Content uploaded by Jiao Zhang
Author content
All content in this area was uploaded by Jiao Zhang on Mar 26, 2014
Content may be subject to copyright.
http://pss.sagepub.com/
Psychological Science
http://pss.sagepub.com/content/early/2013/08/01/0956797613482947
The online version of this article can be found at:
DOI: 10.1177/0956797613482947
published online 1 August 2013Psychological Science
Christopher K. Hsee, Jiao Zhang, Zoe Y. Lu and Fei Xu
Unit Asking: A Method to Boost Donations and Beyond
Published by:
http://www.sagepublications.com
On behalf of:
Association for Psychological Science
can be found at:Psychological ScienceAdditional services and information for
http://pss.sagepub.com/cgi/alertsEmail Alerts:
http://pss.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
What is This?
- Aug 1, 2013OnlineFirst Version of Record >>
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
Psychological Science
XX(X) 1 –8
© The Author(s) 2013
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0956797613482947
pss.sagepub.com
Research Article
Raising charitable donations requires considerable time
and effort. In the United States alone, for example, it costs
more than $50 billion each year to raise roughly $300 bil-
lion in individual donations (Greenfield, 1999; National
Philanthropic Trust, 2012). In the research reported here,
we introduced a simple and virtually costless method for
boosting donations. The method is applicable both to
conventional paper- and telephone-based fund-raisers
and to increasingly popular Web-based fund-raisers. We
wish to make it clear up front that the primary purpose of
this research was not to introduce new psychological the-
ories but to introduce a method that is based on psycho-
logical theories and designed to help people in real life;
nevertheless, we also discuss the theoretical foundation
and implications of this method.
The Method and Hypotheses
Consider a Web-based fundraiser that solicits donations
to help N needy persons. The Web site has two versions
of solicitation: control (conventional) and unit asking
(experimental). The control version asks potential donors
to decide how much to donate for all of the N persons.
The unit-asking version first asks potential donors a
hypothetical question—“How much would you donate
for one of the N persons?”—and then asks them to decide
how much to donate for all of the N persons. Note that
the unit-asking version merely adds a hypothetical ques-
tion that carries no additional external information. Yet
we propose that this hypothetical question will increase
donations to the N persons. We refer to this effect as the
unit-asking effect. The reason unit asking works is not
that it misleads respondents, draws their attention to the
number of persons in need, or renders the needy persons
identifiable. Rather, unit asking works because people
are initially scope insensitive and subsequently scope
consistent.
Specifically, the method builds on two well-established
psychological effects. One is scope insensitivity. When
482947PSS
XXX10.1177/0956797613482947Hsee et al.Unit Asking
research-article2013
Corresponding Author:
Christopher K. Hsee, Booth School of Business, University of Chicago,
5807 S. Woodlawn, Chicago, IL 60637
E-mail: christopher.hsee@chicagobooth.edu
Unit Asking: A Method to Boost Donations
and Beyond
Christopher K. Hsee
1
, Jiao Zhang
2
, Zoe Y. Lu
3
, and Fei Xu
3
1
Booth School of Business, University of Chicago;
2
School of Business Administration, University of Miami;
and
3
Antai School, Shanghai Jiao Tong University
Abstract
The solicitation of charitable donations costs billions of dollars annually. Here, we introduce a virtually costless method
for boosting charitable donations to a group of needy persons: merely asking donors to indicate a hypothetical amount
for helping one of the needy persons before asking donors to decide how much to donate for all of the needy persons.
We demonstrated, in both real fund-raisers and scenario-based research, that this simple unit-asking method greatly
increases donations for the group of needy persons. Different from phenomena such as the foot-in-the-door and
identifiable-victim effects, the unit-asking effect arises because donors are initially scope insensitive and subsequently
scope consistent. The method applies to both traditional paper-based fund-raisers and increasingly popular Web-based
fund-raisers and has implications for domains other than fund-raisers, such as auctions and budget proposals. Our
research suggests that a subtle manipulation based on psychological science can generate a substantial effect in real
life.
