Content uploaded by Thomas Gilovich
Author content
All content in this area was uploaded by Thomas Gilovich on Oct 09, 2014
Content may be subject to copyright.
ATTITUDES AND SOCIAL COGNITION
Buyer’s Remorse or Missed Opportunity?
Differential Regrets for Material and Experiential Purchases
Emily Rosenzweig and Thomas Gilovich
Cornell University
Previous research has established that experiential purchases tend to yield greater enduring satisfaction
than material purchases. The present work suggests that this difference in satisfaction is paralleled by a
tendency for material and experiential purchases to differ in the types of regrets they elicit. In 5 studies,
we find that people’s material purchase decisions are more likely to generate regrets of action (buyer’s
remorse) and their experiential purchase decisions are more likely to lead to regrets of inaction (missed
opportunities). These results were not attributable to differences in the desirability of or satisfaction
provided by the two purchase types. Demonstrating the robustness of this effect, we found that focusing
participants on the material versus experiential properties of the very same purchase was enough to shift
its dominant type of regret. This pattern of regret is driven by the tendency for experiences to be seen
as more singular—less interchangeable—than material purchases; interchangeable goods tend to yield
regrets of action, whereas singular goods tend to yield regrets of inaction.
Keywords: regret, experiential purchases, material purchases, life experience, inaction
Imagine you are torn between two potential purchases, each
costing around $2,000. One is a trip to Mexico; the other a new
professional-style range you have long dreamed of for your
kitchen. At one level, these might seem like rather similar purchase
decisions. Both are for pleasure—you do not need the vacation or
the new range. Both entail the same cost and will require the same
belt tightening to cover the expense. Yet one of the purchases is a
material good—made to be kept in one’s possession—whereas the
other is experiential— designed to provide an experience one lives
through. Previous research indicates that the experiential good—
the vacation—is likely to bring more enduring pleasure than the
material good. The research we present here suggests that this
difference in satisfaction is likely to be compounded by satisfac-
tion’s flip side: regret.
We investigate the possibility that material and experiential
purchases differ not only in the satisfaction they provide but in the
type of regrets they engender as well. Specifically, we predict that
material goods are more likely to result in regrets of action—
buyer’s remorse—and experiential goods are more likely to result
in regrets of inaction—the pain of a missed opportunity. Thus,
buyers who pass up experiential purchases are hit with a double
whammy: Not only do they miss out on the greater satisfaction an
experience might bring but they also are likely to realize and regret
what they missed.
The roots of this prediction lie in research on the causes of
people’s differential satisfaction with material and experiential
purchases (Carter & Gilovich, 2010). Specifically, we maintain
that, on the whole, experiences are seen as less interchangeable
than material purchases: There is a smaller set of items that feel
like effective substitutes for experiential goods. Singular experi-
ences are less likely to prompt counterfactual thoughts that focus
on upward comparisons because the class of items with which an
experience can be compared is small. Instead, the easiest and most
likely comparison is between having missed out on the experience
and not having missed out, yielding regrets of inaction. Con-
versely, the greater interchangeability of material goods affords
myriad opportunities for upward comparisons after a purchase,
making material purchases more likely to spark rumination about
alternative purchases and hence regrets of action.
Material and Experiential Purchases
Over the past decade, research has examined differences in the
amount of enduring satisfaction people derive from material and
experiential purchases (Carter & Gilovich, 2010, 2011; Nicolao,
Irwin, & Goodman, 2009; Van Boven, Campbell, & Gilovich,
This article was published Online First August 15, 2011.
Emily Rosenzweig and Thomas Gilovich, Department of Psychology,
Cornell University.
This research was supported by National Science Foundation Grant
SES-0922323. We thank Shai Davidai, Jun Fukukura, Erik Helzer, Amit
Kumar, Lily Jampol, and Dennis Regan for comments on an earlier version
of this article and Mariyah Ahmad, Jack Cao, Alisha Forster, Allison Graf,
Faaiza Khan, Elizabeth Lewis, and Susanna Li for their help in collecting
and coding these data.
Correspondence concerning this article should be addressed to Thomas
Gilovich, Department of Psychology, Cornell University, 220 Uris Hall,
Ithaca, NY 14853-7601. E-mail: tdg1@cornell.edu
Journal of Personality and Social Psychology, 2012, Vol. 102, No. 2, 215–223
© 2011 American Psychological Association 0022-3514/11/$12.00 DOI: 10.1037/a0024999
215
2010; Van Boven & Gilovich, 2003). Van Boven and Gilovich
(2003) defined material purchases as “those made with the primary
intention of acquiring a material good: a tangible object that is kept
in one’s possession” and experiential purchases as “those made
with the primary intention of acquiring a life experience: an event
or series of events that one lives through” (p. 1194). They found,
across a wide range of subject populations and time frames, that
experiential purchases tend to make people happier than material
purchases. Carter and Gilovich (2010) presented evidence for one
mechanism underlying this phenomenon, finding that people are
more likely to make invidious comparisons when it comes to
material rather than experiential purchases. That is, people dwell
on how their material purchases compare with other people’s and
how they measure up to other purchases they might have made
instead. These thoughts provide the raw material for deflating
upward comparisons that diminish satisfaction.
Our article contributes to this literature in two ways. First, it
elaborates on the ways that experiential purchases might bring
greater happiness than material goods, examining whether expe-
riential purchases are less likely to elicit regrets of action. Al-
though Carter and Gilovich (2010) advanced (but did not test) the
idea that material purchases might be more likely to result in
regrets of action than experiential purchases, the complementary
possibility (that failures to act on experiential purchase opportu-
nities are especially likely to lead to regrets of inaction) has not
been previously discussed, let alone tested. We investigate whether
this influence of purchase type exists over and above any effect
that differential satisfaction for material and experiential purchases
might have on regret. In addition, our work examines the compa-
rability explanation described above. We investigate whether ex-
periential purchases are seen as more singular and whether the
degree to which a purchase is seen as singular versus interchange-
able underlies whether regrets of action or inaction predominate.
