When Consumers Prefer to Give Material Gifts Instead of Experiences: The Role of Social Distance

When Consumers Prefer to Give Material Gifts Instead of Experiences: The Role of Social Distance Abstract Although previous research suggests that there are hedonic and interpersonal benefits to gifting experiences, consumers often give material gifts rather than experiential gifts. Exploring this mismatch, the current research examines when and why consumers prefer to give material versus experiential gifts. The authors propose that gift givers are more likely to give experiential gifts to socially close recipients than socially distant recipients. Since experiences are perceived as more unique than material goods, givers perceive that choosing an experiential gift requires more specific knowledge of a recipient’s preferences to avoid the greater social risk of giving a poorly matched gift. Eight studies provide converging evidence for the proposed effect of social distance on gift preference and demonstrate that this effect is driven by a giver’s knowledge of a recipient’s preferences. Further supporting the mechanism of preference knowledge, the effect of social distance is moderated by the social risk associated with experiential gifts. When experiences contain little social risk—and thus require less knowledge of a recipient—the effect of social distance is significantly mitigated. Together, these results provide answers for why consumers often prefer to give material gifts over experiences, despite the advantage of giving experiences. gift giving, experiential advantage, material and experiential consumption, social distance Gift giving is ubiquitous. It is practiced by consumers of every culture, and researchers from various disciplines have been studying it for decades (Belk and Foa 1980; Coon 1993; Komter 1996; Mauss 1925; Robben and Verhallen 1994). Gift giving serves multiple functions: gifts signal relationships and information, strengthen social bonds, and fulfill social norms, just to name a few (Belk 1996; Camerer 1988; Huang and Yu 2000; Schwartz 1967; Teigen, Olsen, and Solas 2005). Gifts also demonstrate unselfish, agapic love (Belk and Coon 1993), and people often give gifts with the intention of increasing the happiness and enjoyment of their recipients (Baskin et al 2014; Gino and Flynn 2011; Otnes, Lowry, and Kim 1993; Teigen et al. 2005). Considering these social and hedonic functions of gift giving, it seems that consumers should prefer to give experiences (intangible events that a person lives through, e.g., travel, concert) over material goods (tangible products that are retained over time, e.g., clothing, furniture; Tully, Hershfield, and Meyvis 2015). Previous findings have shown that experiential purchases contribute more to consumer happiness than material purchases (Nicolao, Irwin, and Goodman 2009; Van Boven and Gilovich 2003), suggesting that recipients would derive greater happiness from experiential gifts rather than material gifts (Goodman and Lim 2017; Wallman 2015). More recent research highlights another benefit of giving experiences, demonstrating that experiential gifts strengthen social connections between givers and recipients (Chan and Mogilner 2017). Furthermore, consumers prefer to give unique and individualized gifts (Belk 1996; Steffel and LeBoeuf 2014), which is a quality often associated with experiences rather than material goods (Carter and Gilovich 2012; Dai, Chan, and Mogilner 2017). Despite the hedonic and interpersonal benefits of gifting experiences, consumers seem to prefer to give material gifts rather than experiences. For instance, the vast majority of gifts recommended by gift-giving gurus are material in nature (e.g., see New York Magazine’s 2017 Holiday Gift Guide), and in a survey of 100 Americans we found a vast majority of the gifts listed—81% of holiday gifts and 66% of birthday gifts—were material rather than experiential.1 Why do consumers overwhelmingly prefer to give material gifts? Moreover, are there times when consumers do in fact prefer to give experiences? This research examines when and why consumers often give material gifts over experiential gifts. We propose that the social distance between gift givers and recipients—a key component in gift exchange—influences consumers’ preferences for giving experiences versus material goods. In line with previous literature on interpersonal relationships, we conceptualize social distance in terms of interpersonal closeness, which is characterized by frequent, diverse, and strong interactions between two people (Berscheid, Snyder, and Omoto 1989; Kelley et al. 1983). Because experiences, on average, are perceived as more unique and individualized than material goods (Carter and Gilovich 2012; Dai et al. 2017), gift givers might perceive an experiential gift as requiring more specific knowledge of a recipient’s preferences. Thus, they might believe that giving an experience carries greater social risk of choosing an experiential gift that poorly matches their recipient’s preferences.2 Gift givers are more likely to have this preference knowledge when they are socially close to their recipients compared to when they are socially distant. Accordingly, we propose that the closer givers and recipients become, the more likely gift givers are to give an experience compared to a material gift. Across eight studies, we tested these hypotheses and consistently found that gift givers were more likely to give experiences (vs. material goods) when their recipients were socially close compared to socially distant (studies 1a and 1b). We also demonstrated that this effect was due to givers possessing less knowledge about the individual preferences of distant recipients compared to close recipients (studies 2a, 2b, 2c, and 3). Further, when givers were provided with the opportunity to give experiences low on social risk (i.e., experiential gifts that were less likely to have a preference mismatch and incur a social cost), the effect of social distance on gift preference was significantly attenuated (studies 4a and 4b). Thus, despite the advantage of giving experiences, consumers are less likely to choose experiential gifts for socially distant recipients (vs. socially close recipients) in order to avoid the social risk of giving experiential gifts that poorly match their recipients’ preferences. SOCIAL DISTANCE AND GIFT GIVING Gift giving is a social exchange between two parties. In gift giving, consumers make purchases for others, which raises a challenge for gift givers to correctly infer which gifts might best match the preferences of their recipients (Baskin et al. 2014; Otnes et al. 1993). Beyond finding gifts that simply match their recipients’ preferences, gift givers are also motivated to give gifts that are unique or have some individualized elements (Belk 1996; Belk and Coon 1993). For instance, consumers prefer to give gifts that capture unique characteristics of their recipients and shy away from an item in gift registries, particularly when selecting a gift for a close friend (Ward and Broniarczyk 2016). At the extremes, givers have been shown to even prefer gifts that are overly individualized at the expense of greater utility (Steffel and LeBoeuf 2014). A variety of empirical evidence suggests that experiential gifts are likely to be more effective than material gifts, on average, at achieving this goal of giving gifts that uniquely capture recipients’ preferences. First, compared to material goods, consumers perceive experiences to be less comparable (Carter and Gilovich 2010) and less interchangeable (Rosenzweig and Gilovich 2012). For example, it is not that one concert is necessarily of higher quality than another concert, or that one vacation is always better than another. This lower comparability of experiences suggests that the utility of each experience might vary depending on consumers’ individual preferences. Second, previous research has found that consumers expect their evaluations of experiences to be different from others’ evaluations, which prompts them to rely less on consumer reviews for experiential purchases compared to material purchases (Dai et al. 2017). Third, people believe that compared to material goods, experiential purchases are more informative of a purchaser (Carter and Gilovich 2012) and are more intrinsically motivated (Van Boven, Campbell, and Gilovich 2010). In short, these findings suggest that experiential gifts are perceived to be more individualized gifts that uniquely reflect the recipients’ preferences and tastes. Given this uniqueness of experiences, gift givers might perceive selecting an appropriate experiential gift as requiring more specific knowledge of recipients’ preferences. For example, knowing that your colleague enjoys music and concerts, you may think a ticket for a Beyoncé concert would be a good gift; however, you might not be socially close enough to her to know that she is actually a fan of heavy metal bands and would be embarrassed to even consider such a concert. While gift givers are more motivated to give experiences since they are unique to recipients’ individual preferences, this uniqueness of experiential gifts may also increase the social risk of choosing an experiential gift that poorly matches the recipients’ preferences. Accordingly, gift givers would perceive material goods as a safer and less risky alternative when they do not have sufficient knowledge of recipients. Thus, we propose that without the requisite preference knowledge, gift givers would be less likely to give an experiential gift. Whether gift givers have this requisite knowledge of their recipients’ preferences will likely depend on the social distance between givers and receivers, which has been shown to play an important role in gift giving (Chan and Mogilner 2017; Ward and Broniarczyk 2011, 2016). Relationships can be conceptualized as being more or less distant on a continuum of social closeness (Aron, Aron, and Smollan 1992; Berscheid et al. 1989). Close relationships are developed through building extended and intimate interactions (Berscheid et al. 1989; Kelley et al. 1983) and sharing personal experiences (Parks and Floyd 1996). These interactions naturally allow individuals to develop greater knowledge of each other. Indeed, people predict thoughts and feelings of friends more accurately than those of strangers because they have a deeper understanding of close others (Stinson and Ickes 1992). Accordingly, we propose that gift givers will feel that they have better knowledge of close recipients compared to distant recipients. Further, this knowledge will in turn affect their relative preferences for material versus experiential gifts. As the social distance between givers and receivers decreases, givers will perceive more knowledge of their recipients’ preferences, and they will be more willing to give an experience over a material gift. Alternatively, as the social distance between givers and recipients increases, givers’ knowledge of their recipients will decrease and their relative preferences will shift toward a material gift. More formally, we hypothesize that the more socially distant givers are from their recipients, the more likely they are to give a material gift compared to an experiential gift. Further, we hypothesize that this effect of social distance will be mediated by gift givers’ knowledge of their recipients’ preferences. The role of recipients’ preference knowledge suggests an important moderator: the social risk associated with a gift. We define social risk as the extent to which a gift may not match a recipient’s individual preferences, potentially resulting in harm to one’s social relationship or social standing (Hoyer, MacInnis, and Pieters 2012). The mere notion that a gift may not match recipients’ preferences could elicit anxiety in the gift decision process, and an actual mismatch can risk damaging a social relationship (Wooten 2000). We propose that selecting an experiential gift is perceived as requiring more specific knowledge of recipients’ preferences, which makes gift givers feel more susceptible to the social risk of giving a gift that may not be well liked. Hence, to avoid this social risk, gift givers will, in general, shy away from experiential gifts compared to material gifts when they do not possess the necessary knowledge of their recipients. However, there may be some exceptions whereby experiential gifts could be perceived as less risky to give because they are less likely to mismatch recipients’ preferences and harm a social relationship. For instance, compared to a specific experiential gift (e.g., aroma deep facial massage), a more general experiential gift (e.g., a spa gift certificate) can be used in multiple ways and thus would require less knowledge of recipients’ preferences. Accordingly, gift givers might perceive a more general experiential gift as carrying lower social risk. If givers are provided with an opportunity to give such low-risk experiences, the effect of social distance on gift preferences should be mitigated. Eight studies