glow; social relationships; gift
giving.

156 L.B. Aknin et al.
Reference to this paper should be made as follows: Aknin, L.B., Dunn, E.W.,
Sandstrom, G.M. and Norton, M.I. (2013) ‘Does social connection turn good
deeds into good feelings?: On the value of putting the ‘social’ in prosocial
spending’, Int. J. Happiness and Development, Vol. 1, No. 2, pp.155–171.
Biographical notes: Lara B. Aknin is an Assistant Professor at Simon Fraser
University in Canada. Her research interests include happiness, emotions and
prosocial behaviour.
Elizabeth W. Dunn is an Associate Professor at the University of British
Columbia in Canada. Her research interests include happiness and
self-knowledge.
Gillian M. Sandstrom is a PhD candidate at the University of British Columbia
in Canada. Her research interests include social connections, weak social ties
and well-being.
Michael I. Norton is an Associate Professor of Business Administration at
Harvard Business School. His research focuses on money, happiness and
inequality.
1 Introduction
Cause-related marketing has become increasingly ubiquitous, as one corporation after
another links purchases of its products to charitable donations. Companies want to be
associated with positive acts and the warm glow of giving (Andreoni, 1989, 1990) and
such efforts can pay: socially responsible corporate initiatives can influence consumer
choice and increase sales (Barone, et al., 2000; Lev et al., 2010). Several notable cases
suggest a role for a simple factor that influences that payoff: connectedness between the
giver and the recipient. With the Pepsi Refresh Project, for example, PepsiCo allowed
consumers to propose and vote for projects in their communities (such as cleaning up
local rivers); PepsiCo then funded those projects, linking Pepsi to causes close to
consumers’ homes – and hearts. While previous research has focused on how emotion
affects consumers’ propensity to donate to charity or buy cause-related products (Kogut
and Ritov, 2005; Small and Loewenstein, 2003; Small and Simonsohn, 2008; Strahilevitz
and Myers, 1998), we explore the impact of connection not on the initial decision to give
but on the subsequent emotional experience of the giver. Reaping the largest returns from
prosocial acts requires an understanding of when and why charitable behaviour has the
biggest influence: what kind of giving most alters individuals’ affective responses and
subsequent behaviour?
1.1 The emotional impact of giving
One of the most reassuring aspects of human nature is that people not only behave
generously toward one another in daily life, but often seem to enjoy doing so; for
instance, people reap pleasure from doing volunteer work (Thoits and Hewitt, 2001),
engaging in random acts of kindness (Lyubomirsky et al., 2005b), and spending money
on others (Dunn et al., 2008, 2010). But do good deeds always produce good feelings? A

Does social connection turn good deeds into good feelings? 157
recent review notes that “surprisingly, there is little direct evidence that helping others
actually makes the helper feel good” [Dovidio et al., (2006), p.240]. Indeed, much of the
evidence is mixed. For instance, in one of the first papers examining the emotional
consequences of prosocial behaviour, Harris (1977) showed that helping a confederate
search for a missing item led to an increase in happiness, but providing directions to a
campus administration building did not. Similarly, Williamson and Clark (1989) found
that assisting a confederate led to emotional gains when participants desired a communal
relationship with the confederate, but not when they desired an exchange relationship.
We propose that emotional benefits are most likely when prosocial behaviour
facilitates positive social connection between the giver and recipient – and are less likely
when such interaction is precluded. While engaging in prosocial behaviour may not
always lead to emotional gains, the literature on well-being has demonstrated that strong
social relationships are one of the most robust and reliable predictors of happiness
(Lyubomirsky et al., 2005a). Indeed, quality social relationships appear necessary for
achieving high levels of happiness (Diener and Seligman, 2002; Myers, 2000) and
satisfying humans’ fundamental need to belong (Baumeister and Leary, 1995; Ryan and
Deci, 2000). If social relationships are such a strong predictor of well-being, it seems
possible that prosocial behaviour might increase happiness when it facilitates the
development of social relationships through positive social connection.
Re-examining past research through this lens clarifies the seemingly variable
relationship between prosocial behaviour and happiness. Across investigations, prosocial
behaviour commonly produces hedonic gains when enacted in ways that allow for
increased social connection, such as positive social interaction between the giver and
recipient. As Williamson and Clark (1989) suggested, helping someone search for a
missing item may have provided more happiness than giving directions because the
former experience provides more connection between benefactor and beneficiary.
Furthermore, people tend to experience the largest emotional benefits from writing
gratitude letters when they share the letter with the target (Lyubomirsky, 2008), which
allows for greater positive social connection. Thus, greater emotional payoffs seem to
transpire when good deeds involve social connection. Given the plausible but untested
role of social connection, the present research examines whether social connection is
important for experiencing the emotional rewards of generous behaviour.
2 Overview
To examine whether social connection is important for transforming generous behaviour
into positive feelings, we present three studies in which we manipulate levels of social
connection while participants engage in prosocial spending. In Study 1, we examine
whether the emotional benefits of charitable giving are greatest when people have the
opportunity to donate to someone who is personally connected to the charity (vs. giving
to someone who is not). In Study 2, we turn our focus to interpersonal giving and
investigate whether giving more money in a dictator game leads to higher levels of
happiness when participants deliver money to a recipient than when this donation is
delivered via an intermediary, which blocks the opportunity for social connection with
the recipient. Finally, in Study 3, we examine whether participants experience the highest
happiness when they spend a gift card on someone else in a way that maximises social