Keywords
decision making, judgment
Received 12/4/12; Revision accepted 2/11/13
Psychological Science OnlineFirst, published on August 1, 2013 as doi:10.1177/0956797613482947
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
2 Hsee et al.
asked to decide on their willingness to donate (WTD) for
a certain number of needy persons, respondents largely
ignore the number because the number itself is often
independently inevaluable (Hsee, 1996; Hsee & Zhang,
2010). Instead, respondents base their WTD decisions on
easy-to-evaluate cues, such as their feelings toward the
needy persons (Peters, Västfjäll, Gärling, & Slovic, 2006;
Slovic, Finucane, Peters, & MacGregor, 2002, 2007). Such
feelings depend largely on a prototype of the needy per-
sons, regardless of their number (Dunn & Ashton-James,
2008; Hsee & Rottenstreich, 2004; Kahneman & Frederick,
2005). Thus, the unit WTD that respondents in the unit-
asking condition offer for one needy person will likely be
close to the total WTD that respondents in the control
condition offer for all of the N needy persons. We should
mention that WTD for one person can be even higher
than that for N persons if both the one person and each
of the N persons are identified (Kogut & Ritov, 2005a,
2005b). However, in our studies, we did not identify each
of the N persons and, therefore, did not expect WTD to
be higher for one person than for N persons.
The second psychological effect on which the unit-
asking method builds is the desire for consistency (Ariely,
Loewenstein, & Prelec, 2003; Freedman & Fraser, 1966).
When the unit-asking respondents are asked to decide
on their total WTD for all of the N needy persons after
having indicated a unit WTD for one needy person, their
desire for consistency will compel them to contribute
more for N persons, thus generating the unit-asking
effect.
It should be noted that although we expect unit asking
to increase total WTD, we do not expect total WTD to be
N times as large as unit WTD. Donors have budgetary
constraints; if the maximum amount donors can afford to
contribute is $30 and if they have already expressed a
unit WTD of $20 for one person, then the maximum total
WTD they will offer for N persons will be only $30, not
$20 × N.
Our explanation for the unit-asking effect—initial
scope insensitivity and subsequent scope consistency—
resonates with existing research on evaluability (Hsee,
1996; Hsee & Zhang, 2010; Moore, 1999) and arbitrary
coherence (Ariely et al., 2003). Research on evaluability
has suggested that in the single evaluation mode (in
which a single value is presented and evaluated), people
are relatively insensitive to the magnitude of the value,
whereas in the joint evaluation mode (in which multiple
values are juxtaposed and evaluated comparatively),
people are relatively sensitive to the magnitude of the
values. The total WTD that control-condition respondents
indicate is independent of the unit WTD that unit-
asking respondents indicate; therefore, it is in the single-
evaluation mode and, hence, scope insensitive. In contrast,
the total WTD that unit-asking respondents indicate
immediately follows the unit WTD that these respondents
indicate; therefore, it is in the joint-evaluation mode and,
hence, scope sensitive (Kogut & Ritov, 2005b; Moore,
1999).
Research on arbitrary coherence has offered a similar
account. According to that research, the first response to
an unfamiliar stimulus is arbitrary, and subsequent
responses are consistent with the first response. On the
one hand, both the total WTD that control-condition
respondents indicate and the unit WTD that unit-asking
respondents indicate are the first responses in the two
respective conditions; therefore, these responses are arbi-
trary and scope insensitive. On the other hand, the total
WTD that unit-asking respondents indicate is the second
response in their condition; therefore, it is coherent and
scope sensitive. In the current research, we applied these
psychological phenomena to facilitate fund-raising.
We report results from three studies that tested the unit-
asking method. In Study 1, a scenario-based study, we
demonstrated the unit-asking effect and addressed poten-
tial alternative explanations. In Studies 2 and 3, both field
experiments, we applied the unit-asking method and rep-
licated the unit-asking effect in real fund-raisers.
Study 1
Method
Participants were 180 adults in the United States (41%
female, 59% male; mean age = 34.7 years) recruited from
a nationwide online survey service and paid a nominal
amount. The study was conducted shortly before
Christmas. In a questionnaire, respondents were asked to
imagine that the principal of a neighborhood kindergar-
ten had sent them an e-mail requesting that they make a
donation to buy Christmas gifts for the children there.
Respondents were then directed to a Web site and told
that they could revise their answers at any time and that
their answers would not be recorded until they had com-
pleted the entire survey and submitted it.