Patterns of Regret
Foundational work on counterfactual thinking indicated that
regrets of action tend to be stronger and more common than those
of inaction because it is typically easier to imagine undoing an
action taken (and mentally returning to the status quo) than to
imagine what would have resulted from an unchosen option (Kah-
neman, 1995). Other research paints a more complex picture,
documenting a temporal shift in people’s regrets over actions and
inactions—namely, that regrets of action are more intense in the
short term, but regrets of inaction gain prominence and stand out
in the long run (Gilovich & Medvec, 1994, 1995; for an exception,
see Morrison & Roese, 2011). This work on the temporal trajec-
tory of regret demonstrates how regrets in the same domain—
one’s career, for example— can shift over time from regrets of
action (“I shouldn’t have criticized the vice president during the
board meeting”) to regrets of inaction (“I should have applied for
that position in the marketing department”).
Zeelenberg, van den Bos, van Dijk, and Pieters (2002) identified
another moderator of whether regrets of action or inaction tend to
dominate people’s experience, demonstrating that the valence of
prior outcomes (e.g., winning or losing a previous soccer game)
shape what people expect to generate the most regret—action (e.g.,
a coach changing the starting lineup for the next game) or inaction
(keeping the lineup the same). When prior outcomes were positive,
regrets of action tend to be more intense, but when prior outcomes
were negative, regrets of inaction tend to dominate. The research
we report here is designed to add texture to the existing literature
in this area, examining whether the object of the regret itself—
specifically, its material or experiential qualities— can influence
the type of regret people are most likely to experience.
Most of the research on regret has found that people’s greatest
regrets, not surprisingly, often center on major life choices such as
whom to marry, what job to take, or whether to continue with
one’s education (Gilovich & Medvec, 1994, 1995; Gilovich,
Wang, Regan, & Nishina, 2003; Hattiangadi, Medvec, & Gilovich,
1995; Landman, 1993; Roese & Summerville, 2005). Although it
has not been at the forefront of the contemporary regret literature
in psychology, everyday experience tells us that purchase deci-
sions are a frequent source of regret. Buyer’s remorse is something
most people have experienced: Children old enough to have an
allowance are old enough to regret buying a toy that delivered less
joy than anticipated. At the same time, marketers are certainly
aware of the power of regrets of inaction when they suggest that
consumers will regret missing out on a great deal or special offer.
Although economists and marketing researchers have placed more
emphasis than psychologists on the frequent connection between
purchase decisions and regret, they have largely focused on fac-
toring anticipatory regret into their models of consumer purchasing
behavior (e.g., Hetts, Boninger, Armor, Gleicher, & Nathanson,
2000; Loomes & Sugden, 1982). In this article, we focus squarely
on the experience of regret that follows consumer purchases. The
present work contributes to the regret literature by being the first
to examine the systematic ways in which the object of regret—
rather than the time frame of the regret or its valenced anteced-
ents—might shape whether it takes the form of action or inaction.
In five studies, we tested the hypothesis that when purchase
decisions lead to regret, they are more likely to lead to regrets of
action for material purchases and regrets of inaction for experien-
tial purchases. We ruled out differences in perceived desirability of
material and experiential goods as an explanation for this pattern
and investigated whether differences in regret are driven by the
tendency to see experiences as more singular (less interchange-
able) than material goods. In Study 1, we asked participants to
consider their single biggest material or experiential regret and to
indicate whether it was a regret of action or of inaction. In Study
2, we controlled for the possibility that the regrets generated in
Study 1 were for purchases of radically different types and mag-
nitudes by asking participants to consider the purchase of a mate-
rial good and an experiential good that were equated for price and
purchase domain. We also examined whether the phenomenon
extends to predictions for others as well as experiences for the self.
In Study 3, we examined our hypothesis across a broad range of
commonplace regrets, which yielded a naturalistic set of regrets
that we used to test, in Study 3A, the proposed mechanism for our
observed effects. Coders in Study 3A who were unaware of our
hypothesis rated each of the regrets generated in Study 3 on an
interchangeable–singular continuum, and we examined whether
this variable mediated our findings from Study 3. In addition, we
directly compared the potential mediating role of interchangeabil-
ity and of differential perceived value of material and experiential
goods in our reported effects. In Study 4, we manipulated inter-
changeability, examining its influence on regrets for both material
and experiential purchases. Finally, in Study 5, we pushed the
216 ROSENZWEIG AND GILOVICH
boundaries of this phenomenon, testing whether participants asked
to construe the same object as either a material or an experiential
purchase might yield the predicted differential pattern of regret.
Study 1
In this study, we tested our prediction that people’s biggest
regrets about material purchases would tend to be regrets of action,
whereas their biggest regrets about experiential purchases would
tend to be regrets of inaction.
Method
Fifty-six Cornell undergraduates were randomly assigned to
either the material or the experiential condition. They were asked
to think of times they had made or had thought about making a
material or experiential purchase and then read the following text:
Presumably, most of these purchases have worked out well for you.
Occasionally, however, we make decisions that we end up regretting.
And when we do, there are two kinds of regrets we can have. We can
regret: (1) things we did that we wish we hadn’t done, and (2) things
we didn’t do that we wish we had. When you think back on various
decisions you’ve made with respect to your [material or experiential]
purchases, what would you say is your biggest single regret?
Participants responded by checking one of two options: “a
[material or experiential] purchase I made that I wish I hadn’t,” or
“a [material/experiential] purchase I didn’t make that I wish I had.”
Results and Discussion
Twenty-four of the 29 participants (83%) in the experiential
purchase condition indicated that their biggest regret was one of
inaction, a result vastly different than the 10 out of 27 participants
(37%) in the material purchase condition whose biggest regret was
one of inaction,
2
(1) ⫽12.25, p⬍.0001, ⫽.47. Thus, as
predicted, when people thought about experiential purchase deci-
sions, regrets of inaction predominated; when they thought about
material purchase decisions, action regrets were more likely to
come to mind. But note that this design has limitations: Because
we did not control the purchases participants considered, it is
possible that the magnitude or the desirability of the experiential
and material purchases they generated were meaningfully differ-
ent. For example, research suggests that over time, experiential
purchases bring greater satisfaction than material goods (Van
Boven & Gilovich, 2003). Thus, if participants in the experiential
condition were thinking of more desirable items than participants
in the material condition, this might artifactually account for why
they were more likely to have regrets of inaction (over missing out
on a qualitatively better purchase). We addressed this concern in
Study 2.