test our proposed theory (summarized in figure 1). In study 1a, participants indicated greater preferences for experiential gifts relative to material gifts when recipients were socially close compared to socially distant. Similarly, study 1b demonstrated that participants were also more likely to list an experiential gift than a material gift when they were asked to choose a gift for a close friend compared to a distant friend. In studies 2a–c, we tested our preference knowledge theory by manipulating knowledge of recipients’ preferences. Across these three studies, we consistently found that participants exhibited greater preferences for experiential gifts to material gifts when their knowledge of recipients was high compared to low. Further supporting our theory, study 3 revealed that preference knowledge mediated the effect of social distance on gift preference. Study 3 also ruled out joint/shared consumption as an alternative explanation by demonstrating that the effect of social distance on gift preference remained constant regardless of shared consumption of experiential gifts. Finally, studies 4a and 4b tested whether providing gift givers with an opportunity to give experiential gifts low in social risk—less specific gifts that do not require a high degree of knowledge about recipients’ individual preferences—moderates the effect of social distance. We found that the effect of social distance on gift preference was mitigated when givers were provided the opportunity to give less socially risky experiences. FIGURE 1 View largeDownload slide THEORETICAL MODEL AND HYPOTHESES FIGURE 1 View largeDownload slide THEORETICAL MODEL AND HYPOTHESES STUDIES 1A AND 1B: THE ROLE OF SOCIAL DISTANCE IN GIFT PREFERENCE Studies 1a and 1b aimed to provide an initial test of whether social distance influences gift givers’ preferences for experiential and material gifts. We hypothesized that givers’ preferences would shift toward experiential gifts when they choose a gift for socially close recipients compared to socially distant ones. In study 1a, participants rated how much they preferred to give various experiential and material gifts to either a close or distant friend as a holiday gift. Study 1b tested the effect of social distance with a more naturalistic paradigm using an open-ended format. Additionally, study 1b assessed the generalizability of our findings by testing the effect of social distance in another gift-giving occasion (i.e., birthday). Study 1a Method and Procedure Seventy-four undergraduates (Mage = 19.32, 34% female), from various colleges at a midsized private university, participated in this study in exchange for credit in their introductory business course. Sample size was set as large as possible with the constraint of the number of students available to participate in the study. We used the same sample size criteria for all studies using undergraduates. The study was a 2 (social distance: distant vs. close; between subjects) × 2 (gift type: material vs. experiential; within-subject) mixed design. We manipulated social distance by asking participants to think of either a close or distant friend (depending on condition) and write down the initials of that friend. Participants were asked to imagine that they were choosing a holiday gift for the friend and shown a list of 10 gifts of similar cost, five experiential (concert ticket, sightseeing boat ride [available in over 50 cities], ticket for comedy night, wine tour and tasting, cooking classes [four sessions]) and five material (handmade pen, sunglasses, headphones, leather gloves, wallet/purse). A separate pretest (N = 99 via Amazon Mechanical Turk (MTurk), US only, 95% approval or higher) confirmed that the experiential gifts were actually perceived as more experiential than the material gifts (p < .001; see appendix A for details). All gift items were presented in a randomized order, and they were not labeled as experiential or material. Participants were told to assume that costs of the gifts were similar, and they indicated their preferences by rating how much they would like to give each gift to their friend on an 11-point Likert scale (0 = definitely would not give, 10 = definitely would give). Results We averaged participants’ preference ratings of the experiential and material gifts, respectively, and created a preference score for each type of gift. A repeated-measures analysis of variance (ANOVA) was conducted on this preference score with social distance as a between-subjects factor and gift type as a within-subject factor. The analysis yielded a main effect of social distance such that participants indicated greater preferences for all gifts when they were giving a gift to a close friend rather than a distant friend (Mclose = 3.68 vs. Mdistant = 2.47; F(1, 72) = 9.22, p = .003, ηp2= .11). More importantly, this main effect was qualified by a social distance × gift type interaction (F(1, 72) = 15.02, p < .001, ηp2= .17, see figure 2). As predicted, the relative preference for experiences (vs. material goods) was greater for a close friend (Mexperiential = 4.20 vs. Mmaterial = 3.16; F(1, 72) = 8.88, p = .004, ηp2= .11) than for a distant friend (Mexperiential = 1.93 vs. Mmaterial = 3.01; F(1, 72) = 6.58, p = .012, ηp2= .08). FIGURE 2 View largeDownload slide STUDY 1A: THE EFFECT OF SOCIAL DISTANCE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. FIGURE 2 View largeDownload slide STUDY 1A: THE EFFECT OF SOCIAL DISTANCE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. To check the robustness of our effect, we conducted a conceptual replication with a larger sample (300 MTurk workers, 294 responses, Mage = 34.44, 45% female, US only, 95% approval or higher) following the same procedure as study 1a. Consistent with the results of study 1a, we again found a significant social distance by gift type interaction (F(1, 292) = 31.38, p < .001, ηp2= .10). Experiences were more preferred than material gifts for a close friend (Mexperiential = 4.58 vs. Mmaterial = 3.66; F(1, 292) = 27.72, p < .001, ηp2= .09) compared to a distant friend (Mexperiential = 2.43 vs. Mmaterial = 2.89; F(1, 292) = 7.01, p = .009, ηp2= .02). Study 1b Method and Procedure Two hundred seventy-six undergraduate students (Mage = 19.24, 43% female), from various colleges at a midsized private university, participated in this study in exchange for credit in their introductory business course. Depending on condition, participants were asked to think of either a close or distant friend whose birthday was coming up and type the initials of that friend. Participants were asked to indicate a gift they would like to give to the friend as a birthday gift (open-ended; see web appendix B for product categories of the gifts listed by participants). Next, they rated “to what extent the gift was more material or more experiential” on a nine-point Likert scale (1 = more material, 5 = both, 9 = more experiential). Finally, participants indicated the approximate cost of their gift and answered some demographic questions. Results Replicating the findings of study 1a, we found a significant effect of social distance on preference for gift type (t(274) = 4.13, p < .001, Cohen’s d= .50). Participants were more (less) likely to list an experiential gift (a material gift) for a close friend (M = 4.36) than for a distant one (M = 3.08). Moreover, the effect of social distance remained significant (p < .01) even when we controlled for price of the gift (log-transformed), age, and gender. Discussion In sum, the results of studies 1a and 1b provide converging evidence for our hypothesis that a giver’s relative preference for material and experiential gifts changes as a function of his or her social closeness to the recipient. Notably, this pattern was consistently observed across the two different gift-giving occasions (i.e., holidays and birthdays) and regardless of whether the gift items were generated by the experimenter (study 1a) or participants (study 1b). STUDIES 2A, 2B, AND 2C: PREFERENCE KNOWLEDGE Studies 2a–c were designed to test the proposed underlying mechanism of our effect, namely givers’ knowledge of recipients’ preferences. We posited that more knowledge of close recipients (vs. distant recipients) would translate into greater preferences for experiential gifts over material gifts. To test this prediction, we directly manipulated gift givers’ knowledge of their recipients’ preferences. In studies 2a and 2b, participants rated their preferences to give various material and experiential gifts to a friend either whom they knew well or did not know well (between subjects). Study 2c was designed to provide a more conservative test of our effect by using similar material and experiential goods. Namely, participants chose between material and experiential gifts in the same product category (i.e., home spa gift basket vs. aroma deep facial massage). Further, in these three studies, we aimed to rule out several alternative explanations. It is possible that experiential gifts are viewed as more expensive, more hedonic, and/or less utilitarian, or perhaps they are viewed differently based on social distance due to a different mind-set. Therefore, to rule out these alternatives, we tested whether our effect remains significant after controlling for perceived cost and/or the hedonic and utilitarian nature of the gifts, or whether these alternative factors could explain our findings. Study 2a Methods and Procedure One hundred twenty undergraduate students (Mage = 21.48, 52% female) at a large public university participated in this study in exchange for credit in their introductory business course. The study was a 2 (preference knowledge: high knowledge vs. low knowledge) × 2 (product type: material vs. experiential) between-subjects design. To manipulate preference knowledge, we asked participants to write the initials of a friend whom they knew very well or whom they did not know very well, depending on condition. Next, we asked participants to imagine that they were buying a birthday gift for their friend. Depending on condition, participants were presented with either five material gifts or five experiential gifts (identical gifts used in study 1a). For each gift, participants rated the extent to which they would like to give it as a birthday gift to their friend (0 = definitely would not give, 10 = definitely would give). Although we specified that the price of the gifts was similar, participants might perceive the general price of certain gifts as more expensive than others, which might explain our effect. To test this possibility, we asked participants to estimate the cost of each gift (open-ended). In addition, participants rated the extent to which each of the gifts was hedonic and utilitarian (0 = definitely not hedonic/utilitarian at all, 10 = definitely hedonic/utilitarian) after reading the definition of a hedonic and a utilitarian purchase (hedonic was defined as something for pleasure, fantasy, and/or fun; utilitarian was defined as something to fulfill a basic need or to accomplish a functional or practical task, Strahilevitz and Myers 1998). Finally, as a manipulation check, we asked participants to indicate 1) how well they know their friend’s preferences and 2) how well they know what their friend likes and dislikes. Manipulation Check Analyzing participants’ averaged responses to two items on perceived knowledge (r = .95), a 2 (preference knowledge: high vs. low) × 2 (gift type: material vs. experiential) ANOVA revealed only a main effect of preference knowledge: the high-knowledge condition perceived greater knowledge of their recipient’s preferences than those in the low-knowledge condition (Mhigh = 7.95 vs. Mlow = 3.53; F(1, 116) = 248.83, p < .001, ηp2= .68). The interaction and the other main effect were not significant (Fs < 1). Gift Preference For our dependent measure, we created a gift preference index by averaging participants’ preference ratings of the five material or experiential gifts. A 2 (social distance: close vs. distant) × 2 (gift type: material vs. experiential) ANOVA found two main effects: participants in the high-knowledge condition indicated greater preferences for all gifts than those in the low-knowledge condition (Mhigh = 4.27 vs. Mlow = 3.38; (F(1, 116) = 8.28, p = .005, ηp2= .07), and preferences were greater for experiential gifts than for material gifts (Mexperiential = 4.15 vs. Mmaterial = 3.51; F(1, 116) = 4.31, p = .04, ηp2= .04). More importantly, and central to our hypothesis, these main effects were qualified by a marginally significant social distance × gift type interaction (F(1, 116) = 2.92, p = .090, ηp2= .02; see figure 3): relative preferences for experiential gifts (vs. material gifts) were greater in the high-knowledge condition (Mexperiential = 4.85 vs. Mmaterial = 3.69; F(1, 116) = 7.16, p = .009, ηp2= .06) than in the low-knowledge condition (Mexperiential = 3.44 vs. Mmaterial = 3.33; F < 1). FIGURE 3 View largeDownload slide STUDY 2A: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error. FIGURE 3 View largeDownload slide STUDY 