According to the Web site, the kindergarten’s students
included 20 children from low-income families whose
parents had little money with which to buy Christmas
gifts, and the kindergarten principal hoped that the
respondents would make a donation for buying these 20
children Christmas gifts. The Web site included the por-
trait of a young girl and described her as one of the
needy children.
Unbeknownst to the respondents, the Web site had
four between-subjects versions (conditions). The two
main versions were control and unit-asking versions. The
control version asked, “Please think about all of these 20
children. How much are you willing to donate to help
these 20 children? Please enter the amount of money
you decide and agree to donate: $_____.” After entering
an amount of money, respondents were then given the
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
Unit Asking 3
option to revise the amount or submit it. The unit-asking
version was identical to the control version with one
exception: Before respondents were asked how much to
donate for the 20 children, they first were asked about
one child (with what was explicitly described as a hypo-
thetical question): “Before you decide how much to
donate to help these 20 children, please first think about
one such child and answer a hypothetical question: How
much would you donate to help this one child? Please
indicate the amount here: $____.” After respondents
typed the amount, they were asked the same question
about the 20 children that respondents in the control
condition answered.
In addition to the control and unit-asking versions, we
included two other versions to address potential ques-
tions: a thinking-only version and a number-recalling ver-
sion. The thinking-only version was identical to the
unit-asking version except that it asked respondents to
think only about the hypothetical unit-asking question
and did not ask them to actually write down their answer
(their unit WTD for one child); we included this version
to test whether merely thinking about the unit-asking
question without providing an answer would also
increase total donations. The number-recalling version
was identical to the control version except that before
respondents were asked to decide on their total WTD,
they first were asked to recall and write down the num-
ber of children in the kindergarten they were supposed
to help; we included this version to examine whether
merely paying attention to the total number of needy
persons without seeing the unit-asking question would
also increase total donations.
Results and discussion
Figure 1 summarizes the total-WTD results of the four
conditions. A one-way analysis of variance (ANOVA)
revealed a significant effect, F(3, 168) = 11.15, p < .001.
1
To test the unit-asking effect, we conducted a planned
comparison between the two primary conditions, control
and unit asking, and found a significant unit-asking
effect on total WTD; respondents in the unit-asking con-
dition contributed more than twice as much for the 20
children (M = $49.42) as did respondents in the control
condition (M = $18.03), F(1, 168) = 22.68, p < .0001, d =
0.88. Figure 1 also presents the unit-WTD result in the
unit-asking condition. Comparisons of the total-WTD
results with the unit-WTD result revealed that the total
WTD in the control condition was not significantly differ-
ent from the unit WTD in the unit-asking condition, t < 1,
n.s., but the total WTD in the unit-asking condition
was significantly higher than the unit WTD in the unit-
asking condition, t(42) = 5.28, p < .001. These results
corroborated our proposition that respondents were ini-
tially scope insensitive and subsequently scope consis-
tent (although not linearly so).
Two questions arose regarding our findings. First,
would respondents donate more if they were asked to
merely think about the unit-asking question without
answering it? Second, did the unit-asking effect occur
because unit asking drew respondents’ attention to the
number of needy children? In other words, would respon-
dents donate more if they were merely prompted to
attend to the number without the unit-asking manipula-
tion? To address these questions, we examined the other
$0.00
$10.00
$20.00
$30.00
$40.00
$50.00
$60.00
ControlUnit Asking Thinking OnlyNumber Recalling
WTD Amount
Unit WTD
Total WTD
Fig. 1. Results from Study 1: mean amount of money participants were willing to donate (WTD) in
different conditions. Error bars indicate standard errors.
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
4 Hsee et al.
two conditions of the study—the thinking-only condition
(to address the first question) and the number-recalling
condition (to address the second question). Specifically,
we conducted post hoc Duncan tests to compare total
WTD between the different conditions and found that
total WTD in the unit-asking condition was significantly
higher than that in the other three conditions but that
total WTD in the thinking-only condition and in the num-
ber-recalling condition was not significantly different
from that in the control condition. Thus, merely thinking
about the unit-asking question or merely paying atten-
tion to the number of needy children without the unit-
asking manipulation was not sufficient to boost donations.