158 L.B. Aknin et al.
connection. Taking an applied approach, we focus our examination on the broader
conditions that moderate the emotional benefits of prosocial spending. We expect that the
emotional benefits of generosity are most likely to emerge when prosocial spending
provides an opportunity for positive social connection.
3 Study 1: charitable giving
Study 1 explores whether donating to charity produces greater emotional benefits when
givers have the opportunity for positive social connection with the beneficiary – or, as we
examine here, with even a representative of the beneficiary. Although individuals may
rarely have the opportunity for direct social connection with the end recipients of their
charitable donation, people are often asked to give to charity by someone who is
connected to the cause. For example, many food bank volunteers who solicit donations
are individuals who care about hunger and poverty and may serve meals to the needy, but
are not those in need of food themselves. Given that offering a donation to someone who
cares about the cause provides an opportunity for social connection, we hypothesised that
emotional benefits of prosocial spending would be amplified when people gave a
donation to someone who was personally involved with the charity, such that the link
between charitable behaviour and subsequent happiness would be more pronounced in
the presence of social connection.
3.1 Method
• Participants. Sixty-eight individuals (54% female; M
age
= 22.8, SD = 5.7)
participated in this study in exchange for ten Canadian dollars.
• Procedure. Participants were approached in public places on a university campus
and asked to complete a short study investigating how people evaluate charitable
appeals. After providing consent, participants were paid for their participation and
asked to put the payment away; we encouraged participants to put their payment
away because doing so tends to increase the perception that payment for an
experiment is equivalent in value to one’s own hard-earned cash (Raghubir and
Srivastava, 2008). Participants then reported their pre-task general happiness on a
single-item measure [Do you feel happy in general? 1 – no, 5 – yes; Abdel-Khalek
(2006)] and a two-item measure of the Subjective Happiness Scale [
α
= .77; SHS;
Lyubomirsky and Lepper (1999)]. We created a composite measure of baseline
happiness by averaging standardised scores on these two measures, which were
highly correlated (r = .59; composite
α
= .77). Afterward, participants were
presented with a print ad for a real charity called ACTS; they were informed that
ACTS “is devoted to bringing fresh water to rural communities in Africa and sends
many volunteers from North America to East Africa to help with this cause
annually”. As participants examined this ad, the research assistant explained that
previous participants who had seen this ad had asked if they could donate to ACTS,
so they were welcome to make a donation.

Does social connection turn good deeds into good feelings? 159
Participants were than randomly assigned to the high or low social connection condition.
In the high social connection condition, the research assistant (who was blind to our
hypotheses) explained:
“I am actually personally involved with this charity and even though we’re
doing research on charitable appeals, I chose this charity because of my
connection to this charity. My friend just got back from working with this
organization in Africa and I’m helping to raise money on his behalf. So if
you’d like to donate that would be awesome. Would you like to donate?”
If the participant chose to donate, the research assistant accepted the donation, thanked
the participant, and asked them to record their name and the donation amount on a record
form.
Participants in the low social connection condition were not informed that the
research assistant had a personal connection to the charity. Rather, the research assistant
simply explained that any donation the participant chose to make would go to the charity.
Before turning away from the participant, the research assistant provided the participant
with an envelope and instructed the participant to place any money they wished to donate
in the envelope and then to drop the envelope in a collection box a short distance away.
All funds collected went to the advertised charity.
After making their donation decision, all participants completed a final survey,
reporting their positive and negative affect on the Positive and Negative Affect Schedule
[
α
positive
= .86;
α
negative
= .93; PANAS; Watson et al. (1988)], their current
overall happiness on a one-item measure (1 – not at all, 5 – extremely), and their life
satisfaction on the Satisfaction with Life Scale [
α
= .87; SWLS; Diener et al. (1985)].
Consistent with our previous research (Dunn et al., 2008) and the work of others
(e.g., Anderson et al., 2012), we created a composite measure of well-being by averaging
participants’ standardised scores on all positive scales: positive affect, current happiness,
and life satisfaction, which were all positively correlated (r’s from .20–.59; composite
α
= .86).
1
3.2 Results and discussion
Across studies, we measured the construct of subjective well-being using multiple scales.
To maximise the breadth of our measures and the brevity of our paper we standardised
and averaged each scale to create composite measures of well-being, which are reported
in text. This strategy is consistent with previous research (e.g., Aknin et al., in press;
Anderson et al., 2012) and helpful for assessing the full conceptualisation of subjective
well-being (Biswas-Diener et al., 2009; Diener and Biswas-Diener, 2002; Kashdan et al.,
2008), which cannot be captured in one single scale (Diener, 1984). In line with recent
guidelines for maximising transparency (Simmons et al., 2011), we report results on each
independent scale in tables.
Condition had no impact on the amount of money donated to charity, with
participants giving roughly five dollars to charity in both the high social connection
(M = $5.07, SD = 4.32) and the low social connection condition (M = $5.00, SD = 3.40),
F(1, 66) = .01, ns. Our critical question, however, was whether the emotional benefits of
generosity were greater when participants gave to someone who was personally involved
with the charity, thereby enabling a positive social connection. To investigate this
question, we centred donations to a mean of zero and coded condition assignment