2A: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error. FIGURE 4 View largeDownload slide STUDY 2B: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. FIGURE 4 View largeDownload slide STUDY 2B: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. Testing Alternative Accounts We conducted a set of analyses to examine three alternative accounts—hedonic, utilitarian, and cost. First, participants did perceive material and experiential purchases differently in terms of cost and their hedonic and utilitarian nature. Compared to material gifts, experiential gifts were perceived as more hedonic (Mexperiential = 7.18 vs. Mmaterial = 4.44; F(1, 116) = 64.98, p < .001, ηp2 = .36),3 less utilitarian (Mexperiential = 2.76 vs. Mmaterial = 5.55; F(1, 116) = 79.83, p < .001, ηp2 = .41), and more expensive (log-transformed price; Mexperiential = 4.38 vs. Mmaterial = 3.89; F(1, 115) = 17.36, p < .001, ηp2 = .13).4 Second, if our key effect is driven by any of these alternative accounts, then we would expect one of these accounts to interact with preference knowledge to predict gift preference. To test these alternatives, we ran three separate models5—one for hedonic, one for utilitarian, and one for cost—that consisted of our basic model (gift type, preference knowledge, and gift type × preference knowledge interaction) plus each alternative account variable and its interaction with preference knowledge (details of each model are in table 1). The hedonic model did not reveal a significant hedonic nature × preference knowledge interaction (F < 1), and our key interaction remained marginally significant (preference knowledge × gift type, F(1, 114) = 3.12, p = .080, ηp2 = .03). The utilitarian model also did not yield a significant utilitarian nature × preference knowledge interaction (F < 1), but our key interaction also became nonsignificant (F(1, 114) = .75, p = .39, ηp2 = .01). Lastly, the cost model also did not find a significant preference knowledge × cost interaction (F < 1), and our key interaction remained marginally significant (F(1, 113) = 2.88, p = .093, ηp2 = .02). In sum, none of these alternative models predicted gift preference; thus, they did not provide evidence of mediation by any of the alternative explanations. TABLE 1 STUDY 2A: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 a Key hypothesized effect. b Potential alternative explanations. TABLE 1 STUDY 2A: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 a Key hypothesized effect. b Potential alternative explanations. Study 2b Methods and Procedure We recruited 300 MTurk workers and 302 participated (Mage = 35.33, 54% female, US only, 95% approval or higher, previous workers blocked from participation) in a 2 (preference knowledge: high vs. low; between subjects) × 2 (gift type: material vs. experiential; within-subject) mixed design. The procedure was similar to that of study 2a, except that the material-experiential factor was within-subject—in other words, participants rated all 10 gifts (five material goods and five experiences). Participants wrote the initials of a friend either whom they knew very well (high-knowledge condition) or whom they did not know very well (low-knowledge condition). Next, they wrote a short paragraph about how well they knew/did not know their friend. As in study 2a, participants rated the extent to which they would like to give each of the gifts to the friend as a birthday gift (0 = definitely would not give, 10 = definitely would give). Participants estimated the price of each gift and indicated whether their friend was “real, fake, or a little bit of both.” Results Two participants reported that their friend was not a real person but a fake. The results reported below exclude these participants from the subsequence analysis; however, including these participants did not substantially change the findings or significance.6 A repeated-measures ANOVA with preference knowledge (high vs. low) as a between-subjects factor and product type (material vs. experiential) as a within-subject factor yielded a significant effect of preference knowledge (F(1, 298) = 28.23, p < .001, ηp2= .09) and that of gift type (F(1, 298) = 17.02, p < .001, ηp2= .05). Most importantly, these main effects were qualified by a preference knowledge × gift type interaction (F(1, 298) = 4.57, p = .03, ηp2= .02; see figure 4): participants preferred experiential gifts more in the high-knowledge condition (Mexperiential = 4.67 vs. Mmaterial = 3.85; F(1, 298) = 19.62, p < .001, ηp2 = .06) than in the low-knowledge condition (Mexperiential = 3.35 vs. Mmaterial = 3.10; F(1, 298) = 1.98, p > .16, ηp2 = .01). FIGURE 5 View largeDownload slide STUDY 3: GIFT PREFERENCE ACROSS CONDITIONS NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. FIGURE 5 View largeDownload slide STUDY 3: GIFT PREFERENCE ACROSS CONDITIONS NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. Testing an Alternative Account One might argue that gift givers prefer to give more expensive gifts to recipients they have more preference knowledge of and these givers perceive experiential gifts as more expensive. Although we specified that the price of the gifts was similar, participants did perceive the general cost of experiential gifts as higher than material gifts (Mexperiential = $92.52 vs. Mmaterial = $51.23, F(1, 298) = 31.72, p < .001, ηp2 = .10). To test whether cost explains our effect, we calculated a difference score in the estimated price of material and experiential gifts and added this cost difference variable, along with its interaction with knowledge preference, to our basic model (details are in table 2). The results did not show either a significant main effect of cost difference or a cost difference × preference knowledge interaction (Fs < 1). Further, our key interaction between preference knowledge and gift type was still marginally significant (F(1, 296) = 3.50, p = .062, ηp2= .01). TABLE 2 STUDY 2B: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 a Key hypothesized effect. b Potential alternative explanations. TABLE 2 STUDY 2B: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 a Key hypothesized effect. b Potential alternative explanations. Study 2c Methods and Procedure We recruited 400 MTurk workers and 403 participated (Mage = 35.03, 51% female, US only, 95% approval or higher, previous workers blocked from participation). Similar to studies 2a and 2b, participants first wrote the initials of a female friend whom they knew very well (high-knowledge condition) or whom they did not know very well (low-knowledge condition) and then described how well they knew or did not know her. Since the product stimuli (i.e., spa-related product) used in this study were female-oriented purchases, we controlled for potential gender effects by asking participants to imagine that they were buying a birthday gift for a female friend. Participants were told that they narrowed a birthday gift to two items—a home spa gift basket (material gift) and an aroma deep facial massage (experiential gift)—and were asked to make a choice between the two. Although we attempted to minimize any possible differences other than the material-experiential dimension by using similar gifts, participants could still perceive these items differently. To rule out this possibility, we again measured participants’ perception of the hedonic and utilitarian nature of each gift by using similar measures as in study 2a (1 = definitely not hedonic/utilitarian at all, 7 = definitely hedonic/utilitarian). Lastly, participants reported the gender of their friend as well as whether their friend was “real, fake, or a little bit of both.” Results Six participants reported that they listed a fake friend rather than a real person, and 13 participants thought of a male friend instead of a female friend. The results reported below exclude them from the subsequence analysis; however, including these participants did not substantially change the findings or significance.7 As predicted, choice of experiential gift (vs. material gift) was significantly greater in the high-knowledge condition (39.27%) compared to the low-knowledge condition (28.50%; b = .24, Wald χ2(1) = 4.94, p = .03). Testing an Alternative Account To examine whether this effect was driven by differences in the hedonic or utilitarian nature of the gifts, we conducted a set of additional analyses. First, compared to material gifts, experiential gifts were indeed rated as more hedonic (Mexperiential = 5.72 vs. Mmaterial = 4.79; F(1, 382) = 134.89, p < .001, ηp2= .26) and as less utilitarian (Mexperiential = 2.64 vs. Mmaterial = 3.81; F(1, 382) = 166.40, p < .001, ηp2= .30).8 Second, if our key effect is driven by any of these alternative accounts, then we would expect one of these accounts to interact with preference knowledge to predict gift choice. To test these alternatives, we created two difference scores (one hedonic and one utilitarian), subtracting the material gift rating from the experiential gift rating in terms of both the hedonic and utilitarian nature of the gifts. Then we conducted two logistic regressions with gift choice (0 = material gift, 1 = experiential gift) as the dependent variable and preference knowledge (–1 = low, 1 = high), hedonic or utilitarian difference score (mean-centered), and their interaction as predictors (details of each model are in table 3). In the hedonic model, we found that the hedonic difference score was marginally significant (b = .13, Wald χ2(1) = 3.22, p = .073), but the hedonic difference score × preference knowledge interaction was not (p = .83), and our key effect of preference knowledge on gift choice remained significant (b = .26, Wald χ2(1) = 5.61, p = .02). In the utilitarian model, the utilitarian difference score was significant (b = .21, Wald χ2(1) = 10.72, p = .001), but the utilitarian difference score × preference knowledge interaction was not (p > .60). Our key effect of preference knowledge decreased slightly to become marginally significant (b = .20, Wald χ2(1) = 3.34, p = .068). In sum, consistent with studies 2a and 2b, none of these alternative models led to a significant mediator that predicted our hypothesized gift preference × preference knowledge interaction, suggesting that they are not viable alternative explanations. TABLE 3 STUDY 2C: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 a Key hypothesized effect. b Potential alternative explanations. TABLE 3 STUDY 2C: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 a Key hypothesized effect. b Potential alternative explanations. Discussion By manipulating the knowledge that a giver has of his or her recipient’s preferences, studies 2a–c showed that consumers were more likely to give experiential gifts than material gifts when their knowledge of the recipient’s preference was high compared to low. We consistently observed this shift in gift preference across different samples (i.e., undergraduate and MTurk workers) and different methodologies (i.e., preference rating and choice, between-subjects and within-subject). Studies 2a–c also provide evidence suggesting that our effect cannot be fully explained by cost and the differential utilitarian or hedonic dimensions of these gifts. In sum, the results provide support for our theoretical process that consumers are likely to prefer material gifts relatively more than experiential gifts when they do not have sufficient knowledge of recipients’ preferences. STUDY 3: SHARED CONSUMPTION Our goal for study 3 was twofold. First, we sought to capture more direct evidence on our proposed mechanism of preference knowledge. Thus, we measured gift givers’ perceived knowledge of their recipients’ preferences and tested whether this knowledge mediates the effect of social distance on gift type. Second, we aimed to rule out the alternative explanation that the effect of social distance is driven by the shared nature of experiential gifts. Experiences are often consumed with others (Caprariello and Reis 2013), and participants might have preferred to give gifts that they could experience with their recipient (shared consumption) only when they were socially close to the recipient. To examine this alternative account, we manipulated whether the experiential gifts were jointly consumed by givers and recipients or consumed alone. If our findings are due to the shared consumption of experiences, then gift givers should prefer to give experiential gifts for a close recipient only when they can participate in the experiences with the recipient. Alternatively, if social distance affects gift preference independent of shared consumption, then the effect of social distance should not interact with shared consumption. Further, to distinguish sharing between the giver and recipient from sharing in general, we added a third condition where experiential gifts were described as to be shared with someone else (but not with the giver). Method and Procedure We recruited 300 participants via MTurk and 291 responded (Mage = 32.47, 47% female, US only, 95% approval or higher, previous workers blocked from participation). Sample size was set