It seemed necessary for respondents to provide the unit-
WTD amount. What was written down was concrete (the
number was “out there”), and this concreteness accentu-
ated the respondents’ need to be consistent in making
their subsequent total-WTD decisions. We should note,
however, that this need was internal rather than external;
as highlighted later, the unit-asking effect arises even in
situations in which respondents make their decisions pri-
vately and anonymously and can revise their decisions
without scrutiny from others.
Although seemingly related, the unit-asking effect is
distinct from the identifiable-victim effect—the finding
that people are more inclined to help identifiable victims
than unidentifiable victims (Jenni & Loewenstein, 1997;
Kogut & Ritov, 2005a, 2007; Small & Loewenstein, 2003,
2005; Small, Loewenstein, & Slovic, 2007; Small & Verrochi,
2009). According to the existing literature (e.g., Small
et al., 2007), an effective way to render the needy persons
identifiable is to include a picture of a single needy per-
son. As described earlier, we included a picture of a single
child in all of the conditions; therefore, the identifiability
of the children was already high. The fact that we still
observed a robust unit-asking effect rendered identifiabil-
ity an unviable explanation. To further rule out identifi-
ability as an explanation, we conducted a follow-up study
(N = 131) using a 2 (method: control vs. unit asking) × 2
(victim identifiability: high vs. low) factorial design. A 2 ×
2 ANOVA yielded a main effect of method, F(1, 124) =
4.00, p < .05, which replicated the unit-asking effect, and
a main effect of victim identifiability, F(1, 124) = 4.20, p <
.05, which replicated the identifiable-victim effect; the
analysis showed no interaction between the two factors,
F < 1, suggesting that the unit-asking and the identifiable-
victim effects were orthogonal.
The unit-asking effect also differed from the foot-in-
the-door (FITD) effect, the phenomenon that people are
more likely to comply with a large request if they previ-
ously have complied with a small request (Freedman &
Fraser, 1966; Guéguen & Jacob, 2001). Although both
effects capitalize on respondents’ need for consistency,
the FITD effect capitalizes on respondents’ desire to be
attitudinally consistent (e.g., consistent about whether to
donate money for the poor given that they have publicly
expressed their sympathy for the poor), whereas the
unit-asking effect capitalizes on respondents’ desire to be
analytically consistent (e.g., consistent about how much
to donate for 20 children given that they would donate
$10 for one child). Furthermore, FITD typically works in
situations in which, at the time respondents agree to the
small request, they do not know that the requester will
make a large request later, and in situations in which
respondents make their decisions in front of the requester
and, thus, are under social pressure to be consistent.
Conversely, unit asking works even in situations in which
respondents know in advance that the unit-asking ques-
tion is only a hypothetical precursor, and in situations in
which respondents make their decisions in private, with-
out external pressure for consistency.
Study 2
Method
Study 2 was a field experiment in which we applied the
unit-asking method and replicated the unit-asking effect
in a fundraiser of a midsized company (with approxi-
mately 800 employees) in China. The company conducted
annual fund-raisers among its employees that were
designed to assist needy individuals. The employees
made their donations on a company-designated Web site,
and the company sent the donations to the target organi-
zation. We administered our study in conjunction with
one of these fund-raisers. This particular fundraiser aimed
to help 40 students from low-income families at an ele-
mentary school in rural Sichuan, a school the company
had sponsored following the region’s 2008 earthquakes.
The procedure of the study was similar to that fol-
lowed in Study 1 except that the process involved a real
fundraiser and only the two main conditions, control and
unit asking. The company sent a mass e-mail to all of its
employees announcing the fundraiser and encouraging
each employee to help the 40 students by making a
donation on the designated Web site within 1 week of
the announcement. Within 1 week of the announcement,
320 employees (36% female, 64% male; mean age = 32.1
years) visited the Web site. Unbeknownst to the visitors,
some of them were directed to a control version and
some to a unit-asking version of the solicitation.