160 L.B. Aknin et al.
(–1 = low social connection, 1 = high social connection). Then, we entered baseline
happiness, donation amount, condition, and a Donation X Condition interaction term into
a regression equation predicting post-donation well-being. As expected, baseline
happiness (
β
= .27, p < .05) and the Donation X Condition interaction term (
β
= .34,
p < .01) were the only significant predictors of well-being ratings, suggesting that the
impact of donation size on happiness depends upon whether the charitable donation is
given in the high or low social connection condition.
We examined the relationship between donation size and post-donation well-being
while controlling for baseline happiness in each condition. We found that larger
donations were associated with higher well-being ratings in the high social connection
condition (
β
= .32, p < .05), but not in the low social connection condition (
β
= –.33,
p = .10). As predicted, the link between giving and subsequent happiness was strongest
when participants gave the donation directly to someone who was personally connected
to the charity (Figure 1; see Table 1 for results on each individual measure). Given that
participants in the low social connection condition were also committing prosocial acts, it
is striking that the relationship between donation size and life satisfaction was negative
(though not significantly so), highlighting the critical role of social connection in
producing positive feelings from giving.
Figure 1 Donation size (centred to a mean of zero) and social connection interact in predicting
the emotional rewards of charitable giving while controlling for baseline happiness
(Study 1)
Note: Larger donations lead to higher levels of happiness in the high social connection
condition but not in the low social connection condition.

Does social connection turn good deeds into good feelings? 161
Table 1 Results from Study 1 on individual measures of well-being
Simple effects
Donation X
condition interaction
Low social connection High social connection
Baseline measure: 1-item general happiness
Post-donation measure
SWLS
β
= .31, p < .02
β
= –.22, p = .25
β
= .31, p = .06
PA
β
= .21, p = .13
β
= –.25, p = .26
β
= .11, p = .50
1-item happiness
β
= .33, p < .02
β
= –.34, p = .11
β
= .29, p = .08
NA
β
= –.21, p = .12
β
= .10, p = .65
β
= –.28, p = .09
Baseline measure: 2-item subjective happiness scale
Post-donation measure
SWLS
β
= .23, p < .05
β
= –.29, p = .11
β
= .18, p = .13
PA
β
= .21, p = .15
β
= –.22, p = .28
β
= .12, p = .47
1-item happiness
β
= .35, p < .02
β
= – .38, p = .07
β
= .25, p = .13
NA
β
= –.17, p = .20
β
= .11, p = .59
β
= –.21, p = .17
4 Study 2: classroom dictator game
Using a paradigm that mirrors common real-world instances of charitable giving, Study 1
provides initial evidence for the role of social connection: people who donated more
money to charity experienced higher well-being afterward only if they gave their
donation to a charitable representative who was personally connected with the cause.
Although giving to charitable organisations is common, people more often give to other
individuals, including friends and acquaintances (Dunn et al., 2008). To generalise the
role of connection to interpersonal contexts, Study 2 used a dictator game paradigm in
which we paid participants $10 and let them decide how much of their payment to give to
a peer. We manipulated whether participants delivered the money to the recipient
personally (enabling brief social connection) or through an intermediary (precluding
social connection). Our account suggests that the emotional benefits of prosocial
spending are most likely to occur when generous acts provide positive social connection
between the giver and recipient; therefore, we predicted that greater generosity would
predict higher levels of happiness only when givers were allowed to interact with
recipients.
4.1 Method
• Participants. Forty-eight undergraduates (63% female; M
age
= 23.0, SD = 5.7)
participated in exchange for a chocolate bar.
• Procedure. Participants reported their pre-task general happiness on the same single-
item pre-task measure as used in Study 1 (Abdel-Khalek, 2006). Half the participants
were then informed that they had been randomly assigned to the decision maker role
in a dictator game, and were given ten one-dollar Canadian coins as additional

162 L.B. Aknin et al.
compensation for participation.
2
Decision makers were told that they had each been
randomly paired with a fellow student, who had not received any money, and that
they should decide how much money to keep and how much (if any) to give this
other student. We manipulated the degree of social connection that this transaction
allowed by informing half of the decision makers that they would personally deliver
the funds to the recipient (high social connection condition), and informing the other
half that a research assistant would deliver the funds on their behalf (low social
connection condition). Decision makers placed whatever amount they wished to
donate (from $0–$10) in an envelope and gave the donation directly to the recipient
or to the research assistant who delivered the funds. All participants then reported
their current happiness on the same one-item post-task measure used in Study 1 and
their current positive and negative affect on the PANAS (
α
positive
= .83;
α
negative
= .89;
Watson, et al., 1988). Consistent with Study 1, we created a composite measure of
post-donation happiness by averaging standardised scores on positive
scales – positive affect and current happiness – which were positively correlated
(r = .25; composite
α
= .83).
Figure 2 Donation size (centred to a mean of zero) and social connection interact in predicting
post-donation well-being while controlling for baseline happiness in the dictator game
(Study 2)
Note: Larger donations lead to higher levels of happiness in the high social connection
condition but not in the low social connection condition.
4.2 Results and discussion
In line with Study 1, decision makers gave away approximately half their windfall to
their paired recipient, regardless of whether they were in the high (M = $5.50, SD = 2.71)
or low social connection condition (M = $5.25, SD = 2.67), F(1, 22) = .05, ns. Again, our
primary interest was in whether the emotional benefits of generosity were greater when