Study 2
Study 2 provided a more controlled test of our hypothesis
because participants were asked to consider a specific material or
experiential purchase decision—from the same domain and of the
same monetary value—and to tell us whether they thought a regret
of action or inaction would be more intense. By framing these
specific purchases in the context of a third party’s choice, we could
also determine whether the pattern of results from Study 1 extends
from the self to judgments about others. To investigate the impact
of purchase desirability, we also asked participants to rate how
much they would enjoy receiving the material or experiential
purchase we described.
Method
Eighty-four participants (46 women and 38 men, ages ranging
from 18 to 61 years) were recruited through Amazon’s Mechanical
Turk and paid for their participation. Participants were randomly
assigned to read one of the following purchase scenarios:
I’d like you to imagine two people—Joe, who bought a new iPod
Shuffle ($55) but now wishes he hadn’t, and Mark, who chose not to
buy a new iPod Shuffle ($55) but now wishes he had.
I’d like you to imagine two people—Joe, who bought a ticket to a rock
concert ($55) but now wishes he hadn’t, and Mark, who chose not to
buy a ticket to a rock concert ($55) but now wishes he had.
In both conditions, participants were asked to indicate on a
7-point scale from 1 (the person who acted)to7(the person who
failed to act) which of the two individuals was more likely to
regret his decision. Participants were then asked to imagine that
someone gave them an iPod Shuffle or a concert ticket and to rate
how much they would enjoy it ona1(not at all)to7(extremely)
scale. Note that the products selected for the two scenarios be-
longed to the same general domain—music—and had the same
price attached to them. These prices were very close to the actual
product values; according to a leading music-industry trade pub-
lication, the average price for a rock concert ticket at the end of
2010 was $60 and the price of an iPod Shuffle was $55. Finally,
both purchases are common enough that we expected participants
to have no difficulty imagining the purchase and the potential for
regret.
Results and Discussion
Participants in the material condition thought that the experience
of buyer’s remorse would be more intense than the experience of
regret over a missed opportunity to buy the iPod (M⫽2.47),
one-sample t(37) against the midpoint value of 4 ⫽⫺4.88, p⬍
.000, d⫽1.60. Participants in the experiential condition, in con-
trast, predicted that action and inaction regrets for the concert
purchase would be equal (M⫽4.00), t⬍1. The difference
between the ratings of the two groups of participants was signif-
icant, unequal variances t(82) ⫽3.28, p⫽.002, d⫽0.71.
Critically, there was no difference in how much participants felt
they would enjoy receiving the iPod (M⫽5.16) versus the concert
ticket (M⫽5.37), t⬍1. Furthermore, controlling for the desir-
ability of the iPod or concert ticket left the effect of condition
unchanged, ⫽⫺1.50, t(81) ⫽⫺3.15, p⫽.002.
These data thus further support our hypothesis using a paradigm
in which the material and experiential purchases were from the
same domain and of equal monetary value. They augment the
results from Study 1, showing that people not only expect their
own regrets to be different for material and experiential purchases
but also expect the same pattern for the regrets of others. These
data also suggest that the effect is not driven by differences in the
217
MATERIAL AND EXPERIENTIAL PURCHASE REGRETS
perceived desirability of material and experiential goods—a find-
ing that receives further support in Study 3A and Study 5.
Study 3
Study 3 was designed to extend the results of Study 1. Partici-
pants were asked to generate a number of specific regrets from
their own lives and to indicate whether each was a regret of action
or inaction (rather than to specify their single biggest regret). We
also used this set of naturalistic material and experiential regrets in
further analyses (in Study 3A) to shed light on the mechanism
underlying our findings.
Method
Seventy-five Cornell undergraduates participated in exchange
for extra course credit and were randomly assigned to either the
material or the experiential condition. After reading a definition of
a material purchase (“a tangible object that you buy and keep in
your possession”) or an experiential purchase (“a life experience”),
participants read that regrets about purchases can come in the form
of regrets of action or inaction. Specifically, the text read as
follows:
Presumably you’ve been happy with most of your decisions about
whether or not to buy [material or experiential] objects. Occasionally,
however, we make decisions that we end up regretting. Sometimes,
for example, we spend money on something and afterwards realize the
purchase was a mistake, and we end up regretting the action we took.
Other times we don’t make a [material or experiential] purchase that
we had thought about; afterwards we realize that we should have
made the purchase and we end up regretting our inaction.
Participants were then asked to list three specific material or
experiential regrets that came to mind. After they wrote them
down, we asked them to go back and label each with an Afor
action or an Ifor inaction.
Results and Discussion
Two participants were excluded because they did not provide
specific regrets, instead using generalities such as “buying an
object that you get sick of.” Another participant was excluded
because he accidentally participated twice, completing both con-
ditions. Of the remaining 72 participants, the mean number of
regrets of action was significantly higher for participants in the
material purchase condition (M⫽2.24) than in the experiential
purchase condition (M⫽1.54), t(70) ⫽3.48, p⬍.001, d⫽0.83,
and the reverse was (of course) true for regrets of inaction. These
results demonstrate that even fairly commonplace purchase deci-
sions gone awry—such as bad meals, clothes participants never
wore, campus concerts, and computer games— conform to the
same differentiated pattern of regrets for material and experiential
purchases: People experience more regrets of action over material
goods and more regrets of inaction over experiential goods. These
findings are consistent with those of Study 1 even though partic-
ipants were asked to list several regrets and did so prior to
encoding them as regrets of action or inaction.