based on a power analysis indicating that we would need 139 participants to replicate the effect with 80% power, but given that we were testing a potential moderator as an alternative explanation, we more than doubled our sample size to be conservative. The study was a 2 (social distance: close vs. distant; between subjects) × 3 (shared consumption of experience: nonshared vs. shared but not with giver vs. shared with giver; between subjects) × 2 (gift type: material vs. experiential; within-subject) mixed design. The procedure of study 3 was similar to that of study 1a. Participants first provided the initials of either a close or distant friend (depending on condition) and then were presented with a similar set of gifts to those used in study 1a (see appendix B). In the nonshared condition, the experiential gifts were described as for a single person’s use (e.g., individual massage/spa pass for one person), and participants were told that they would not be able to join the experiences. In the shared-but-not-with-giver condition, the experiential gifts were described as for two people, but participants were told that they would not be able to participate in these experiences, suggesting that the recipient would share the experiences with someone other than the giver. In the shared-with-giver condition, the experiential gifts were described as for two people (e.g., massage/spa day pass with the friend), and participants were asked to assume that they would participate in these experiences with the recipient. For each gift, participants reported their preferences for the gift by using the same scale as in study 2b. In addition, participants reported their knowledge of their friend by indicating 1) how well they knew their friend’s preferences and 2) how well they knew what their friend likes and dislikes (r = .98). Results Gift Preference Our key dependent measure was relative gift preference for experiential gifts compared to material gifts. To calculate this measure, we averaged participants’ ratings for the five experiential gifts and for the five material gifts, respectively, and then created a difference score by subtracting the average rating of material gifts from that of experiential gifts. Thus, a larger number indicates a greater relative preference for experiences over material gifts. Consistent with our previous findings, relative preference for experiential gifts compared to material gifts depended on social distance (F(1, 285) = 21.18, p < .001, ηp2= .07; see table 4 and figure 5). Participants’ relative preference for experiential gifts over material gifts was greater when selecting a gift for a close friend (M = .57) than for a distant friend (M = –.83). TABLE 4 STUDY 3: GIFT PREFERENCE MEAN ACROSS CONDITIONS Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 TABLE 4 STUDY 3: GIFT PREFERENCE MEAN ACROSS CONDITIONS Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 Supporting the notion that experiences are often associated with shared consumption (Caprariello and Reis 2013), we also found a significant main effect of shared consumption on preference for experiences over material gifts (F(2, 285) = 4.07, p = .02, ηp2= .03). Participants preferred experiences more (relative to material goods) when experiential gifts were described as to be shared with the giver than not shared (Mshared-with-giver = .31 vs. Mnonshared = –.71; F(1, 285) = 7.83, p = .006). We found a similar pattern between the shared-but-not-with-giver condition and nonshared condition, but the difference was marginally significant (Mshared-but-not-with-giver = –.03 vs. Mnonshared = –.71; F(1, 285) = 3.49, p = .064). We did not find a significant difference between the two shared conditions (Mshared-with-giver = .31 vs. Mshared-but-not-with-giver = –.03; F < 1). For our purpose, we were interested in whether shared consumption could explain the effects of social distance—that is, whether our key effect (social distance on relative gift preference) was moderated by shared consumption. While this effect was not significantly moderated by a shared consumption (F(2, 285) = 1.91, p = .15), to be thorough we examined two planned contrasts and the simple effect of social distance in each condition. The effect of social distance in the shared-with-giver condition was marginally stronger than the nonshared condition (F(1, 285) = 2.93, p = .088, ηp2= .01) and marginally stronger than the shared-but-not-with-giver condition (F(1, 285) = 2.76, p = .098, ηp2= .01). Next, we tested the simple effect of social distance in each condition. We found a significant social distance effect in the shared-with-giver condition: participants preferred experiential gifts more when selecting a gift for a socially close friend rather than a socially distant friend (Mclose = 1.41 vs. Mdistant = –.78; F(1, 285) = 18.57, p < .001, ηp2= .06). More importantly, and consistent with our hypotheses, the same pattern was found in the shared-but-not-with-giver condition (Mclose = .48 vs. Mdistant = –.50; F(1, 285) = 3.35, p = .068, ηp2= .01) and in the nonshared condition (Mclose = –.22 vs. Mdistant = –1.18; F(1, 285) = 3.51, p = .062, ηp2= .01), although the effect became marginal in these conditions. Mediation Analysis We proposed that the effect of social distance on relative gift preference is driven by perceived knowledge of recipients’ preferences. To test this hypothesis, we first examined the effect of social distance on preference knowledge. Consistent with our prediction, the analysis revealed a main effect of social distance: participants reported better knowledge about close friends than distant friends (Mclose = 9.22 vs. Mdistant = 5.67, F(1, 284) = 212.56, p < .001, ηp2= .43).9 There was no significant main effect of shared consumption or interaction of shared consumption and social distance on preference knowledge (Fs < 1). Next, we tested whether knowledge of recipients’ preferences mediates the effect of social distance on gift preference. A mediation analysis on the relative preference difference score using the PROCESS macro (Hayes 2013, model 4, with shared consumption as a covariate) revealed a significant indirect effect of social distance on relative gift preference through preference knowledge (95% CI = [.1956, 1.1798]; 5,000 samples). Discussion Study 3 provides more compelling evidence for our theory. Consistent with our previous results, participants exhibited stronger relative preferences for experiential gifts over material gifts when they were socially close to their recipients rather than distant. In addition, the mediation analysis revealed that this preference pattern was driven by the increase in knowledge about recipients’ preferences. Although the effect of social distance increases when experiences are shared (with the gift giver), this effect still holds even when experiential gifts are not shared. Taken together, these results suggest that our key effect of social distance cannot be solely explained by the social nature of experiences and this effect is at least partly driven by preference knowledge of recipients. STUDIES 4A AND 4B: SOCIAL RISK AS A MODERATOR We proposed that a gift giver’s relative preference for experiences over material gifts is greater for a close recipient rather than for a distant one because choosing experiential gifts requires a deeper knowledge of a recipient’s individual tastes and preferences. The mediating process of preference knowledge suggests that social risk is an important moderator—specifically, the social risk of choosing a gift that may not match a recipient’s individual preferences. Given that experiential gifts are perceived as more unique (Carter and Gilovich 2010; Dai et al. 2017), we proposed that material goods would be a safer, less socially risky alternative than experiences when givers are socially distant. To test this social risk component, in studies 4a and 4b, we manipulated the social risk associated with experiential gifts by varying the specificity of experiential gifts. Compared to specific experiences (e.g., aroma deep facial massage), general experiences (e.g., spa gift certificate) should require less knowledge of the recipient, involving relatively little social risk. Accordingly, although givers tend to shy away from giving experiential gifts to a distant recipient (vs. close recipient), they are more likely to give experiences when these gifts carry relatively low social risk. Study 4a Method and Procedure We recruited 400 MTurk workers and 399 participated (Mage = 40.17, 50% female, US only, 95% approval or higher, previous workers blocked from participation). The study was a 2 (social distance: close vs. distant; between subjects) × 2 (gift type: low-risk experiential vs. high-risk experiential; between subjects) × 2 (product replicate: spa domain vs. glass accessories; within-subject) mixed design. Similar to the previous studies, participants typed the initials either of a close or distant female friend. Since the product stimuli used in this study (i.e., spa-related product and glass accessories) were female-oriented purchases, we asked participants to imagine that they were buying a birthday gift for a female friend. All participants were presented with two pairwise choices that consisted of one material gift and one experiential gift in each choice. To manipulate the social risk associated with experiential gifts, we varied the specificity of experiential gifts. In the low-risk condition, participants chose between (1a) a spa gift basket (material) and (1b) a gift certificate for a massage, facial, or any other spa treatment (low-risk experience), and between (2a) a glass pendant necklace (material) and (2b) a glass ornament–making class, other art class, or special exhibit entrance at a local museum (low-risk experience). In the high-risk condition, participants chose between (1a) a spa gift basket (material) and (1b) an aroma deep facial massage (high-risk experience), and between (2a) a glass pendant necklace (material) and (2b) a glass pendant–making class (high-risk experience). To test whether participants followed instructions, we asked whether their friend was (1) male or female and (2) “real, fake, or a little bit of both.” Results Three participants did not follow instructions and generated a male friend instead of a female friend, and six participants admitted to listing a fake friend rather than a real person. Consistent with our previous studies, the results reported below exclude these participants from the subsequence analysis, but including these responses leads to similar results.10 Using a generalized estimating equation (GEE) model, we examined participants’ choices between material and experiential gifts. The analysis yielded a significant main effect of social distance (Wald χ2(1) = 13.49, p < .001) and of social risk (Wald χ2(1) = 12.57, p < .001). More importantly, these main effects were qualified by a marginally significant social distance and social risk interaction (Wald χ2 (1) = 2.72, p = .099; see figure 6). Consistent with the previous findings, when experiential gifts carried high social risk, participants were more likely to choose experiential gifts for a close friend rather than for a distant friend (Mclose = 44.4 % vs. Mdistant = 25.5%; Wald χ2(1) = 14.80, p < .001). However, when experiential gifts involved low social risk, the choice of experiential gifts did not differ for a close and distant friend (Mclose = 52.1% vs. Mdistant = 43.9%; Wald χ2 (1) = 2.08, p > .15). The product replicate factor did not have a significant effect, nor did it interact with other factors (ps > .1; see appendix C for the choice distributions across product replicates). FIGURE 6 View largeDownload slide STUDY 4A: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE FIGURE 6 View largeDownload slide STUDY 4A: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE Study 4b Method and Procedure We recruited 500 MTurk workers and 501 participated (Mage = 34.80, 47% female, US only, 95% approval or higher, previous workers blocked from participation). This study was a 2 (social distance: close vs. distant; between subjects) × 2 (gift type: low-risk experiential vs. high-risk experiential; between subjects) × 3 (product replicate: spa domain vs. glass accessories vs. coffee domain; within-subject) mixed design. The procedure was identical to study 4a, except that participants were presented with three pairwise choices instead of two. The third pairwise choice was between (3a) a tea and coffee mug gift basket (material) and (3b) a local café gift card (low-risk experience), or between (3a) a tea and coffee mug gift basket (material) and (3b) a local coffee tasting (high-risk experience). To ensure that our social risk manipulation was effective, we conducted a separate pretest in which 101 MTurk workers rated how risky they thought each gift was to give an average female. These ratings confirmed that high-risk experiences were perceived as more risky than low-risk experiences (Mhigh = 3.47 vs. Mlow = 2.45; F(1, 100) = 73.34, p < .001, ηp2= .42; see appendix D for details). Results