Does social connection turn good deeds into good feelings? 163
donations were made in person. As in Study 1, we centred donations to a mean of zero
and coded condition assignment (–1 = low social connection, 1 = high social connection).
We then entered pre-task happiness, donation amount, condition, and a Donation X
Condition interaction term into a regression equation predicting post-task positive affect.
The interaction term was the only significant predictor of positive affect (
β
= .47,
p < .05), indicating that the impact of donation amount on positive affect may depend
upon the level of social connection.
We split decision makers into high and low social connection groups and entered pre-
task happiness and donation amount as predictors of post-task positive affect. In line with
our predictions, larger donations were associated with higher post-task positive affect in
the high social connection condition (
β
= .83, p < .005), but this effect was again
eliminated – and reversed – in the low social connection condition (
β
= –.60, p < .01).
Thus, giving larger donations appears to produce higher levels of positive affect (above
and beyond baseline levels of happiness), but only when a social interaction took place
between benefactor and beneficiary (Figure 2; see Table 2 for results on each individual
measure).
Table 2 Results from Study 2 on individual measures of post-donation well-being
Simple effects
Interaction
Low social connection High social connection
Baseline measure: 1-item general happiness
PA
β
= .43, p < .04
β
= –.52, p = .13
β
= .68, p <.02
1-item happiness
β
= .32, p = .06
β
= –.42, p = .07
β
= .63, p < .01
NA
β
= .02, p = .94
β
= –.19, p = .62
β
= –.01, p = .98
5 Study 3: Starbucks gift cards
Study 2 replicated the importance of social connection for the emotional benefits of
giving and extended these findings to the realm of interpersonal giving; people who made
more generous financial decisions experienced more positive affect afterward if they
were allowed to interact with their beneficiary. While this design allowed us to
investigate our hypothesis in a controlled context, the artificiality of this paradigm
represents a limitation. Therefore, in Study 3, we conducted a field study to examine the
emotional consequences of generous spending in a more naturalistic setting.
Study 3 also allowed us to address two additional limitations of Studies 1 and 2. First,
in both studies, the opportunity for social connection was confounded with social
recognition of committing a kind act: in the high social connection condition, the
benefactor could both interact with and receive positive recognition from the beneficiary,
whereas there was no such opportunity in the low social connection condition. Second,
donation sizes were not randomly assigned; participants were allowed to decide how
much money to give away to charity (Study 1) or peers (Study 2). Therefore, in Study 3,
participants were instructed to give the same gift (a Starbucks gift card) directly to
someone else, while we manipulated the degree of interaction between giver and
recipient. Finally, in Study 3 we examined happiness at the end of the day, rather than
immediately after the prosocial spending experience.

164 L.B. Aknin et al.
5.1 Method
• Participants. Fifty-eight individuals agreed to participate in a study on gift card
spending. Eight of these participants – evenly dispersed across conditions – could not
be reached for follow-up calls and were removed from analyses, leaving a final
sample of 50 participants (66% female; M
age
= 21.0, SD = 2.5).
• Procedure. Participants were approached in the morning on campus and given a
Starbucks gift card valued at $10 Canadian to use by the end of the day. In a 2
(spending type: personal vs. prosocial) X 2 (social connection: high vs. low) design,
participants were randomly assigned to spend the gift card on either themselves or
someone else in a way that either minimised or maximised social connection
(Table 3). Specifically, participants who were told to use the card to benefit someone
else were instructed either
a to give the gift card in its entirety to someone else as a gift and not to
accompany the recipient to Starbucks, thereby minimising social connection
b to visit Starbucks with this person and spend the gift card on both themselves
and the recipient, thereby maximising social connection.
Participants who were told to spend the card on themselves were instructed either to
a go to Starbucks by themselves
b go to Starbucks with a friend but to spend the gift card only on themselves.
3
Table 3 Spending directions and contrast weights the four conditions in Study 3
Social connection
Low High
Please use this $10.00 Starbucks
gift card to buy yourself a
coffee/treat while alone.
Please use this $10.00 Starbucks
gift card to buy yourself a
coffee/treat while visiting
Starbucks with a friend.
Oneself
Contrast weight = –1 Contrast weight = –1
Please give this $10.00 Starbucks
gift card to someone else as a gift.
Please use this $10.00 Starbucks
gift card to buy yourself and
someone else a coffee/treat to
have while visiting Starbucks.
Gift card
target
Someone else
Contrast weight = –1 Contrast weight = +3
Note: Participants were given a thorough explanation of their condition instructions in
person, with the brief summaries above provided in writing as reminders.
Participants were contacted in the evening by a research assistant, blind to condition, and
rated their current positive and negative affect on the PANAS [
α
positive
= .80;
α
negative
= .83; Watson et al. (1988)], overall happiness on the SHS [
α
= .77; Lyubomirsky
and Lepper (1999)], and life satisfaction on the SWLS [
α
= .86; Diener
et al. (1985)]. Again, we created a composite of well-being by averaging participants’
standardised scores of positive affect, overall happiness, and life satisfaction (r’s from
.09–.57; composite
α
= .83).