Study 3A
Previous work by Carter and Gilovich (2010) found that part of
the reason experiential goods bring greater happiness is that they
are less likely to elicit invidious comparisons to other goods,
resulting in less time spent thinking about other purchases that
might have been better than the chosen option. Carter and Gilovich
noted that experiences are less likely to spark such comparisons
because, being less tangible, they are literally harder to liken to one
another—it is harder to line up and compare, feature by feature, the
different possible options. Good luck comparing the ambience at
Tetsuya’s with that at Charlie Trotter’s,the quality of light at
Bondi versus Hanalei, or the view from the Blue Mountains versus
Muir Pass.
In Study 3A, we pursued this idea further, examining whether
experiences tend to be seen as more sui generis— of their own
kind, or unique—than material goods and, in turn, whether mate-
rial goods are seen as more interchangeable. The idea of compa-
rability that Carter and Gilovich (2010) referred to reflects whether
a purchase has features that can be easily aligned and compared
with others in its class. Here we focus on the notion of interchange-
ability, which reflects the size of the class and the uniqueness of its
members. Are there many members of the class and are they seen
as ready substitutes for one another? We maintain there is a
smaller set of things that provide the same benefits of the average
experience than there is of the average material good.
How might this influence the type of regrets people are likely to
have over material and experiential purchases? When goods are
interchangeable with others of the same type, regrets of action
become more likely, as there is a large pool of alternatives with
which to compare the purchase—any of which might look more
appealing if the current purchase does not meet expectations.
Conversely, when a good is seen as singular or unique, regrets of
action are less likely because it is harder to think of a counterfac-
tual world in which a better outcome would have resulted if only
a different purchase had been made. Instead, when the items or
events in question are not interchangeable, it is likely to be failures
to act that stand out, as the individual comes to the realization that
a unique opportunity has passed.
Returning to the stimuli used in Study 2, a great many products
serve the same function as an iPod shuffle, including similar
models from Sony or Samsung and a variety of smartphones that
play music. In contrast, there are many fewer acceptable substi-
tutes for a specific rock concert. Although there are certainly other
concerts one could attend, even artists in the same genre do not
provide the same experience: Performances by David Byrne and
Regina Spektor just do not feel interchangeable. We contend that
people see experiential goods, on the whole, as less interchange-
able than material goods. We tested this hypothesis in this study by
having coders evaluate each of the purchases participants listed in
Study 3 on the dimension of interchangeability. We then tested
whether the interchangeability of the items or events in question is
related to whether the purchaser’s regret was one of action or
inaction. We predicted that the interchangeability of the items
would mediate the relationship between type of purchase (material
vs. experiential) and type of regret (action vs. inaction). We also
examined whether differential desirability of material and experi-
ential purchases may have played a role in the different types of
218 ROSENZWEIG AND GILOVICH
regrets people tend to experience over material and experiential
purchase decisions.
Method
Three research assistants who were unaware of our hypothesis
and previous findings coded the full set of regrets generated in
Study 3 on the dimension of interchangeability. More specifically,
they read the following text:
Some things you can purchase are largely interchangeable—there are
many other things just like it that could substitute and serve essentially
the same function. Things that are interchangeable are easily replace-
able. Other things you can purchase are much more singular—there
are not many things like it or that would be a good substitute. Things
that are singular feel unique and hard to replace.
For each of the purchasing decisions you read about, we’d like you to
rate the object or experience for how interchangeable it is. Please use
a scale between 1 and 5, where the values mean the following:
1⫽Completely Interchangeable
2⫽Mostly Interchangeable
3⫽Somewhat Interchangeable
4⫽Not Very Interchangeable
5⫽Not Interchangeable At All
The responses from the 72 participants in Study 3, who each
provided three regrets, were given to the coders in a single random
order. The coders were given the regrets exactly as written by the
Study 3 participants and all 216 regrets were rated by each coder.
A different set of three coders, who were also unaware of our
hypothesis and previous findings, coded the full set of purchases
generated by participants in Study 3 on the dimension of desir-
ability. Coders were asked to rate the object or experience by
answering the question “Cost aside, how desirable would this be to
the average Cornell student? How much would they enjoy it?” The
5-point scale was anchored at 1 ⫽very little and 5 ⫽extremely.
The responses from the 72 participants in Study 3 were stripped of
any reference to regret, leaving only the purchase description. A
list of these purchases was given to the coders in a single random
order, and all 216 purchases were rated by each coder.
Results and Discussion
The ratings made by both sets of coders were reliable (inter-
changeability ␣⫽.77, desirability ␣⫽.81) and so they were
averaged to create two indices: one of the interchangeability and
one of the desirability of each purchase.
1
As predicted, material
purchases were rated as significantly more interchangeable than
experiential purchases (M
material
⫽2.09, M
experiential
⫽3.14), ⫽
.624, t(214) ⫽⫺11.68, p⬍.0001. In addition, the more inter-
changeable a purchase was, the more likely the regret associated
with it was one of action, ⫽.374, t(214) ⫽5.87, p⫽⬍.0001.
2
As detailed in Study 3, material purchases were significantly more
likely to result in regrets of action than experiential purchases, ⫽
.248, t(214) ⫽3.73, p⬍.001. To test whether interchangeability
mediated the relationship between purchase type and regret type,
we used the Baron and Kenny (1986) procedure, with the correc-
tion specified by MacKinnon and Dwyer (1993) to account for the
fact that our dependent variable (regret type) was dichotomous.
Interchangeability fully mediated the relationship between pur-
chase type and regret type, Sobel Z⫽3.76, p⬍.001, such that
when interchangeability was included in the model, purchase type
was no longer a significant predictor, ⫽.02, p⬎.7.