Results and discussion
Unit asking made a significant difference in total WTD;
mean total WTD was approximately 65% higher in the
unit-asking condition (M = 326.69 yuan) than in the con-
trol condition (M = 197.37 yuan), t(315) = 2.70, p < .01,
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
Unit Asking 5
d = 0.31 (at the time of the study, 1.00 yuan was worth
approximately $0.16). Because this study involved real
donations and because some employees visited the Web
site without making donations, we also analyzed the data
after removing such nondonors and still found a robust
unit-asking effect: Mean total WTD was approximately
90% higher in the unit-asking condition (M = 601.30
yuan) than in the control condition (M = 315.13 yuan),
t(186) = 4.12, p < .0001, d = 0.63. Unit asking did not
influence the percentage of Web-site visitors who made
donations (62.6% vs. 54.3%), χ
2
(1, N = 317) = 2.17, n.s.
We should note that the employees may have felt obli-
gated to make donations because the solicitation came
from their company. However, the fact that the solicita-
tion came from their company would have influenced
only the participation rate or the amount of donations,
not the unit-asking effect, namely, the difference between
the unit-asking and the control conditions.
Study 2 replicated the unit-asking effect in a real fun-
draiser. As in Study 1, respondents in the unit-asking con-
dition knew in advance that their donation was for all of
the students and that the unit-asking question was hypo-
thetical. Moreover, respondents made their decisions in
private and could revise their responses before submis-
sion. Thus, it was unlikely that the unit-asking effect was
a result of confusion or social pressure.
Study 3
Method
To further replicate the unit-asking effect and to rule out
external social pressure as a potential cause, we ran
another field experiment. Unlike Study 2, Study 3 involved
a paper-based fundraiser carried out in anonymity. We
conducted the study in conjunction with the doctoral-
studies office of a large business school in China. The
office solicited donations from a group of 101 wealthy
executive-business-administration students (15% female,
85% male; mean age = 40.8 years) to support the research
of 70 doctoral students. Potential donors (the executive-
business-administration students) received a printed form
and an envelope and were told that those wishing to
donate should fill out the form anonymously, place it in
the envelope along with their (cash) donation, and drop
it at a designated location within a week. We designed
the form so that it had a control and a unit-asking ver-
sion, as in Studies 1 and 2. Unbeknownst to the respon-
dents, some respondents received one version and the
rest received the other version. The respondents were
reminded that they had to fill out the form if they wished
to donate, that the amount they said they would donate
on the form had to match the amount placed in the enve-
lope, and that they should not put their names on
the form.
Results and discussion
Again, unit asking greatly increased donations. Mean
total WTD in the unit-asking condition (M = 746.94 yuan)
was more than 4 times as high as that in the control con-
dition (M = 168.20 yuan), t(97) = 2.39, p < .05, d = 0.48.
When we excluded those who did not make donations,
we still found a strong effect: Mean total WTD was more
than 5 times as high in the unit-asking condition (M =
2588.89 yuan) as in the control condition (M = 442.63
yuan), t(35) = 3.20, p < .01, d = 1.08. As in Study 2, unit
asking did not affect the percentage of respondents who
made donations (38.0% vs. 34.7%), χ
2
(1, N = 99) < 1, n.s.
Because this study was conducted anonymously, the
results further ruled out demand or social pressure as an
explanation for the unit-asking effect.
It is notable that this study exhibited a greater unit-
asking effect than did the other two studies. One possible
reason for this result is that the number of needy persons
in this study (i.e., 70) was greater than that in Study 2
(i.e., 40) and that in Study 1 (i.e., 20; see further discus-
sion in the General Discussion section). Another possible
reason for the greater unit-asking effect is that partici-
pants in this study were more affluent and had fewer
budgetary constraints than did participants in the other
studies. We surmise that the size of the unit-asking effect
depends, at least in part, on the elasticity of the amount
donors can contribute, which in turn depends on their
level of wealth. The unit-asking effect should enjoy a
greater upward potential among affluent donors than
among financially restricted donors.
General Discussion
The present work joins an increasing body of psychologi-
cal research that has applied psychological science to the
creation of strategies and policies to improve the welfare
of people in real life (e.g., Arkes & Gaissmaier, 2012;
Cialdini, 2003; Gneezy, Gneezy, Nelson, & Brown, 2010;
Herzog & Hertwig, 2009; Johnson & Goldstein, 2003; Liu
& Aaker, 2008; Schultz, Nolan, Cialdini, Goldstein, &
Griskevicius, 2007; Smith, Faro, & Burson, 2013; Thaler &
Sunstein, 2009). Specifically, the current work applies
existing psychological research on scope insensitivity
and need for consistency to the creation of a simple and
costless fundraising method and demonstrates the effec-
tiveness of the method in both scenario-based and field
experiments. The remainder of the article discusses the
implications of our research.