Does social connection turn good deeds into good feelings? 165
5.2 Results and discussion
We predicted that participants would experience the highest levels of happiness when
they spent the gift card on someone else in a way that maximised social connection. To
test this question, we first used a 2 (spending type: personal vs. prosocial) X 2 (social
connection: high vs. low) analysis of variance (ANOVA). Results revealed a
non-significant main effect of spending type F(1,46) = 0.68, p > .40, a non-significant
main effect of social connection F(1,46) = 2.32, p > .12, and a non-significant interaction
F(1,46) = 1.32, p > .25. Given our specific prediction, however, we used a planned
contrast to analyse well-being with weights reflecting our hypothesised pattern, as shown
in Table 3. As expected, this contrast was significant, t(46) = 2.01, p = .05, demonstrating
that participants randomly assigned to spend the gift card on someone else by going to
Starbucks with that individual were significantly happier at the end of the day than
participants in the other three spending conditions (Figure 3; see Table 4 for results of
each individual measure).
4
Thus, participants who spent on others in a way that allowed
for social connection experienced the highest levels of happiness at the end of the day.
Figure 3 Happiness levels reported after gift card spending in each of the four conditions in
Study 3 on composite measure of well-being
Table 4 Results from Study 3 on individual measures of well-being
Means (standard deviation)
Target: self
connection:
maximum
Target: self
connection:
minimum
Target: other
connection:
maximum
Target: other
connection:
minimum
Contrast value
SHS 4.84 (.84) 4.88 (.66) 5.34 (1.02) 4.46 (1.00) t(46) = 2.12, p < .04
SWLS 4.79 (1.33) 4.70 (1.09) 5.45 (.90) 4.84 (.72) t(46) = 1.91, p = .06
PA 2.81 (.67) 2.70 (.44) 2.83 (.51) 2.78 (.70) t(46) = 0.34, p = .73
NA 1.47 (.57) 1.45 (.60) 1.41 (.24) 1.45 (.39) t(46) = –0.25, p = .80

166 L.B. Aknin et al.
Using a field study designed to mimic everyday spending behaviours, Study 3 provided
additional support for our hypothesis that the emotional benefits of spending are greatest
when people spend on others in a way that facilitates social connection. That said, asking
participants to engage in everyday spending behaviours allows for several confounds. For
instance, participants who gave the gift card as a gift may have exerted less effort, since
they did not actually go to Starbucks themselves; many of the confounds present in this
field study, however, are absent from the more controlled Study 2. Taken together with
Studies 1 and 2, Study 3 supports our hypothesis that spending leads to the greatest
hedonic benefits when people engage in prosocial spending that promotes positive social
connection.
6 General discussion
Across three studies, we found support for the hypothesis that social connection is
important for transforming good deeds into good feelings. In Study 1, we showed that
participants who gave larger donations to a charitable organisation reported higher levels
of happiness after doing so when they gave their donations directly to someone who was
connected to the charity. In Study 2, we found that participants who gave more money to
a recipient were happier when they delivered the funds directly to the beneficiary;
participants did not experience the emotional benefits of giving when this same
transaction was performed by an intermediary, thus inhibiting social connection. Finally,
in Study 3, participants who spent a Starbucks gift card on someone else, and spent time
with that person while doing so, were happiest at the end of the day, again suggesting that
prosocial spending that allows for positive social connection leads to higher levels of
happiness. These findings help to clarify the inconsistent support for the claim that
prosocial behaviour leads to an increase in well-being by demonstrating that how good
deeds are enacted is important for understanding when emotional benefits are
experienced.
While alternative explanations for each individual study are easy to generate, the
results of all three studies presented here can be most parsimoniously explained by the
hypothesis that positive social connection is important for reaping the emotional benefits
of generous financial behaviour. Although we did not directly assess levels of social
connection in each study, we used face-valid manipulations that approximate socially
connected giving opportunities in daily life. Indeed, in Study 1 we were interested in the
consequences of direct social connection with a charity representative. Thus, we reasoned
that having some participants give a donation to an experimenter who made it known that
she had a personal connection to the cause while having other participants donate through
the same experimenter without her disclosing her connection would allow us to test this
question. Similarly, in Study 2 we were interested in the consequences of direct social
connection with a beneficiary and therefore included one condition with social interaction
and one without. Finally, in Study 3 we used face-valid manipulations of social
connection by asking participants in the high social connection condition to spend time
with their beneficiary. These manipulations are consistent with previous work examining
the consequences of meaningful contact with a beneficiary (Grant et al., 2007).
Although it is conceivable that demand characteristics may have affected our results,
we think they are unlikely to account for our findings for several reasons. First, if