Ten participants did not think of a female friend, and seven participants admitted to listing a fake friend rather than a real person. Consistent with our previous studies, the results reported below exclude these participants from the subsequence analysis, and including them does not change the results.11 A GEE model analyzing participants’ choices between material and experiential gifts revealed main effects of social distance (Wald χ2(1) = 4.76, p = .03), social risk (Wald χ2 (1) = 16.42, p < .001), and product replicate (Wald χ2(2) = 40.64, p < .001). Additionally, a social distance × product replicate interaction emerged (Wald χ2(2) = 7.10, p = .03), but a social risk × product replicate interaction was not significant (Wald χ2(2) = 2.83, p > .24). Most importantly, replicating the results of study 4a, a social distance × social risk interaction was marginally significant (Wald χ2(1) = 3.67, p = .055; see figure 7). When experiential gifts carried high social risk, participants were more likely to choose experiential gifts for a close friend rather than for a distant friend (Mclose = 44.1 % vs. Mdistant = 31.6%; Wald χ2(1) = 7.93, p = .005). However, when experiential gifts were associated with low social risk, social distance did not affect the choice of experiential gifts (Mclose = 50.8% vs. Mdistant = 49.9%; Wald χ2(1) = .04, p > .84). The three-way interaction was not significant, which means the social distance × social risk interaction did not vary by product replicate (Wald χ2(2) = 1.27, p > .53; see appendix C for the choice distributions across product replicates). FIGURE 7 View largeDownload slide STUDY 4B: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE FIGURE 7 View largeDownload slide STUDY 4B: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE Single-Paper Meta-Analysis, Replication, and Discussion While the simple effects in studies 4a and 4b were significant at traditional levels, we should note that our interactions were both marginally significant at traditional alpha levels. Thus, for reliability we conducted both a single-paper meta-analysis (SPM; McShane and Böckenholt 2017) and a conceptual replication. An SPM of our studies estimates the social distance × social risk interaction at –.11 (95% CI: –.03, –.20). The confidence interval did not contain zero, demonstrating the robustness of our findings. Examining the simple effects also showed robust results. When experiential gifts carried high social risk, the SPM showed that participants were more likely to choose experiential gifts for a close friend rather than for a distant friend (.14, 95% CI: .09, .21); however, when experiential gifts were associated with low social risk, social distance did not affect the choice of experiential gifts (.03, 95% CI: –.02, .09). For our conceptual replication of studies 4a and 4b, we used 10 replicates in a within-subject design similar to studies 1a and 3 (see web appendix C for study details). We again found that social risk moderated the effect of social distance on gift type preference (F(1, 364) = 15.19, p < .001, ηp2= .04). When experiences were high risk, relative preference for experiential gifts (vs. material gifts) was greater for close recipients (Mexperiential = 4.57 vs. Mmaterial = 3.72) than for distant recipients (Mexperiential = 2.04 vs. Mmaterial = 2.80; F(1, 364) = 23.75, p < .001). But when experiences were low risk, there was no relative preference difference for close recipients (Mexperiential = 3.78 vs. Mmaterial = 3.24) compared to distant recipients (Mexperiential = 2.64 vs. Mmaterial = 1.90; F < 1). In sum, we replicated the effect of social distance when the risk associated with experiential gifts was high, and the effect of social distance was mitigated when the risk was low. In sum, studies 4a and 4b provide further support for our theorization by showing the moderating effect of the social risk associated with experiential gifts. Namely, when the social risk associated with experiences was high, participants’ relative preference for experiential gifts over material gifts was greater for close friends than for distant friends (consistent with our previous studies). However, when experiential gifts did not involve such risk, the effect of social distance on relative preference was mitigated. In other words, when experiential gifts carry the higher social risk of giving a gift that poorly matches a recipient’s preference, consumers shy away from giving these experiential gifts to distant recipients whom they do not know well; however, when this social risk is reduced, socially distant recipients are more likely to be given experiential gifts. GENERAL DISCUSSION Despite evidence suggesting that givers should prefer to give experiential gifts because experiences tend to be more individualized and unique, we found that on average gift givers prefer to give material gifts instead of experiences. The current research explored this mismatched preference for material gifts by focusing on the social distance between givers and recipients, a key component in gift exchange. We proposed that gift givers’ preferences shift toward experiential gifts when they are socially close to their recipients compared to socially distant from their recipients. Further, we hypothesized that this effect of social distance on gift type preference is due to givers’ perceived knowledge of the recipients’ individual preferences—knowledge that is necessary for avoiding the social risk of giving a gift that poorly matches recipients’ preferences. We provided evidence for our hypotheses across eight studies. Study 1a compared preferences for various material and experiential gifts and found that relative preferences for experiential gifts were greater when the recipients were socially close compared to socially distant. Study 1b used open-ended responses that allowed participants to generate a gift that they were willing to give. Consistent with the findings of study 1a, we again found that gifts chosen for close recipients were more experiential than those chosen for distant recipients. By manipulating (studies 2a–c) and measuring (study 3) perceived preference knowledge, we tested whether the effect of social distance was driven by a giver’s perceived knowledge of the recipient’s preferences. Studies 2a–c provide converging evidence that participants’ relative preferences for experiential gifts over material gifts were greater when their knowledge of recipient was high compared to low. In these studies, we also ruled out an alternative explanation based on cost and the differential utilitarian-hedonic dimensions of these gifts. Further supporting the preference knowledge process, study 3 showed that the effect of social distance was mediated by givers’ knowledge of their recipients. Additionally, study 3 ruled out joint/shared consumption as an alternative explanation by demonstrating that the mediating effect of preference knowledge remained constant regardless of shared consumption of experiential gifts. Studies 4a and 4b identified a moderator—the social risk of experiences—that further tested our proposed process. While we found the effect of social distance with relatively high-social-risk experiential gifts, this effect was significantly mitigated with low-social-risk experiential gifts. In other words, when experiential gifts were less socially risky to give, the difference in gift preference due to social distance was reduced. In sum, these results suggest that consumers are willing to give experiences, but only when they feel close enough to the recipient to choose appropriate experiential gifts that avoid the social risk of giving experiences that might not match the recipient’s individual preferences. Alternative Explanations and Future Research We should note that, in addition to social distance, there might be other independent factors influencing consumers’ choices of material versus experiential gifts. In the current research, we focused on social distance because it is a key component to the social exchange of gift giving. To choose unique and individualized gifts, consumers must have knowledge about their recipients, which made social distance a natural and critical variable to study. Nonetheless, given other differences between material and experiential gifts, we did address and attempt to rule out several alternative accounts as to why social distance affects preference for material and experiential gifts. For example, experiential gifts can be consumed with others (Caprariello and Reis 2013), and givers might prefer to share an experience only with close recipients, but not with distant recipients. We tested this explanation in study 3 by manipulating whether the experience was shared with others or consumed alone. We found that the effect of social distance replicated regardless of joint consumption, suggesting that shared consumption does not drive our findings. Another possible explanation is that material goods better adhere to social norms or are more memorable because of their tangible and physical nature (Goodman, Malkoc, and Stephenson 2016). We ran a follow-up study to rule out these two alternatives by examining whether socially distant givers were more concerned about social norms and/or giving a physical reminder (and thus less likely to give experiences). The follow-up study used the same procedures as study 1a but with additional measures. At the end of the survey, we asked participants, “When you were evaluating the gifts, how much did you consider the following dimensions?” (1 = not at all, 10 = very much). We collected social norms and physical reminder data using two measures each (social norms: “Whether the gift is socially acceptable,” and “Whether the gift would be appropriate for our friendship”; physical reminder: “Whether the gift is a physical reminder of me or the relationship,” and “Whether the gift physically persists over time.”). As expected, we replicated our previous effect of social distance on gift type preference such that the relative preference to give experiential gifts (vs. material gifts) was significantly greater when the recipient was socially close (Mexperiential = 4.36 vs. Mmaterial = 3.48) compared to socially distant (Mexperiential = 2.65 vs. Mmaterial = 2.86; F(1, 398) = 24.08, p < .001, ηp2= .06). More importantly, in terms of a physical reminder, the alternative explanation did not pan out: we found that socially distant gift givers actually indicated less of a need for a physical reminder (M = 2.38), compared to close givers (M = 3.07; F(1, 398) = 7.79, p = .006, ηp2= .02). In terms of social norms, we did find that participants were more concerned with social norms when they were selecting a gift for a socially distant friend (M = 5.80) compared to a socially close friend (M = 5.18; F(1, 398) = 4.61, p = .03, ηp2= .01); however, a concern for social norms did not predict preference for material versus experiential gifts (p > .6). Thus, neither a need for a physical reminder nor a concern for social norms is a viable alternative explanation for our findings. Implications and Contribution This research has important contributions to both the gift giving and happiness literatures. First, in terms of happiness, the current findings highlight that an experiential gift might not always be a viable option for givers, despite evidence suggesting that experiences contribute more to recipients’ happiness and strengthen relationships between givers and recipients (Chan and Mogilner 2017; Goodman and Lim 2017; Wallman 2015). Given that money spent on others leads to more happiness than money spent on oneself (Dunn, Aknin, and Norton 2008), it seems that the optimal strategy would be to double down—not only spend money on others, but also buy them experiential gifts. However, based on the current findings, buying experiential gifts seems to come with a tradeoff: gift givers believe that choosing an experiential gift requires more knowledge about recipients’ individual preferences, and thus giving an experiential gift carries greater social risk. Of course, gift givers could be wrong and unnecessarily concerned, and distant receivers may still be better off with an experiential gift. Future research may want to investigate the accuracy and consequences of such choices. To give experiences, gift givers believe that they need the proper knowledge of what their recipients like and dislike, and acquiring this knowledge may require a large investment (in time or social resources). If givers do not have such information, they will try to avoid the social risks in giving an experiential gift that does not match their recipient’s preference. However, if this is merely a perception, then it is possible that simply thinking about all the ways that a giver feels more knowledgeable about the recipient could increase the gifting of experiential gifts. Future research needs to investigate other methods to reduce social risks associated with giving experiential gifts. Second, the studies extend the social relevance theories to experiential purchases. Previous literature has