Consistent with the findings reported by Van Boven and Gilov-
ich (2003), the experiential purchases that participants listed were
rated as significantly more desirable (M⫽3.75) than their material
purchases (M⫽3.17), t(211) ⫽⫺4.35, p⬍.0001. In addition,
desirability was a significant predictor of regret type, with espe-
cially desirable purchases being more likely to be associated with
regrets of inaction (⫽.139, t(209) ⫽4.37, p⬍.0001. However,
desirability did not mediate the relationship between purchase type
(material or experiential) and regret type: Purchase type remained
significant when desirability was included in the model,

condition
⫽.115, t(208) ⫽3.49, p⫽.001; 
desirability
⫽.173,
t(208) ⫽2.66, p⫽.008. Finally, interchangeability continued to
mediate the effects of condition on regret type even when desir-
ability was included as a covariate in the model, Sobel z⫽2.78,
p⬍.005. These findings thus support our contention that it is
interchangeability, not desirability, that drives the relationship
between material and experiential purchases and regret type (see
Figure 1).
Although not the focus of this article, the reasons that experi-
ential purchases are seen as less interchangeable than material
purchases are worth considering. One reason doubtless stems from
the ephemeral nature of experiences. The fact that they do not
persist makes it both harder to compare them with foregone
experiential purchases and harder to imagine “returning” one and
having a different experience instead. Another likely reason is that
experiences feel closer to the self than material goods, and their
close association with the self makes them seem more unique
(Carter & Gilovich, 2011).
Study 4
To conduct a more controlled test of the importance of inter-
changeability in the type of purchase regrets people are likely to
experience, we manipulated the interchangeability of both a ma-
terial purchase and an experiential purchase. Both material and
experiential purchases vary along the dimension of interchange-
1
A fourth coder also rated the interchangeability of the full set of
purchases, but her ratings were poorly correlated with those of the other
three (average r⫽.15) and reduced the overall reliability of the four sets
of ratings to .61. The analyses above were therefore conducted using just
the ratings of the three reliable coders, but the findings and pvalues do not
change if all four coders’ ratings are included in the composite index of
interchangeability.
2
When material and experiential goods are analyzed separately, inter-
changeability is a highly significant predictor of regret type for experiential
goods, ⫽.386, t(103) ⫽4.25, p⬍.0001, but it is a much weaker
predictor of regret type for material goods, ⫽.154, t(107) ⫽1.61, p⫽
.11. This appears to be caused by the fact that the range of interchange-
ability ratings for material goods (Min 1 to Max 3.33) was much smaller
than the range for experiential goods (Min 1 to Max 4.67).This underscores
our contention that not only are material goods as a whole seen as
significantly more interchangeable than experiences, but they are also less
variable in how interchangeable they seem.
219
MATERIAL AND EXPERIENTIAL PURCHASE REGRETS
ability. Dinner out at a local chain restaurant is a fairly inter-
changeable experience (if you cannot get into Chili’s, you can
always eat at Applebee’s), whereas dinner at a local Ethiopian
restaurant is much less so. Similarly, a summer sundress is a fairly
interchangeable material good (many dresses would have roughly
the same appeal), but a wedding dress is more singular: Most
women do not think that just any old wedding dress will do. Given
that interchangeability predicted regret type for both material and
experiential goods in Study 3A, we expected that making either
type of purchase less interchangeable (more singular) would in-
crease the likelihood that it would elicit regrets of inaction.
Method
Sixty-six participants (38 women, 28 men, mean age ⫽34
years) were recruited through Amazon’s Mechanical Turk and
paid for their participation. Participants were randomly assigned to
one cell of a 2 ⫻2 design. The 36 participants in the material
purchase condition were asked to imagine that they were looking
to buy a dresser and were randomly assigned to read about either
a dresser they found at the local mall (interchangeable) or an
antique dresser they found at an estate sale (singular). The 30
participants in the experiential purchase condition were asked to
imagine that they were trying to decide whether to buy a plane
ticket to either their family’s yearly (interchangeable) or their
family’s first ever (singular) reunion getaway in California. All
participants were asked to consider the two types of regret—
purchasing the dresser or ticket and wishing they had not or not
purchasing it and wishing they had—and to indicate on a 1–7 scale
which regret would be stronger, from 1 (buying the [dresser or
ticket])to7(not buying the [dresser or ticket]). A separate sample
of participants from Mechanical Turk rated the four scenarios
using the same interchangeability scale described in Study 3A.
These participants confirmed that the dresser from the mall is
considered more interchangeable than the antique dresser, t(19) ⫽
6.75, p⬍.0001, and that the ticket to the annual family reunion is
considered more interchangeable than the ticket to the first-time
reunion t(19) ⫽2.46, p⬍.03.
Results and Discussion
As predicted, there was a main effect of interchangeability, such
that participants were more likely to believe that they would
experience regrets of inaction for purchases that were singular than
for purchases that were interchangeable, F(62) ⫽19.59, p⬍
.0001,
p
2
⫽.24. Participants who read about the dresser framed as
a singular purchase thought the regret of inaction would be stron-
ger (M⫽4.94) than did those who read about the dresser framed
as an interchangeable good (M⫽2.39), t(62) ⫽3.65, p⬍.001,
d⫽0.927. Similarly, participants who considered buying a ticket
to fly home for a first-time family reunion thought the regret of
inaction would be stronger (M⫽5.43) than did those who read
about buying a ticket to the annual event (M⫽3.38), t(62) ⫽2.67,
p⬍.01, d⫽0.68. Neither the main effect of purchase type nor the
interaction between interchangeability and purchase type was sig-
nificant, both ps⬎.15.
Study 5
Study 5 was designed to explore the boundaries of our main
finding by examining whether the very same object, when viewed
through a material or an experiential lens, might yield different
patterns of regret. Although many purchases are unambiguously
material or experiential, others straddle the line between the two
categories, having both material and experiential properties. In
Study 5, we manipulated whether one such ambiguous pur-
chase—a 3-D TV—was framed either as an experience or as a
material good. As in Study 2, participants were then presented with
two individuals who regretted their decision and were asked whose
regret would be stronger—the person who made the purchase or
the person who did not. We expected that focusing on the expe-
riential features of what is generally seen as a material good would
increase the predicted likelihood and strength of regrets of inac-
tion. We also included questions that would allow us to examine
whether our framing manipulation had an effect on perceived
product value and whether any such effect might have artifactually
yielded the predicted difference in type of regret.