Increasing scope sensitivity
Besides increasing total donations, unit asking carries
another potential benefit—increasing scope sensitivity. A
common challenge in fund-raising is that donors are
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
6 Hsee et al.
insensitive to the number of people in need of help; they
tend to donate similar amounts regardless of whether the
number in need is 10 or 100 (Baron & Greene, 1996;
Cameron & Payne, 2011; Desvousges et al., 1993). On the
basis of the notion that people are initially scope insensi-
tive and subsequently scope consistent, we hypothesized
that unit asking could alleviate the scope-insensitivity
problem. When respondents in the unit-asking condition
are asked to indicate their unit WTD for one needy per-
son, they will likely offer the same amount, regardless of
whether the total number of needy persons is 10 or 100.
Then, when asked to decide on their total WTD for all of
the needy persons, they will likely donate more if the
total number is 100 than if the total number is 10. In other
words, the unit-asking question sets a common unit-level
baseline by which to decide how much to donate for N
persons, hence heightening respondents’ sensitivity to
the size of N.
We tested this hypothesis in a recent study (N = 319;
Hsee, Zhang, & Lu, 2013). The procedure was similar to
that of Study 1 except that it adopted a 2 (method: con-
trol vs. unit asking) × 2 (scope: 10 vs. 100 children)
design. Replicating the unit-asking effect, results from an
ANOVA on total WTD revealed a main effect of method,
F(1, 310) = 15.25, p < .0001, and no main effect for scope,
F(1, 310) = 1.89, n.s. More important, the analysis yielded
a Method × Scope interaction effect, F(1, 310) = 4.15, p <
.05, revealing greater scope sensitivity in the unit-asking
condition than in the control condition. It is intriguing to
note that unit asking takes advantage of people’s initial
scope insensitivity to increase their eventual scope
sensitivity.
We should also note, however, that the ability of unit
asking to increase scope sensitivity is likely limited; if the
target numbers are large—for example, 1,000 versus
10,000—respondents may encode either number as “a
lot” and not differentiate the two. Moreover, large num-
bers may even reduce donations (e.g., Cameron & Payne,
2011; Fetherstonhaugh, Slovic, Johnson, & Friedrich,
1997; Slovic, 2007). For a discussion of possible shapes
of the donation functions under the unit-asking and con-
trol conditions, see the Supplemental Material available
online.
Other applications and implications
The unit-asking method may be applied to domains other
than fund-raisers. For example, auctioneers may elicit a
higher final bid for a set of 12 wine glasses if they first ask
prospective bidders to consider how much they would
pay for 1 of the glasses than if they ask only how much
bidders would pay for the whole set. Likewise, psychol-
ogy professors may obtain more funding from their
department chair for running a set of eight experiments if
they first ask the chair how much he or she would fund
one such experiment than if they ask only how much the
chair would fund the whole set of experiments.
We conjecture that the unit-asking method probably
works better in some situations than in others and would
like to see future research devoted to the following spec-
ulations. First, because unit-asking capitalizes on initial
scope insensitivity, which in turn capitalizes on lack of
evaluability, the method would perhaps work better if the
need of the target persons (or the value of the target
items) were hard to determine. For example, the method
may exert a greater effect on fund-raising if the donors
do not know exactly how much each needy person
needs. Second, the method would probably work better
if the number of target persons (or the number of target
items) were specified and comprehensible (e.g., 30 chil-
dren) than if it were unspecified (e.g., a group of chil-
dren) or incomprehensibly large (e.g., 3,000 children).
Third, although unit asking works even without social
pressure for consistency, it would probably be more
effective in the presence of social pressure. For example,
the method may create a greater effect on donations if
the requester is present (as in person-to-person fund-
raisers) as opposed to not present (as in Web-based
fund-raisers). Finally, the method would likely work bet-
ter on donors who have not experienced the method or
have forgotten it than on donors who just encountered
the method because the latter donors may discount the
unit-asking question.