highlighted that experiential consumption promotes a sense of social connectedness and therefore brings greater happiness to consumers (Caprariello and Reis 2013; Chan and Mogilner 2017). Adding to this line of research, the current research suggests that social factors can affect people’s experiential purchases. Our research demonstrates that greater social closeness increases the rate at which consumers give experiential gifts, highlighting social closeness as an important antecedent of experiential giving. The present findings, coupled with the recent finding that experiential gifts can increase social connectedness between givers and receivers (Chan and Mogilner 2017), suggest that further research is needed to explore experiential gift giving and consumption. There is a vast literature exploring the nuances of gift giving, but most of this literature has assumed (perhaps indirectly) that gifts are material in nature—that is, they are, and should be, physically given in a box wrapped neatly with a bow. Our research challenges this assumption and identifies situations where experiential gifts may be preferred. The results also suggest that there is an opportunity for firms to provide more experiential gifts, particularly those types of gifts that entail low social risk. Websites that facilitate giving experiences (such as Excitations, Cloud9 Living, and Blueboard) seem to have already noticed the value of experiential giving. Further, they seem to be aware that experiential gifts are still not the norm, as Excitations’ tagline is “Give the unexpected.” Yet giving experiences does not have to be unexpected or a niche business, and these perceptions may change over time. Just as retailers have adapted strategies to be friendlier to giving material gifts—offering exchanges, gift receipts, gift boxes, and gift cards—perhaps marketers of experiential purchases (e.g., Ticketmaster, Travelocity, or the local theater) should be doing the same: finding ways to lower the risk of experiential gifts by offering exchanges, gift cards, and gift receipts. If firms can provide better offerings that lower the social risk of experiential gift giving, thus lowering the knowledge needed to give an experience, then we may find a future where giving experiences becomes mainstream. DATA COLLECTION INFORMATION Two pilot studies in the introduction were conducted on Amazon Mechanical Turk. The first one was run in 2013 and the second in 2017. Studies 1a and 1b were collected at Washington University in St. Louis in 2015 and 2014, respectively. The conceptual replication in the discussion of study 1a was conducted on MTurk in 2015. Study 2a was collected at The Ohio State University in 2017. Studies 2b, 2c, 3, 4a, and 4b were collected on MTurk, along with the conceptual replication of study 4 and the follow-up study in the General Discussion. The authors conducted studies 2b and 2c in 2017, study 3 in 2016, studies 4a and 4b in 2017, the replication of study 4 in 2016, and the follow-up study in 2015. The data were analyzed by the second author with oversight from the first author. The authors wish to thank Selin Malkoc, Kathleen Vohs, Cindy Cryder, the editorial team, and attendees of the OSU Marketing Seminar series, GAP, and CB Research Lab. Supplementary materials are included in the web appendix accompanying the online version of this article. APPENDIX A: STIMULI AND PRETEST RESULTS (STUDY 1A) MEAN RATING OF EXPERIENTIAL AND MATERIAL GIFTS Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* NOTE.—11-point Likert scale (0 = Material, 10 = Experiential); one-sample t-test; *p < .001. Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* NOTE.—11-point Likert scale (0 = Material, 10 = Experiential); one-sample t-test; *p < .001. APPENDIX B: STIMULI (STUDY 3) Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) View Large Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) View Large APPENDIX C: GIFT CHOICE ACROSS CONDITIONS (STUDIES 4A AND 4B) Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% View Large Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% View Large APPENDIX D: SOCIAL RISK PRETEST (STUDY 4B) Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) View Large Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) View Large Footnotes 1 We asked 100 Mechanical Turk (MTurk) workers (US only, 95% approval or higher) to list up to five gifts they recently received for the holiday season and their last birthday. After reading definitions of material and experiential gifts, they rated each gift as more material (1) or more experiential (7) on a seven-point scale: 81% of holiday gifts and 66% of birthday gifts were rated below the midpoint of 4. 2 To test this assumption (i.e., whether gift givers perceive experiential gifts as involving greater social risk), 188 MTurk workers (US only, 95% approval rating or higher) listed one material gift and one experiential gift and then rated “Which gift would be more risky to give your friend?” (on a 1–9 scale). Confirming our assumption, participants perceived experiential gifts as carrying higher social risk relative to material gifts (t(187) = 5.93, p < .001; see web appendix A for details). 3 Unexpectedly, we also found a marginal interaction between preference knowledge and gift type (F(1, 116) = 3.64, p = .059, ηp2 = .03) on the hedonic nature of the gift. Differences (experience vs. material) in the hedonic nature were slightly smaller for participants in the high-knowledge condition compared to the low-knowledge condition. 4 One respondent did not answer the price estimation measures, leaving 119 usable responses. 5 We ran three separate models, instead of one large model, due to our relatively modest sample size and potential collinearity between the potential alternative factors. 6 The results of study 2b do not change substantially when we include the two participants who imagined a fake friend rather than thought of a real friend. Results revealed a main effect of preference knowledge (F(1, 300) = 29.29, p < .001, ηp2= .09), a main effect of gift type (F(1, 300) = 17.48, p < .001, ηp2= .06), and the hypothesized interaction between gift type and social distance (F(1, 300) = 4.38, p = .037, ηp2= .01). 7 The results of study 2c do not change substantially when we include the 19 participants that did not follow instructions. The hypothesized effect of preference knowledge remains significant (b = .24, Wald χ2(1) = 5.19, p = .02). 8 A preference knowledge × product type interaction was marginally significant on hedonic rating (F(1, 382) = 3.42, p = .065, ηp2 = .01) as well as on utilitarian rating (F(1, 382) = 4.47, p = .04, ηp2 = .01). Compared to low knowledge, high-knowledge participants showed smaller differences for both hedonic and utilitarian natures. 9 One respondent did not answer both preference knowledge measures, leaving 290 usable responses. 10 When we include the nine participants that did not follow instructions, we found a significant main effect of social distance (Wald χ2(1) = 13.00, p < .001) and of social risk (Wald χ2(1) = 11.15, p < .001). The hypothesized interaction between social distance and social risk was marginally significant and virtually unchanged (Wald χ2(1) = 2.58, p = .108). 11 When we include those 17 participants that did not follow instructions, we found a significant main effect of social distance (Wald χ2(1) = 4.53, p = .03), of social risk (Wald χ2 (1) = 16.24, p < .001), and of product replicate (Wald χ2 (2) = 40.65, p < .001). Additionally, a social distance × product replicate interaction was significant (Wald χ2(2) = 8.66, p = .01). Most importantly, the hypothesized interaction between gift type and social distance was marginally significant and virtually unchanged (Wald χ2(1) = 3.77, p = .052). References Aron Arthur , Aron Elaine N. , Smollan Danny ( 1992 ), “Inclusion of Other in the Self Scale and the Structure of Interpersonal Closeness,” Journal of Personality and Social Psychology , 63 4 , 596 – 612 . Google Scholar CrossRef Search ADS Baskin Ernest , Wakslak Cheryl J. , Trope Yaacov , Novemsky Nathan ( 2014 ), “Why Feasibility Matters More to Gift Receivers than to Givers: A Construal-Level Approach to Gift Giving,” Journal of Consumer Research , 41 1 , 169 – 82 . Google Scholar CrossRef Search ADS Belk Russell W. ( 1996 ), “The Perfect Gift,” in Gift Giving: A Research Anthology , ed. Otnes Cele , Beltramini Richard F. , Bowling Green, OH : Bowling Green University Popular Press , 59 – 84 . Belk Russell W. , Coon Gregory S. ( 1993 ), “Gift Giving as Agapic Love: An Alternative to the Exchange Paradigm Based on Dating Experiences,” Journal of Consumer Research , 20 3 , 393 – 417 . Google Scholar CrossRef Search ADS Berscheid Ellen , Snyder Mark , Omoto Allen M. ( 1989 ), “The Relationship Closeness Inventory: Assessing the Closeness of Interpersonal Relationships,” Journal of Personality and Social Psychology , 57 5 , 792 – 807 . Google Scholar CrossRef Search ADS Camerer Colin ( 1988 ), “Gifts as Economic Signals and Social Symbols,” American Journal of Sociology , 94 , S180 – 214 . Google Scholar CrossRef Search ADS Caprariello Peter A. , Reis Harry T. ( 2013 ), “To Do, to Have, or to Share? Valuing Experiences over Material Possessions Depends on the Involvement of Others,” Journal of Personality and Social Psychology , 104 2 , 199 – 215 . Google Scholar CrossRef Search ADS PubMed Carter Travis J. , Gilovich Thomas ( 2010 ), “The Relative Relativity of Experiential and Material Purchases,” Journal of Personality and Social Psychology , 98 1 , 146 – 59 . Google Scholar CrossRef Search ADS PubMed Carter Travis J. , Gilovich Thomas ( 2012 ), “I Am What I Do, Not What I Have: The Centrality of Experiential Purchases to the Self-Concept,” Journal of Personality and Social Psychology , 102 6 , 1304 – 17 . Google Scholar CrossRef Search ADS PubMed Chan Cindy , Mogilner Cassie ( 2017 ), “Experiential Gifts Foster Stronger Social Relationships than Material Gifts,” Journal of Consumer Research , 43 6 , 913 – 31 . Dai, Hengchen, Cindy Chan, and Cassie Mogilner (2017), “People Rely Less on Consumer Reviews for Experiential Purchases than for Material Purchases,” working paper, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2493536. Dunn Elizabeth W. , Aknin Lara B. , Norton Michael I. ( 2008 ), “Spending Money on Others Promotes Happiness,” Science , 319 5870 , 1687 – 8 . Google Scholar CrossRef Search ADS PubMed Foa Edna , Foa Uriel G. ( 1980 ), “Resource Theory: Interpersonal Behavior as Exchange,” in Social Exchange: Advances in Theory and Research , ed. Gergen Kenneth , Greenberg Martin , Willis Richard , New York : Plenum Press , 215 – 37 . Gino Francesca , Flynn Francis J. ( 2011 ), “Give Them What They Want: The Benefits of Explicitness in Gift Exchange,” Journal of Experimental Social Psychology , 47 5 , 915 – 22 . Google Scholar CrossRef Search ADS Goodman Joseph K. , Lim Sarah ( 2017 ), “Experiential Gifts Lead to More Happiness than Material Gifts,” working paper, Fisher College of Business, The Ohio State University, Columbus, OH 43210. Goodman Joseph K. , Malkoc Selin A. , Stephenson Brittney ( 2016 ), “Celebrate or Commemorate? A Material Purchase Advantage When Honoring Special Life Events,” Journal of the Association for Consumer Research , 1 4 , 497 – 508 . Google Scholar CrossRef Search ADS Hayes Andrew F. ( 2013 ), Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach , New York : Guilford Press . Huang Ming-Hui , Yu Shihti ( 2000 ), “Gifts in Romantic Relationships: A Survival Analysis,” Journal of Consumer Psychology , 9 3 , 179 – 88 . Google Scholar CrossRef Search ADS Hoyer Wayne D. , MacInnis Deborah J. , Pieters Rik ( 2012 ), Consumer Behavior , New York : Cengage Learning . Kelley Harold H. , Berscheid Ellen , Christensen Andrew , Harvey John H. , Huston Ted L. , Levinger George ( 1983 ), “Analyzing Close Relationships,” in Close Relationships , ed. Kelley Harold H. , Berscheid Ellen , Christensen Andrew , Harvey John H. , Huston Ted L. , Levinger George , New York : Freeman , 20 – 67 . Komter Aafke E. ( 1996 ), The Gift. An Interdisciplinary Perspective , Amsterdam : Amsterdam University Press . Mauss Marcel ( 1925 ), The Gift: Forms and Functions of Exchange in Archaic Societies , tr. Cunnison Ian , New York : W. W. Norton . McShane Blakeley B. , Böckenholt Ulf ( 2017 ), “Single Paper Meta-Analysis: Benefits for Study Summary, Theory Testing, and Replicability,” Journal of Consumer Research , 43 6 , 1048 – 63 . New York Magazine ( 2017 ), “Gifts, the Ultimate Catalogue,” http://nymag.com/guides/holidays/gifts/2017/. Nicolao Leonardo , Irwin Julie R. , Goodman Joseph K. ( 2009 ), “Happiness for Sale: Do Experiential Purchases Make Consumers Happier than Material Purchases?” Journal of Consumer Research , 36 2 , 188 – 98 . Google Scholar CrossRef Search ADS Otnes Cele , Lowrey Tina M. , Kim Young Chan ( 1993 ), “Gift Selection for Easy and Difficult Recipients: A Social Roles Interpretation,” Journal of Consumer Research , 20 2 , 229 – 44 . Google Scholar CrossRef Search ADS Parks Malcolm R. , Floyd Kory ( 1996 ), “Meanings for Closeness and Intimacy in Friendship,” Journal of Social and Personal Relationships , 13 1 , 85 – 107 . Google Scholar CrossRef Search ADS Robben Henry S. J. , Verhallen Theo M. M. ( 1994 ), “Behavioral Costs as Determinants of Cost Perception and Preference Formation for Gifts to Receive and Gifts to Give,” Journal of Economic Psychology , 15 2 , 333 – 50 . Google Scholar CrossRef Search ADS Rosenzweig Emily , Gilovich Thomas ( 2012 ), “Buyers Remorse or Missed Opportunity? Differential Regrets for Material and Experiential Purchases,” Journal of Personality and Social Psychology , 102 2 , 215 – 23 . Google Scholar CrossRef Search ADS PubMed Schwartz Barry ( 1967 ), “The Social Psychology of the Gift,” American Journal of Sociology , 73 1 , 1 – 11 . Google Scholar CrossRef Search ADS Steffel Mary , Le Boeuf Robyn A. ( 2014 ), “Overindividuation in Gift Giving: Shopping for Multiple Recipients Leads Givers to Choose Unique but Less Preferred Gifts,” Journal of Consumer Research , 40 6 , 1167 – 80 . Google Scholar CrossRef Search ADS Stinson Linda , Ickes William ( 1992 ), “Empathic Accuracy in the Interactions of Male Friends versus Male Strangers,” Journal of Personality and Social Psychology , 62 5 , 787 – 97 . Google Scholar CrossRef Search ADS PubMed Strahilevitz Michal , Myers John G. ( 1998 ), “Donations to Charity as Purchase Incentives: How Well They Work May Depend on What You Are Trying to Sell,” Journal of Consumer Research , 24 4 , 434 – 46 . Google Scholar CrossRef Search ADS Teigen Karl Halvor , Olsen Marina V. G. , Solås Odd Egil ( 2005 ), “Giver–Receiver Asymmetries in Gift Preferences,” British Journal of Social Psychology , 44 1 , 125 – 44 . Google Scholar CrossRef Search ADS PubMed Tully Stephanie M. , Hershfield Hal E. , Meyvis Tom ( 2015 ), “Seeking Lasting Enjoyment with Limited Money: Financial Constraints Increase Preference for Material Goods over Experiences,” Journal of Consumer Research , 42 1 , 59 – 75 . Google Scholar CrossRef Search ADS Van Boven Leaf , Campbell Margaret C. , Gilovich Thomas ( 2010 ), “Stigmatizing Materialism: On Stereotypes and Impressions of Materialistic and Experiential Pursuits,” Personality and Social Psychology Bulletin , 36 4 , 551 – 63 . Google Scholar CrossRef Search ADS PubMed Van Boven Leaf , Gilovich Thomas ( 2003 ), “To Do or to Have? That Is the Question,” Journal of Personality and Social Psychology , 85 6 , 1193 – 202 . Google Scholar CrossRef Search ADS PubMed Wallman James ( 2015 ), Stuffocation: Living More with Less , London : Penguin Books . Ward Morgan K. , Broniarczyk Susan M. ( 2011 ), “It’s Not Me, It’s You: How Gift Giving Creates Giver Identity Threat as a Function of Social Closeness,” Journal of Consumer Research , 38 1 , 164 – 81 . Google Scholar CrossRef Search ADS Ward Morgan K. , Broniarczyk Susan M. ( 2016 ), “Ask and You Shall (Not) Receive: Relationship Signaling Results in Registry Rejection and Worse Gift Choices,” Journal of Marketing Research , 53 6 , 1001 – 18 . Google Scholar CrossRef Search ADS Wooten David B. ( 2000 ), “Qualitative Steps toward an Expanded Model of Anxiety in Gift Giving,” Journal of Consumer Research , 27 1 , 84 – 95 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Consumer Research Oxford University Press

When Consumers Prefer to Give Material Gifts Instead of Experiences: The Role of Social Distance

Loading next page...
 
/lp/ou_press/when-consumers-prefer-to-give-material-gifts-instead-of-experiences-LW6RFjplmK
Publisher
University of Chicago Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0093-5301
eISSN
1537-5277
D.O.I.
10.1093/jcr/ucy010
Publisher site
See Article on Publisher Site

Abstract

Abstract Although previous research suggests that there are hedonic and interpersonal benefits to gifting experiences, consumers often give material gifts rather than experiential gifts. Exploring this mismatch, the current research examines when and why consumers prefer to give material versus experiential gifts. The authors propose that gift givers are more likely to give experiential gifts to socially close recipients than socially distant recipients. Since experiences are perceived as more unique than material goods, givers perceive that choosing an experiential gift requires more specific knowledge of a recipient’s preferences to avoid the greater social risk of giving a poorly matched gift. Eight studies provide converging evidence for the proposed effect of social distance on gift preference and demonstrate that this effect is driven by a giver’s knowledge of a recipient’s preferences. Further supporting the mechanism of preference knowledge, the effect of social distance is moderated by the social risk associated with experiential gifts. When experiences contain little social risk—and thus require less knowledge of a recipient—the effect of social distance is significantly mitigated. Together, these results provide answers for why consumers often prefer to give material gifts over experiences, despite the advantage of giving experiences. gift giving, experiential advantage, material and experiential consumption, social distance Gift giving is ubiquitous. It is practiced by consumers of every culture, and researchers from various disciplines have been studying it for decades (Belk and Foa 1980; Coon 1993; Komter 1996; Mauss 1925; Robben and Verhallen 1994). Gift giving serves multiple functions: gifts signal relationships and information, strengthen social bonds, and fulfill social norms, just to name a few (Belk 1996; Camerer 1988; Huang and Yu 2000; Schwartz 1967; Teigen, Olsen, and Solas 2005). Gifts also demonstrate unselfish, agapic love (Belk and Coon 1993), and people often give gifts with the intention of increasing the happiness and enjoyment of their recipients (Baskin et al 2014; Gino and Flynn 2011; Otnes, Lowry, and Kim 1993; Teigen et al. 2005). Considering these social and hedonic functions of gift giving, it seems that consumers should prefer to give experiences (intangible events that a person lives through, e.g., travel, concert) over material goods (tangible products that are retained over time, e.g., clothing, furniture; Tully, Hershfield, and Meyvis 2015). Previous findings have shown that experiential purchases contribute more to consumer happiness than material purchases (Nicolao, Irwin, and Goodman 2009; Van Boven and Gilovich 2003), suggesting that recipients would derive greater happiness from experiential gifts rather than material gifts (Goodman and Lim 2017; Wallman 2015). More recent research highlights another benefit of giving experiences, demonstrating that experiential gifts strengthen social connections between givers and recipients (Chan and Mogilner 2017). Furthermore, consumers prefer to give unique and individualized gifts (Belk 1996; Steffel and LeBoeuf 2014), which is a quality often associated with experiences rather than material goods (Carter and Gilovich 2012; Dai, Chan, and Mogilner 2017). Despite the hedonic and interpersonal benefits of gifting experiences, consumers seem to prefer to give material gifts rather than experiences. For instance, the vast majority of gifts recommended by gift-giving gurus are material in nature (e.g., see New York Magazine’s 2017 Holiday Gift Guide), and in a survey of 100 Americans we found a vast majority of the gifts listed—81% of holiday gifts and 66% of birthday gifts—were material rather than experiential.1 Why do consumers overwhelmingly prefer to give material gifts? Moreover, are there times when consumers do in fact prefer to give experiences? This research examines when and why consumers often give material gifts over experiential gifts. We propose that the social distance between gift givers and recipients—a key component in gift exchange—influences consumers’ preferences for giving experiences versus material goods. In line with previous literature on interpersonal relationships, we conceptualize social distance in terms of interpersonal closeness, which is characterized by frequent, diverse, and strong interactions between two people (Berscheid, Snyder, and Omoto 1989; Kelley et al. 1983). Because experiences, on average, are perceived as more unique and individualized than material goods (Carter and Gilovich 2012; Dai et al. 2017), gift givers might perceive an experiential gift as requiring more specific knowledge of a recipient’s preferences. Thus, they might believe that giving an experience carries greater social risk of choosing an experiential gift that poorly matches their recipient’s preferences.2 Gift givers are more likely to have this preference knowledge when they are socially close to their recipients compared to when they are socially distant. Accordingly, we propose that the closer givers and recipients become, the more likely gift givers are to give an experience compared to a material gift. Across eight studies, we tested these hypotheses and consistently found that gift givers were more likely to give experiences (vs. material goods) when their recipients were socially close compared to socially distant (studies 1a and 1b). We also demonstrated that this effect was due to givers possessing less knowledge about the individual preferences of distant recipients compared to close recipients (studies 2a, 2b, 2c, and 3). Further, when givers were provided with the opportunity to give experiences low on social risk (i.e., experiential gifts that were less likely to have a preference mismatch and incur a social cost), the effect of social distance on gift preference was significantly attenuated (studies 4a and 4b). Thus, despite the advantage of giving experiences, consumers are less likely to choose experiential gifts for socially distant recipients (vs. socially close recipients) in order to avoid the social risk of giving experiential gifts that poorly match their recipients’ preferences. SOCIAL DISTANCE AND GIFT GIVING Gift giving is a social exchange between two parties. In gift giving, consumers make purchases for others, which raises a challenge for gift givers to correctly infer which gifts might best match the preferences of their recipients (Baskin et al. 2014; Otnes et al. 1993). Beyond finding gifts that simply match their recipients’ preferences, gift givers are also motivated to give gifts that are unique or have some individualized elements (Belk 1996; Belk and Coon 1993). For instance, consumers prefer to give gifts that capture unique characteristics of their recipients and shy away from an item in gift registries, particularly when selecting a gift for a close friend (Ward and Broniarczyk 2016). At the extremes, givers have been shown to even prefer gifts that are overly individualized at the expense of greater utility (Steffel and LeBoeuf 2014). A variety of empirical evidence suggests that experiential gifts are likely to be more effective than material gifts, on average, at achieving this goal of giving gifts that uniquely capture recipients’ preferences. First, compared to material goods, consumers perceive experiences to be less comparable (Carter and Gilovich 2010) and less interchangeable (Rosenzweig and Gilovich 2012). For example, it is not that one concert is necessarily of higher quality than another concert, or that one vacation is always better than another. This lower comparability of experiences suggests that the utility of each experience might vary depending on consumers’ individual preferences. Second, previous research has found that consumers expect their evaluations of experiences to be different from others’ evaluations, which prompts them to rely less on consumer reviews for experiential purchases compared to material purchases (Dai et al. 2017). Third, people believe that compared to material goods, experiential purchases are more informative of a purchaser (Carter and Gilovich 2012) and are more intrinsically motivated (Van Boven, Campbell, and Gilovich 2010). In short, these findings suggest that experiential gifts are perceived to be more individualized gifts that uniquely reflect the recipients’ preferences and tastes. Given this uniqueness of experiences, gift givers might perceive selecting an appropriate experiential gift as requiring more specific knowledge of recipients’ preferences. For example, knowing that your colleague enjoys music and concerts, you may think a ticket for a Beyoncé concert would be a good gift; however, you might not be socially close enough to her to know that she is actually a fan of heavy metal bands and would be embarrassed to even consider such a concert. While gift givers are more motivated to give experiences since they are unique to recipients’ individual preferences, this uniqueness of experiential gifts may also increase the social risk of choosing an experiential gift that poorly matches the recipients’ preferences. Accordingly, gift givers would perceive material goods as a safer and less risky alternative when they do not have sufficient knowledge of recipients. Thus, we propose that without the requisite preference knowledge, gift givers would be less likely to give an experiential gift. Whether gift givers have this