Method
Sixty-two participants (33 men, 29 women) were recruited
through Amazon’s Mechanical Turk and paid for their participa-
tion. Participants were randomly assigned to either the material or
the experiential condition. In the material scenario, Mark (who
ultimately bought the TV) and Joe (who did not buy the TV) were
both described as “imagining where a 3-D TV set would go in their
apartments, what it would look like, and what their friends would
think.” In the experiential condition, Mark and Joe were described
as “imagining the fun they’d have watching it with friends, and
how cool it would be to experience TV in a whole new way.” In
both conditions, participants read that “Mark ended up buying one,
but for various reasons now wishes he hadn’t. Joe did not buy one,
but for various reasons now wishes he had.” Participants were
asked to indicate which person would regret their decision more on
a 1–7 scale, with 7 representing more regret on the part of Joe, who
chose not to buy, and 1 representing more regret on the part of
Mark, who chose to buy. Finally, participants were then asked to
imagine that they were in the market for a new TV and to report
how much they would be willing to pay for a 3-D TV. They also
indicated how much they would enjoy owning a new 3-D TV and
how much satisfaction a new 3-D TV would bring them, on scales
Figure 1. The mediating role of interchangeability on the relationship
between type of purchase and type of regret. The beta weight in parenthe-
ses reflects the value for type of purchase when the mediator is included in
the regression. The Sobel Zwas calculated using the MacKinnon and
Dwyer correction for mediation with dichotomous outcome variables.
ⴱⴱⴱ
p⬍.001.
220 ROSENZWEIG AND GILOVICH
ranging from 1 (not at all/none)to5(extremely/an extreme
amount).
Results and Discussion
Participants who read about the 3-D TV framed as a material
good thought the regret of action would be stronger (M⫽2.10)
than did those who read about the 3-D TV framed as an experience
(M⫽3.34), unequal variances t(60) ⫽2.28, p⫽.02, d⫽0.62.
Predicted enjoyment and satisfaction were highly correlated (r⫽
.89) and so we averaged them together to form an index of product
desirability. We also calculated the natural log of the prices par-
ticipants indicated they were willing to pay for a 3-D TV to
normalize that distribution. Our framing manipulation did not
significantly influence perceived desirability or willingness to pay
(both ps⬎.6). Furthermore, purchase framing remained a signif-
icant predictor of regret type when both desirability and log pay
were included in the relevant multiple-regression analysis, ⫽
.129, t(54) ⫽2.425, p⫽.02.
3
Note that the mean ratings of both
groups were below the midpoint, indicating more overall antici-
pated regret of action. This may reflect participants’ familiarity
with buyer’s remorse when it comes to the latest and greatest of
new technologies—technologies that are often quickly rendered
obsolete. Nevertheless, that the framing of the purchase to focus on
its experiential properties shifted its regret profile suggests that
keeping an object’s experiential properties in mind when making
(and later evaluating) a purchase might lead to less buyer’s re-
morse.
General Discussion
Understanding and predicting possible regrets is an important
part of extracting as much satisfaction and pleasure from our
purchasing power as possible. Regrets, whether of action or inac-
tion, are painful, and their pain needs to be factored into the
hedonic equation underlying people’s purchasing decisions. Our
research suggests that when it comes to such decisions, the regret
people are most likely to experience is indeed predictable—
broadly by whether the purchase is a material or an experiential
good and more narrowly by how interchangeable the purchase is
with others in its class. This knowledge might make it easier to
avoid some purchase regrets in the first place: Tilt toward expe-
riences over material goods when the two types of expenditures are
in close competition and there are not enough funds to cover both.
Studies 1 and 3 demonstrate that people are prone to different
types of regret for material and experiential purchases, both with
respect to subjects’ greatest purchase regrets and with respect to
more mundane, everyday regrets that they supplied for us. Partic-
ipants in Study 2 anticipated this same pattern of regrets for others,
even when the magnitude and domain of the material and experi-
ential purchases were held constant. Focusing participants’ atten-
tion in Study 5 on either the material or the experiential features of
the very same purchase also yielded this differential pattern of
regret. In Study 3A, we coded the specific regrets listed by par-
ticipants in Study 3 to test a mechanism responsible for these
effects. The experiential purchases participants generated in Study
3 were rated as less interchangeable than the material purchases
they described, and this difference in interchangeability mediated
the relationship between purchase type (material or experiential)
and regret type (action or inaction). To gain experimental control
over this mediator, in Study 4, we presented participants with one
of two scenarios in which material or experiential purchases were
more or less interchangeable. The interchangeability of the pur-
chase predicted whether participants thought it would elicit action
or inaction regrets, independent of whether it was an experience or
a material good.
These studies extend the literature on regret, illuminating the
objects of regret—rather than the time frame in which they are
evaluated—as a determinant of whether action or inaction regrets
are likely to predominate. We chose to study this in the realm of
consumer purchases, a context that often generates regret but has
not (surprisingly) received much attention in the contemporary
psychological literature. Our work also complements existing re-
search on material and experiential goods, showing that the pre-
viously documented tendency for experiences to induce more
enduring satisfaction than possessions is mirrored in satisfaction’s
flip side: regret. Experiences tend to provide more satisfaction than
material goods, and the failure to realize experiences tends to elicit
regrets of inaction. Conversely, not only do material purchases
typically lead to less satisfaction than experiential goods but they
are more likely to lead to outright regret over having made the
purchase in the first place. These findings thus parallel the message
that experiential purchases yield more satisfaction than material
goods, constituting an analogous result on what is, in essence, an
additional measure of satisfaction: type of regret.