Fund-raising solicits hundreds of billions of dollars
annually, yet it is effortful and costly. As our studies dem-
onstrate, unit asking can significantly raise donations
without adding extra costs. Therefore, even if only a
small fraction of fund-raisers adopts the method, it can
increase total donations by a large amount. A subtle
manipulation based on psychological science can make a
substantial difference in real life.
Author Contributions
All the authors, especially C. K. Hsee, contributed to develop-
ing and designing the studies. J. Zhang and Z. Y. Lu did most
of the work collecting and analyzing data. C. K. Hsee drafted
the manuscript; J. Zhang, Z. Y. Lu, and F. Xu provided critical
revisions. All authors approved the final version of the manu-
script for submission.
Acknowledgments
The authors thank their respective schools and the Templeton
Foundation for research support and thank the following indi-
viduals (in alphabetical order) for helpful comments: Laura
Faas, Uri Gneezy, David Levari, Tony Lian, Carey Morewedge,
Luxi Shen, Yanping Tu, Adelle Yang, and Yang Yang.
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
Unit Asking 7
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with
respect to their authorship or the publication of this article.
Supplemental Material
Additional supporting information may be found at http://pss
.sagepub.com/content/by/supplemental-data
Note
1. Because WTD had no upper bound, some responses were
very high. To prevent the influence of outliers, we excluded
responses that were more than 3 standard deviations from the
mean from all of the studies before analyses. We also analyzed
the results by Winsorizing the data at the 5%-to-95% level and
found the same pattern of results in all of the studies.
References
Ariely, D., Loewenstein, G., & Prelec, D. (2003). Coherent arbi-
trariness: Stable demand curves without stable preferences.
The Quarterly Journal of Economics, 118, 73–106.
Arkes, H. R., & Gaissmaier, W. (2012). Psychological
research and the prostate-cancer screening controversy.
Psychological Science, 23, 547–553.
Baron, J., & Greene, J. (1996). Determinants of insensitivity to
quantity in valuation of public goods: Contribution, warm
glow, budget constraints, availability, and prominence.
Journal of Experimental Psychology: Applied, 2, 107–125.
Cameron, C. D., & Payne, B. K. (2011). Escaping affect: How
motivated emotion regulation creates insensitivity to mass
sufferings. Journal of Personality and Social Psychology,
100, 1–15.
Cialdini, R. B. (2003). Crafting normative messages to pro-
tect the environment. Current Directions in Psychological
Science, 12, 105–109.
Desvousges, W. H., Johnson, F., Dunford, R., Hudson, S.,
Wilson, K. N., & Boyle, K. J. (1993). Measuring resource
damages with contingent valuation: Test of validity and reli-
ability. In J. Hausman (Ed.), Contingent valuation: A criti-
cal assessment (pp. 91–164). Amsterdam, The Netherlands:
North-Holland.
Dunn, E. W., & Ashton-James, C. (2008). On emotional innu-
meracy: Predicted and actual affective responses to grand-
scale tragedies. Journal of Experimental Social Psychology,
44, 692–698.
Fetherstonhaugh, D., Slovic, P., Johnson, S. M., & Friedrich, J.
(1997). Insensitivity to the value of human life: A study of
psychophysical numbing. Journal of Risk and Uncertainty,
14, 283–300.
Freedman, J. L., & Fraser, S. C. (1966). Compliance without pres-
sure: The foot-in-the-door technique. Journal of Personality
and Social Psychology, 4, 195–202.
Gneezy, A., Gneezy, U., Nelson, L. D., & Brown, A. (2010).
Shared social responsibility: A field experiment in pay-
what-you-want pricing and charitable giving. Science, 329,
325–327.
Greenfield, J. M. (1999). Fund raising: Evaluating and manag-
ing the fund development process (2nd ed.). New York, NY:
John Wiley & Sons.
Guéguen, N., & Jacob, C. (2001). Fund-raising on the web:
The effect of an electronic foot-in-the-door on donation.
CyberPsychology & Behavior, 4, 705–709.
Herzog, S., & Hertwig, R. (2009). The wisdom of many in one
mind: Individual judgments with dialectical bootstrapping.
Psychological Science, 20, 231–237.
Hsee, C. K. (1996). The evaluability hypothesis: An explanation
for preference reversals between joint and separate evalua-
tions of alternatives. Organizational Behavior and Human
Decision Processes, 67, 247–257.