Does social connection turn good deeds into good feelings? 167
participants were trying to please the experimenter by reporting higher levels of
happiness after giving more money to others, we would expect to see a positive
relationship between donation size and happiness in both conditions of Studies 1 and 2.
However, we see that larger donations lead to higher levels of happiness only when
donations are given directly to the beneficiary or representative. Moreover, it is unlikely
that our results are due to demand effects because previous work has shown that people
do not believe that spending money on others produces greater happiness than spending
money on oneself (Dunn et al., 2008). In addition, it is unlikely that researchers were
shaping participant responses; in Study 3, the research assistant assessing happiness at the
end of the day was blind to condition assignment, meaning they could not influence
happiness reports. Thus it is unlikely that the higher levels of happiness seen after
socially connected prosocial spending reflect demand effects from the participant or
experimenter.
Employing three of the most widely used happiness scales, our studies demonstrate
that the effects of prosocial spending on happiness may differ depending upon the degree
of social connection between the benefactor and beneficiary. These results are
particularly striking given the similarity of the behaviours enacted in Studies 1 and 2.
That is, when participants had the opportunity to decide how much money to give to
charity or another participant, donations were nearly indistinguishable in the high and
low social connection conditions ($5.07 vs. $5.00 in Study 1 and $5.50 vs. $5.25 in
Study 2). Yet, generous behaviour produced drastically different emotional consequences
depending on the degree of social connection the situations allowed.
While our investigation focused on the emotional consequences of direct
interpersonal giving, a growing body of research suggests that a similar relationship
between giving and well-being may exist at the macro-level as well. Indeed, looking
across nations, Oishi et al. (2012) demonstrate that residents in countries with higher
progressive taxation are happier than residents in countries with less progressive taxation,
even while controlling for important variables such as income inequality and national
wealth. More directly, Arvin and Lew (2010) assessed the link between prosocial
spending and well-being at the national level by surveying foreign aid and citizens’
happiness levels in nine donor countries. Their findings suggest that the amount of
foreign aid distributed by a nation may predict donor citizens’ happiness (see also Arvin
and Lew, 2009). While the results were only significant in one European nation (France),
this may be due, in part, to the fact that foreign aid was disbursed to countries that
citizens did not feel connected to. Indeed, it is possible that citizens in donor nations
would feel happier if aid was given to a country that many citizens feel connected to,
such as a country where many citizens have travelled. One implication of our research is
that foreign aid is likely to lead to the greatest well-being benefits when donors feel
psychologically connected to the recipients of their aid.
Because our goal was to understand the broad conditions that moderate the emotional
benefits of prosocial spending rather than isolate the specific mechanisms responsible, we
manipulated social connection in ways that parallel real giving opportunities. For
example, although most major charities allow people the convenience of making
donations online, one of the most common reasons people give to charity is that they are
asked for a contribution by someone they know (Independent Sector, 1999), who may be
connected to the charity. This latter form of socially connected giving differs from the
former in multiple ways, including anonymity, social pressure, the likelihood of a

168 L.B. Aknin et al.
positive social interaction, and the opportunity for a direct expression of gratitude. We
designed our high and low social connection conditions to mimic these real world
contexts by including many of the features that socially connected giving entails; further
investigation is needed to identify the specific aspects of social connection that might
yield the largest emotional rewards in particular giving situations.
Importantly, our research suggests that happiness is not an inevitable outcome of
prosocial spending; generosity produced the greatest happiness benefits when positive
social connection was facilitated. This yields important practical implications for both
individuals and philanthropic organisations. A simple way to apply this research may be
to use one’s prosocial spending experiences as opportunities for positive social
interactions with the beneficiary of one’s gift. For example, if you choose to buy your
niece a basketball for her birthday, you would be more likely to experience a happiness
boost from taking the time to shoot some hoops with her rather than simply mailing the
ball to her as a gift. Similarly, our findings suggest that charities should strive to increase
the interpersonal connection between donors and beneficiaries, such that donors
experience an emotional boost from giving. For instance, one author recently gave her
mother a gift certificate to DonorsChoose.org, a charity that funds various public school
classroom projects across the USA. After helping a third-grade classroom fund an art and
reading project, the author’s mom received hand-written thank-you cards from the
students and teacher who benefitted from her donation. The personalised messages
increased the social connection between the donor and beneficiary, making the donation
more rewarding and potentially increasing the likelihood that she will give again. Indeed,
previous research suggests that when people experience greater happiness after giving
they are more likely to act generously in the future (Aknin et al., 2011). Thus, social
connection may not only make the initial generous act more rewarding, it may also
encourage subsequent donations and initiate a positive feedback loop between giving and
happiness.
7 Conclusions
The present research provides the first direct investigation of whether social connection
acts as a moderator in transforming generous behaviour into positive feelings. Examining
this hypothesis in three contexts designed to map instances of real-world giving, we find
that financial generosity leads to the largest happiness gains when acts of giving involve
increased social connection with a representative of the recipient (Study 1) or with the
recipients themselves (Studies 2 and 3). Our findings dovetail with research suggesting
that positive social interactions, expanded social networks, and reduced isolation mediate
the emotional benefits of volunteer work (Musick and Wilson, 2003). While additional
factors other than social connection likely influence the happiness gained from prosocial
spending, our findings suggest that putting the social in prosocial is one way to transform
good deeds into good feelings.