Role of Interchangeability in Counterfactual
Thinking and Regret
Research on regret is closely linked to the literature on coun-
terfactual thinking, with the signature finding being that people are
most likely to regret a negative outcome when it is easy to imagine
counterfactual states of the world in which the outcome would
have been better (Kahneman & Tversky, 1982; Miller & Taylor,
1995; Roese & Olson, 1995). This ease of counterfactual genera-
tion is driven by the perceived mutability of the event: How easily
can the outcome or its antecedents be mentally undone (Kahne-
man, 1995)? Past research has focused on such determinants of
mutability as departures from normality (Kahneman & Miller,
1986), position in a temporal sequence (Miller & Gunasegaram,
1990), and proximity to a notable outcome (Medvec, Madey, &
Gilovich, 1995). The interchangeability of an item or event is
another determinant of mutability, influencing the likelihood that
an individual will consider alternative purchases that might have
been made. The extent to which a purchase is seen as one of a
broad set of substitutable items or as something more one of kind
influences how easy it is to imagine a counterfactual world in
which a different and perhaps better purchase was made. And as
we have shown, this has implications for the types of regrets
people tend to have over their purchase decisions. There are, no
doubt, other features of purchasing decisions, such as the extent to
which the buyer deliberated over a purchase or chose mindlessly,
3
Note that desirability was a significant predictor of regret type, such
that the more participants rated the 3-D TV as desirable, the more likely
they were to believe that regrets of inaction would predominate, ⫽.533,
t(54) ⫽2.21, p⫽.03.
221
MATERIAL AND EXPERIENTIAL PURCHASE REGRETS
or whether specific alternative purchases were considered, that
might also influence the likelihood of regret. It is unclear, how-
ever, whether these other determinants of purchase mutability
would differ across material and experiential goods and hence
influence the type of regret that each type of purchase tends to
elicit.
Limited and Lost Opportunities
Two recent articles highlight the importance of lost opportuni-
ties in determining the nature and intensity of regret. Beike,
Markman, and Karadogan (2009) have put forward evidence that
people’s most intense regrets are ones involving lost opportunities
and that the life domains that produce the greatest number of
regrets (e.g., education) are those in which people perceive fewer
opportunities in the future. The authors did not distinguish between
regrets of action and inaction, and previous research on the tem-
poral aspects of regret indicates that regrets in the domains they
reference can be of either type. It is likely, furthermore, that part
of what makes an opportunity feel truly lost—and thus what
amplifies the regret—is how interchangeable it is. For example, a
regret of inaction over a lost opportunity to study abroad in Kenya
might be more intense than regret over a failure to study abroad in
England because, for most U.S. citizens at least, time spent in
England seems more interchangeable with other experiences they
might have. Even with respect to regrets of action, regret intensity
might be moderated by the interchangeability of the action that
now cannot be remedied. Regrets about majoring in English might
be stronger than regrets about majoring in medieval architecture
because although both resulted in the lost opportunity of choosing
a major with better job prospects, the English major— being more
commonplace—feels more interchangeable with other popular but
more practical majors.
Recent marketing research suggests that in the case of limited-
opportunity purchases, the established temporal pattern of regret
can flip, such that regrets of inaction dominate in the short term
and regrets of action can grow stronger as time passes after the
purchase (Abendroth & Diehl, 2006). This work has several inter-
esting points of intersection with what we report here. First, it is
notable that Abendroth and Diehl (2006) based their conclusions
on three material purchases that are all essentially markers of an
experience (souvenirs from a vacation, a live concert CD, and a
concert t-shirt from a performance the participant imagined attend-
ing). The fact that these purchases, even in the short term, elicited
regrets of inaction underscores our findings from Study 5 in which
the framing of a (primarily) material purchase in experiential terms
influenced the type of regret it provoked.
The idea of limited opportunity is certainly related to inter-
changeability—the two often go hand in hand. Indeed, limited
purchasing opportunities often derive their power from the degree
to which other items cannot serve as substitutes. There is no great
loss in a limited opportunity to buy a Samsung TV—perhaps the
model is being closed out—if a comparable Sony TV remains
available. On the flip side, imagine that you live in San Diego and
can visit Sea World whenever you want. While at Sea World, you
debate whether to spend the money to swim with dolphins and
ultimately decide not to. Although nothing limits your ability to go
back there the very next day, it is still easy to imagine sitting at
home that evening regretting not having purchased such an excep-
tional experience.
Future Directions
Several avenues of future research merit exploration. First, the
pattern of results we report may be moderated by materialism.
Materialists may be more inclined than the general population to
see material goods as singular and thus experience greater regret
than less materialist people over missed opportunities to buy them.
Conversely, people who are dispositionally experience seeking
should generally be less likely to experience regrets of action and
more likely to experience regrets of inaction. Second, it would also
be worthwhile to explore how the opposing temporal patterns
outlined by Gilovich and Medvec (1994, 1995) on the one hand
and Abendroth and Diehl (2006) on the other apply to material and
experiential purchases. It may be that a fair number of the short-
term regrets of action that Gilovich and Medvec reported involve
material purchases, and the bulk of their corpus of long-term
regrets of inaction involve missed experiential opportunities. An-
other potentially fruitful area of future research would be to
examine the impact of the differential amount of social interaction
that tends to accompany material and experiential purchases. Our
findings suggest that sociality might influence people’s likely
regrets because social experiences— by virtue of the unique com-
bination of personalities attendant at each one— often seem more
singular. Going to the movies by oneself on Thursday is not all that
different from doing it on Friday, but going with one group of
friends rather than another or even the same group of friends who
are in a different mood is not nearly as interchangeable. Finally,
the results of Study 5 highlight important opportunities for the
study of behavioral interventions in this domain. Might an inten-
tional focus on the experiential elements of even clearly material
purchases lead to greater satisfaction and diminished regret?
In the end, we hope our research helps inform people’s future
purchase decisions. As this article was being written, one of the
authors was the subject of our opening example. She was debating
whether to take a trip to Mexico with her husband and, on seeing
the price tag for tickets, had a hard time deciding to go. Around the
same time, she considered replacing the range in her kitchen, a
rusty unit from the 1980s with a tendency to spew gas for 20
minutes while deciding whether to light. After her husband re-
minded her of this very line of research, she sheepishly bought the
tickets, had a wonderful week on the beach, and does not regret a
penny she spent getting there. She still has not bought the stove of
her dreams but has no regrets over that inaction—the kitchen has
yet to explode. Until it does, there are still plenty of ranges from
which to choose.