Hsee, C. K., & Rottenstreich, Y. (2004). Music, pandas, and
muggers: On the affective psychology of value. Journal of
Experimental Psychology: General, 133, 23–30.
Hsee, C. K., & Zhang, J. (2010). General evaluability theory.
Perspectives on Psychological Science, 5, 343–355.
Hsee, C. K., Zhang, J., & Lu, Y. (2013). [Unit asking and scope
sensitivity]. Unpublished raw data.
Jenni, K. E., & Loewenstein, G. (1997). Explaining the iden-
tified victim effect. Journal of Risk and Uncertainty, 14,
235–257.
Johnson, E. J., & Goldstein, D. (2003). Do defaults save life?
Science, 302, 1338–1339.
Kahneman, D., & Frederick, S. (2005). A model of heuristic
judgment. In K. J. Holyoak & R. G. Morrison (Eds.), The
Cambridge handbook of thinking and reasoning (pp. 267–
293). New York, NY: Cambridge University Press.
Kogut, T., & Ritov, I. (2005a). The “identified victim” effect:
An identified group, or just a single individual? Journal of
Behavioral Decision Making, 18, 157–167.
Kogut, T., & Ritov, I. (2005b). The singularity effect of identified
victims in separate and joint evaluations. Organizational
Behavior and Human Decision Processes, 97, 106–116.
Kogut, T., & Ritov, I. (2007). One of us: Outstanding willingness
to help save a single identified compatriot. Organizational
Behavior and Human Decision Processes, 104, 150–157.
Liu, W., & Aaker, J. (2008). The happiness of giving: The time-
ask effect. Journal of Consumer Research, 35, 543–557.
Moore, D. A. (1999). Order effects in preference judgments:
Evidence for context dependence in the generation of pref-
erences. Organizational Behavior and Human Decision
Processes, 78, 146–165.
National Philanthropic Trust. (2012). Charitable giving statis-
tics. Retrieved from http://www.nptrust.org/philanthropic-
resources/charitable-giving-statistics
Peters, E., Västfjäll, D., Gärling, T., & Slovic, P. (2006). Affect
and decision making: A “hot” topic. Journal of Behavioral
Decision Making, 19, 79–85.
Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J.,
& Griskevicius, V. (2007). The constructive, destructive,
and reconstructive power of social norms. Psychological
Science, 18, 429–434.
Slovic, P. (2007). If I look at the mass I will never act: Psychic
numbing and genocide. Judgment and Decision Making,
2, 1–17.
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
8 Hsee et al.
Slovic, P., Finucane, M. L., Peters, E., & MacGregor, D. G.
(2002). The affect heuristic. In T. Gilovich, D. Griffin, &
D. Kahneman (Eds.), Heuristics and biases: The psychol-
ogy of intuitive judgment (pp. 397–420). New York, NY:
Cambridge University Press.
Slovic, P., Finucane, M. L., Peters, E., & MacGregor, D. G. (2007).
The affect heuristic. European Journal of Operational
Research, 177, 1333–1352.
Small, D. A., & Loewenstein, G. (2003). Helping a victim or
helping the victim: Altruism and identifiability. Journal of
Risk and Uncertainty, 26, 5–16.
Small, D. A., & Loewenstein, G. (2005). The devil you know:
The effect of identifiability on punitiveness. Journal of
Behavioral Decision Making, 18, 311–318.
Small, D. A., Loewenstein, G., & Slovic, P. (2007). Sympathy
and callousness: The impact of deliberative thought on
donations to identifiable and statistical victims. Organi-
zational Behavior and Human Decision Processes, 102,
143–153.
Small, D. A., & Verrochi, N. M. (2009). The face of need:
Emotional expression on charity advertisements. Journal of
Marketing Research, 46, 777–787.
Smith, R. W., Faro, D., & Burson, K. (2013). More for the many:
The influence of entitativity on charitable giving. Journal of
Consumer Research, 39, 961–976.
Thaler, R. H., & Sunstein, C. R. (2009). Nudge: Improving
decisions about health, wealth, and happiness. London,
England: Penguin.
at UNIV OF MIAMI on August 17, 2013pss.sagepub.comDownloaded from