Does social connection turn good deeds into good feelings? 169
References
Abdel-Khalek, A.M. (2006) ‘Measuring happiness with a single-item scale’, Social Behavior and
Personality, Vol. 34, No. 2, pp.139–150.
Aknin, L.B., Barrington-Leigh, C.P., Dunn, E.W., Helliwell, J.F., Burns, J., Biswas-Diener, R.,
Kemeza, I., Nyende, P., Ashton-James, C.E. and Norton, M.I. (in press) ‘Prosocial spending
and well-being: cross-cultural evidence for a psychological universal’, Journal of Personality
and Social Psychology.
Aknin, L.B., Dunn, E.W. and Norton, M.I. (2011) ‘Happiness runs in a circular motion: evidence
for a positive feedback loop between prosocial spending and happiness’, Journal of Happiness
Studies, Vol. 13, No. 2, pp.347–355.
Anderson, C., Kraus, M.W., Galinsky, A.D. and Keltner, D. (2012) ‘The local ladder effect: social
status and subjective well-being’, Psychological Science, Vol. 23, No. 7, pp.764–771.
Andreoni, J. (1989) ‘Giving with impure altruism: applications to charity and Ricardian
equivalence’, The Journal Political Economy, Vol. 97, No. 6, pp.1447–1458.
Andreoni, J. (1990) ‘Impure altruism and donations to public goods: a theory of warm-glow
giving’, The Economic Journal, Vol. 100, No. 401, pp.464–477.
Arvin, B.M. and Lew, B. (2009) ‘Happiness and foreign aid’, Atlantic Economic Journal, Vol. 37,
pp.325–326.
Arvin, B.M. and Lew, B. (2010) ‘Aid and happiness: untangling the causal relationship in
nine European donor countries’, Global Business and Economics Review, Vol. 12, No. 4,
pp.341–358.
Barone, M.J., Miyazaki, A.D. and Taylor, K.A. (2000) ‘Does one good turn deserve another?
Examining the influence of cause-related marketing efforts on consumer choice’, Journal of
the Academy of Marketing Science, Vol. 28, No. 2, pp.250–264.
Baumeister, R.F. and Leary, M.R. (1995) ‘The need to belong: desire for interpersonal attachments
as a fundamental human motivation’, Psychological Bulletin, Vol. 117, No. 3, pp.497–529.
Biswas-Diener, R., Kashdan, T.B. and King, L.A. (2009) ‘Two traditions of happiness research, not
two distinct types of happiness’, Journal of Positive Psychology, Vol. 4, No. 3, pp.208–211.
Diener, E. (1984) ‘Subjective well-being’, Psychological Bulletin, Vol. 95, No. 3, pp.542–575.
Diener, E. and Biswas-Diener, R. (2002) ‘Will money increase subjective well-being?’, Social
Indicators Research, Vol. 57, No. 2, pp.119–169.
Diener, E. and Seligman, M.E.P. (2002) ‘Very happy people’, Psychological Science, Vol. 13,
No. 1, pp.81–84.
Diener, E., Emmons, R.A., Larson, R.J. and Griffin, S. (1985) ‘The satisfaction with life scale’,
Journal of Personality Assessment, Vol. 49, No. 1, pp.71–75.
Dovidio, J.F., Piliavin, J.A., Schroeder, D.A. and Penner, L.A. (2006) The Social Psychology of
Prosocial Behavior, The Social Psychology of Prosocial Behavior, Lawrence Erlbaum
Associates, Mahwah, NJ.
Dunn, E.W., Aknin, L.B. and Norton, M.I. (2008) ‘Spending money on others promotes happiness’,
Science, Vol. 319, No. 5870, pp.1687–1688.
Dunn, E.W., Ashton-James, C.E., Hanson, M.D. and Aknin, L.B. (2010) ‘On the costs of
self-interested economic behavior: how does stinginess get under the skin?’, Journal of Health
Psychology
, Vol. 15, No. 4, pp.627–633.
Grant, A.M., Campbell, E.M., Chen, G., Cottone, K., Lapedis, D. and Lee, K. (2007) ‘Impact and
the art of motivation maintenance: the effects of contact with beneficiaries on persistence
behaviour’, Organizational Behaviour and Human Decision Processes, Vol. 103, No. 1,
pp.53–67.
Harris, M.B. (1977) ‘Effects of altruism on mood’, Journal of Social Psychology, Vol. 102,
pp.197–208.