References
Abendroth, L. J., & Diehl, K. (2006). Now or never: Effects of limited
purchase opportunities on patterns of regret over time. Journal of Con-
sumer Research, 33, 342–351. doi:10.1086/508438
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
distinction in social psychological research: Conceptual, strategic, and
statistical considerations. Journal of Personality and Social Psychology,
51, 1173–1182.
Beike, D. R., Markman, K. D., & Karadogan, F. (2009). What we regret
most are lost opportunities: A theory of regret intensity. Personality and
222 ROSENZWEIG AND GILOVICH
Social Psychology Bulletin, 35, 385–397. doi:10.1177/
0146167208328329
Carter, T. J., & Gilovich, T. (2010). The relative relativity of material and
experiential purchases. Journal of Personality and Social Psychology,
98, 146 –159. doi:10.1037/a0017145
Carter, T. J., & Gilovich, T. (2011). I am what I do, not what I have: The
differential centrality of experiential and material purchases to the self.
Manuscript submitted for publication.
Gilovich, T., & Medvec, V. H. (1994). The temporal pattern to the
experience of regret. Journal of Personality and Social Psychology, 67,
357–365. doi:10.1037/0022-3514.67.3.357
Gilovich, T., & Medvec, V. H. (1995). The experience of regret: What,
when, and why. Psychological Review, 102, 379 –395. doi:10.1037/
0033-295X.102.2.379
Gilovich, T., Wang, R. F., Regan, D., & Nishina, S. (2003). Regrets of
action and inaction across cultures. Journal of Cross-Cultural Psychol-
ogy, 34, 61–71. doi:10.1177/0022022102239155
Hattiangadi, N., Medvec, V. H., & Gilovich, T. (1995). Failing to act:
Regrets of Terman’s geniuses. International Journal of Aging and Hu-
man Development, 40, 175–185. doi:10.2190/4U4E-N77B-PKJ2-CJXM
Hetts, J. J., Boninger, D. S., Armor, D. A., Gleicher, F., & Nathanson, A.
(2000). The influence of anticipated counterfactual regret on behavior.
Psychology and Marketing, 17, 345–368.
Kahneman, D. (1995). Varieties of counterfactual thinking. In N. J. Roese
& J. M. Olson (Eds.), What might have been: The social psychology of
counterfactual thinking (pp. 375–396). Hillsdale, NJ: Erlbaum.
Kahneman, D., & Miller, D. T. (1986). Norm theory: Comparing reality to
its alternatives. Psychological Review, 93, 136 –153. doi:10.1037/0033-
295X.93.2.136
Kahneman, D., & Tversky, A. (1982). The simulation heuristic. In D.
Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under uncertainty:
Heuristics and biases (pp. 201–208).New York, NY: Cambridge Uni-
versity Press.
Landman, J. (1993). Regret: The persistence of the possible. New York,
NY: Oxford University Press.
Loomes, G., & Sugden, R. (1983). Regret theory: An alternative theory of
rational choice under uncertainty. The Economic Journal, 92, 805– 824.
doi:10.2307/2232669
MacKinnon, D. P., & Dwyer, J. H. (1993). Estimating mediated effects in
prevention studies. Evaluation Review, 17, 144 –158. doi:10.1177/
0193841X9301700202
Medvec, V. H., Madey, S., & Gilovich, T. (1995). When less is more:
Counterfactual thinking and satisfaction among Olympic medal winners.
Journal of Personality and Social Psychology, 69, 603– 610. doi:
10.1037/0022-3514.69.4.603
Miller, D. T., & Gunasegaram, S. (1990). Temporal order and the per-
ceived mutability of events: Implications for blame assignment. Journal
of Personality and Social Psychology, 59, 1111–1118. doi:10.1037/
0022-3514.59.6.1111
Miller, D. T., & Taylor, B. R. (1995). Counterfactual thought, regret, and
superstition: How to avoid kicking yourself. In N. J. Roese & J. M.
Olson (Eds.), What might have been: The social psychology of counter-
factual thinking (pp. 305–332). Mahwah, NJ: Erlbaum.
Morrison, M., & Roese, N. J. (2011). Regrets of the typical American:
Findings from a nationally representative survey. Social Psychological
and Personality Science. Advance online publication. doi:10.1177/
1948550611401756
Nicolao, L., Irwin, J. R., & Goodman, J. K. (2009). Happiness for sale: Do
experiential purchases make consumers happier than material pur-
chases? Journal of Consumer Research, 36, 188 –198. doi:10.1086/
597049
Roese, N. J., & Olson, J. M. (1995). Counterfactual thinking: A critical
overview. In N. J. Roese & J. M. Olson (Eds.), What might have been:
The social psychology of counterfactual thinking (pp. 1–55). Mahwah,
NJ: Erlbaum.
Roese, N. J., & Summerville, A. (2005). What we regret most . . . and why.
Personality and Social Psychology Bulletin, 31, 1273–1285. doi:10
.1177/0146167205274693
Van Boven, L., Campbell, M. C., & Gilovich, T. (2010). The social costs
of materialism: On people’s assessments of materialistic and experiential
consumers. Personality and Social Psychology Bulletin, 36, 551–563.
doi:10.1177/0146167210362790
Van Boven, L., & Gilovich, T. (2003). To do or to have? That is the
question. Journal of Personality and Social Psychology, 85, 1193–1202.
doi:10.1037/0022-3514.85.6.1193
Zeelenberg, M., van den Bos, K., van Dijk, E., & Pieters, R. (2002). The
inaction effect in the psychology of regret. Journal of Personality and
Social Psychology, 82, 314 –327. doi:10.1037/0022-3514.82.3.314
Received March 17, 2011
Revision received June 15, 2011
Accepted June 15, 2011 䡲
223
MATERIAL AND EXPERIENTIAL PURCHASE REGRETS
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Personality and Social Psychology
This content is subject to copyright. Terms and conditions apply.