170 L.B. Aknin et al.
Independent Sector (1999) Giving and Volunteering in the United States [online]
http://www.independentsector.org/GandV/default.htm (accessed 9 January 2010).
Kashdan, T.B., Biswas-Diener, R. and King, L.A. (2008) ‘Reconsidering happiness: the costs
of distinguishing between hedonics and eudaimonia’, Journal of Positive Psychology, Vol. 3,
No. 4, pp.219–233.
Kogut, T. and Ritov, I. (2005) ‘The ‘identified victim’ effect: an identified group, or just a single
individual?’, Journal of Behavioral Decision Making, Vol. 18, No. 3, pp.157–167.
Lev, B., Petrovits, C. and Radhakrishnan, S. (2010) ‘Is doing good, good for you? How corporate
charitable contributions enhance revenue growth’, Strategic Management Journal, Vol. 31,
No. 2, pp.182–200.
Lyubomirsky, S. (2008) The How of Happiness: A Scientific Approach to Getting the Life You
Want, Penguin Press, New York, NY.
Lyubomirsky, S. and Lepper, H.S. (1999) ‘A measure of subjective happiness: preliminary
reliability and construct validation’, Social Indicators Research, Vol. 46, No. 2, pp.137–155.
Lyubomirsky, S., King, L. and Diener, E. (2005a) ‘The benefits of frequent positive affect: does
happiness lead to success?’, Psychological Bulletin, Vol. 131, No. 6, pp.803–851.
Lyubomirsky, S., Sheldon, K.M. and Schkade, D. (2005b) ‘Pursuing happiness: the architecture of
sustainable change’, Review of General Psychology, Vol. 9, No. 2, pp.111–131.
Musick, M.A. and Wilson, J. (2003) ‘Volunteering and depression: the role of psychological
and social resources in different age groups’, Social Sciences and Medicine, Vol. 56, No. 2,
pp.259–269.
Myers, D.G. (2000) ‘The funds, friends, and faith of happy people’, American Psychologist,
Vol. 55, No. 1, pp.56–67.
Oishi, S., Schimmack, U. and Diener, E. (2012) ‘Progressive taxation and the subjective well-being
of nations’, Psychological Science, Vol. 23, No. 1, pp.86–92.
Raghubir, P. and Srivastava, J. (2008) ‘Monopoly money: the effect of payment coupling and
form on spending behavior’, Journal of Experimental Psychology: Applied, Vol. 14, No. 3,
pp.213–225.
Ryan, R.M. and Deci, E.L. (2000) ‘Self-determination theory and the facilitation of intrinsic
motivation, social development, and well-being’, American Psychologist, Vol. 55, No. 1,
pp.68–78.
Simmons, J.P., Nelson, L.D. and Simonsohn, U. (2011) ‘False-positive psychology: undisclosed
flexibility in data collection and analysis allows presenting anything as significant’,
Psychological Science, Vol. 22, No. 11, pp.1359–1366.
Small, D.A. and Loewenstein, G. (2003) ‘Helping a victim or helping the victim: altruism and
identifiability’, Journal of Risk and Uncertainty, Vol. 26, No. 1, pp.5–16.
Small, D.A. and Simonsohn, U. (2008) ‘Friends of victims: personal experience and prosocial
behavior’, Journal of Consumer Research, Vol. 35, No. 3, pp.532–542.
Strahilevitz, M. and Myers, J.G. (1998) ‘Donations to charity as purchase incentives: how well they
work may depend on what you are trying to sell’, Journal of Consumer Research, Vol. 24,
No. 4, pp.434–446.
Thoits, P.A. and Hewitt, L.N. (2001) ‘Volunteer work and well-being’, Journal of Health and
Social Behavior, Vol. 42, No. 2, pp.115–131.
Watson, D., Clark, L.A. and Tellegen, A. (1988) ‘Development and validation of brief measures of
positive and negative affect: the PANAS scales’, Journal of Personality and Social
Psychology, Vol. 54, No. 6, pp.1063–1070.
Watson, D., Clark, L.A., McIntyre, C.W. and Hamaker, S. (1992) ‘Affect, personality, and social
activity’, Journal of Personality and Social Psychology, Vol. 63, No. 6, pp.1011–1025.
Williamson, G.M. and Clark, M.S. (1989) ‘Providing help and desired relationship type as
determinants of changes in moods and self-evaluations’, Journal of Personality and Social
Psychology, Vol. 56, No. 5, pp.722–734.

Does social connection turn good deeds into good feelings? 171
Notes
1 Although negative affect was not the focus of the present research and reported levels of NA
were typically quite low, the results on NA for all studies are provided in the tables for
interested readers.
2 Participants signed a receipt to acknowledge this payment and ensure a sense of ownership.
3 To confirm that our four conditions were ecologically valid, we asked an additional sample of
40 undergraduates to report how frequently they engaged in the four types of coffee buying
behaviours assigned in Study 3. Participants reported that they engage in all four types of
spending behaviour at least 10% of the time. We also confirmed that buying a coffee for
oneself while with others, an experience that some may consider awkward, represents a
familiar spending behaviour. When participants were asked to rate how common or
uncommon this spending behaviour is on a 7-point likert scale (1 – very uncommon, 7 – very
common), participants rated this behaviour significantly above the midpoint (X = 5.38,
SD = 1.92), t(39) = 4.54, p < .001.
4 Given that past research has shown a link between social interaction and happiness (e.g.,
Watson et al., 1992), it is likely that participants experienced increased positive mood
immediately after visiting Starbucks with a friend, even in the absence of prosocial spending.
We expected the hedonic effects of this single interaction to last through the end of the
day – when happiness was assessed – only when coupled with the positive benefits of
prosocial spending.