requisite knowledge of their recipients’ preferences will likely depend on the social distance between givers and receivers, which has been shown to play an important role in gift giving (Chan and Mogilner 2017; Ward and Broniarczyk 2011, 2016). Relationships can be conceptualized as being more or less distant on a continuum of social closeness (Aron, Aron, and Smollan 1992; Berscheid et al. 1989). Close relationships are developed through building extended and intimate interactions (Berscheid et al. 1989; Kelley et al. 1983) and sharing personal experiences (Parks and Floyd 1996). These interactions naturally allow individuals to develop greater knowledge of each other. Indeed, people predict thoughts and feelings of friends more accurately than those of strangers because they have a deeper understanding of close others (Stinson and Ickes 1992). Accordingly, we propose that gift givers will feel that they have better knowledge of close recipients compared to distant recipients. Further, this knowledge will in turn affect their relative preferences for material versus experiential gifts. As the social distance between givers and receivers decreases, givers will perceive more knowledge of their recipients’ preferences, and they will be more willing to give an experience over a material gift. Alternatively, as the social distance between givers and recipients increases, givers’ knowledge of their recipients will decrease and their relative preferences will shift toward a material gift. More formally, we hypothesize that the more socially distant givers are from their recipients, the more likely they are to give a material gift compared to an experiential gift. Further, we hypothesize that this effect of social distance will be mediated by gift givers’ knowledge of their recipients’ preferences. The role of recipients’ preference knowledge suggests an important moderator: the social risk associated with a gift. We define social risk as the extent to which a gift may not match a recipient’s individual preferences, potentially resulting in harm to one’s social relationship or social standing (Hoyer, MacInnis, and Pieters 2012). The mere notion that a gift may not match recipients’ preferences could elicit anxiety in the gift decision process, and an actual mismatch can risk damaging a social relationship (Wooten 2000). We propose that selecting an experiential gift is perceived as requiring more specific knowledge of recipients’ preferences, which makes gift givers feel more susceptible to the social risk of giving a gift that may not be well liked. Hence, to avoid this social risk, gift givers will, in general, shy away from experiential gifts compared to material gifts when they do not possess the necessary knowledge of their recipients. However, there may be some exceptions whereby experiential gifts could be perceived as less risky to give because they are less likely to mismatch recipients’ preferences and harm a social relationship. For instance, compared to a specific experiential gift (e.g., aroma deep facial massage), a more general experiential gift (e.g., a spa gift certificate) can be used in multiple ways and thus would require less knowledge of recipients’ preferences. Accordingly, gift givers might perceive a more general experiential gift as carrying lower social risk. If givers are provided with an opportunity to give such low-risk experiences, the effect of social distance on gift preferences should be mitigated. Eight studies test our proposed theory (summarized in figure 1). In study 1a, participants indicated greater preferences for experiential gifts relative to material gifts when recipients were socially close compared to socially distant. Similarly, study 1b demonstrated that participants were also more likely to list an experiential gift than a material gift when they were asked to choose a gift for a close friend compared to a distant friend. In studies 2a–c, we tested our preference knowledge theory by manipulating knowledge of recipients’ preferences. Across these three studies, we consistently found that participants exhibited greater preferences for experiential gifts to material gifts when their knowledge of recipients was high compared to low. Further supporting our theory, study 3 revealed that preference knowledge mediated the effect of social distance on gift preference. Study 3 also ruled out joint/shared consumption as an alternative explanation by demonstrating that the effect of social distance on gift preference remained constant regardless of shared consumption of experiential gifts. Finally, studies 4a and 4b tested whether providing gift givers with an opportunity to give experiential gifts low in social risk—less specific gifts that do not require a high degree of knowledge about recipients’ individual preferences—moderates the effect of social distance. We found that the effect of social distance on gift preference was mitigated when givers were provided the opportunity to give less socially risky experiences. FIGURE 1 View largeDownload slide THEORETICAL MODEL AND HYPOTHESES FIGURE 1 View largeDownload slide THEORETICAL MODEL AND HYPOTHESES STUDIES 1A AND 1B: THE ROLE OF SOCIAL DISTANCE IN GIFT PREFERENCE Studies 1a and 1b aimed to provide an initial test of whether social distance influences gift givers’ preferences for experiential and material gifts. We hypothesized that givers’ preferences would shift toward experiential gifts when they choose a gift for socially close recipients compared to socially distant ones. In study 1a, participants rated how much they preferred to give various experiential and material gifts to either a close or distant friend as a holiday gift. Study 1b tested the effect of social distance with a more naturalistic paradigm using an open-ended format. Additionally, study 1b assessed the generalizability of our findings by testing the effect of social distance in another gift-giving occasion (i.e., birthday). Study 1a Method and Procedure Seventy-four undergraduates (Mage = 19.32, 34% female), from various colleges at a midsized private university, participated in this study in exchange for credit in their introductory business course. Sample size was set as large as possible with the constraint of the number of students available to participate in the study. We used the same sample size criteria for all studies using undergraduates. The study was a 2 (social distance: distant vs. close; between subjects) × 2 (gift type: material vs. experiential; within-subject) mixed design. We manipulated social distance by asking participants to think of either a close or distant friend (depending on condition) and write down the initials of that friend. Participants were asked to imagine that they were choosing a holiday gift for the friend and shown a list of 10 gifts of similar cost, five experiential (concert ticket, sightseeing boat ride [available in over 50 cities], ticket for comedy night, wine tour and tasting, cooking classes [four sessions]) and five material (handmade pen, sunglasses, headphones, leather gloves, wallet/purse). A separate pretest (N = 99 via Amazon Mechanical Turk (MTurk), US only, 95% approval or higher) confirmed that the experiential gifts were actually perceived as more experiential than the material gifts (p < .001; see appendix A for details). All gift items were presented in a randomized order, and they were not labeled as experiential or material. Participants were told to assume that costs of the gifts were similar, and they indicated their preferences by rating how much they would like to give each gift to their friend on an 11-point Likert scale (0 = definitely would not give, 10 = definitely would give). Results We averaged participants’ preference ratings of the experiential and material gifts, respectively, and created a preference score for each type of gift. A repeated-measures analysis of variance (ANOVA) was conducted on this preference score with social distance as a between-subjects factor and gift type as a within-subject factor. The analysis yielded a main effect of social distance such that participants indicated greater preferences for all gifts when they were giving a gift to a close friend rather than a distant friend (Mclose = 3.68 vs. Mdistant = 2.47; F(1, 72) = 9.22, p = .003, ηp2= .11). More importantly, this main effect was qualified by a social distance × gift type interaction (F(1, 72) = 15.02, p < .001, ηp2= .17, see figure 2). As predicted, the relative preference for experiences (vs. material goods) was greater for a close friend (Mexperiential = 4.20 vs. Mmaterial = 3.16; F(1, 72) = 8.88, p = .004, ηp2= .11) than for a distant friend (Mexperiential = 1.93 vs. Mmaterial = 3.01; F(1, 72) = 6.58, p = .012, ηp2= .08). FIGURE 2 View largeDownload slide STUDY 1A: THE EFFECT OF SOCIAL DISTANCE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. FIGURE 2 View largeDownload slide STUDY 1A: THE EFFECT OF SOCIAL DISTANCE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. To check the robustness of our effect, we conducted a conceptual replication with a larger sample (300 MTurk workers, 294 responses, Mage = 34.44, 45% female, US only, 95% approval or higher) following the same procedure as study 1a. Consistent with the results of study 1a, we again found a significant social distance by gift type interaction (F(1, 292) = 31.38, p < .001, ηp2= .10). Experiences were more preferred than material gifts for a close friend (Mexperiential = 4.58 vs. Mmaterial = 3.66; F(1, 292) = 27.72, p < .001, ηp2= .09) compared to a distant friend (Mexperiential = 2.43 vs. Mmaterial = 2.89; F(1, 292) = 7.01, p = .009, ηp2= .02). Study 1b Method and Procedure Two hundred seventy-six undergraduate students (Mage = 19.24, 43% female), from various colleges at a midsized private university, participated in this study in exchange for credit in their introductory business course. Depending on condition, participants were asked to think of either a close or distant friend whose birthday was coming up and type the initials of that friend. Participants were asked to indicate a gift they would like to give to the friend as a birthday gift (open-ended; see web appendix B for product categories of the gifts listed by participants). Next, they rated “to what extent the gift was more material or more experiential” on a nine-point Likert scale (1 = more material, 5 = both, 9 = more experiential). Finally, participants indicated the approximate cost of their gift and answered some demographic questions. Results Replicating the findings of study 1a, we found a significant effect of social distance on preference for gift type (t(274) = 4.13, p < .001, Cohen’s d= .50). Participants were more (less) likely to list an experiential gift (a material gift) for a close friend (M = 4.36) than for a distant one (M = 3.08). Moreover, the effect of social distance remained significant (p < .01) even when we controlled for price of the gift (log-transformed), age, and gender. Discussion In sum, the results of studies 1a and 1b provide converging evidence for our hypothesis that a giver’s relative preference for material and experiential gifts changes as a function of his or her social closeness to the recipient. Notably, this pattern was consistently observed across the two different gift-giving occasions (i.e., holidays and birthdays) and regardless of whether the gift items were generated by the experimenter (study 1a) or participants (study 1b). STUDIES 2A, 2B, AND 2C: PREFERENCE KNOWLEDGE Studies 2a–c were designed to test the proposed underlying mechanism of our effect, namely givers’ knowledge of recipients’ preferences. We posited that more knowledge of close recipients (vs. distant recipients) would translate into greater preferences for experiential gifts over material gifts. To test this prediction, we directly manipulated gift givers’ knowledge of their recipients’ preferences. In studies 2a and 2b, participants rated their preferences to give various material and experiential gifts to a friend either whom they knew well or did not know well (between subjects). Study 2c was designed to provide a more conservative test of our effect by using similar material and experiential goods. Namely, participants chose between material and experiential gifts in the same product category (i.e., home spa gift basket vs. aroma deep facial massage). Further, in these three studies, we aimed to rule out several alternative explanations. It is possible that experiential gifts are viewed as more expensive, more hedonic, and/or less utilitarian, or perhaps they are viewed differently based on social distance due to a different mind-set. Therefore, to rule out these alternatives, we tested whether our effect remains significant after controlling for perceived cost and/or the hedonic and utilitarian nature of the gifts, or whether these alternative factors could explain our findings. Study 2a Methods and Procedure One hundred twenty undergraduate students (Mage = 21.48, 52% female) at a large public university participated in this study in exchange for credit in their introductory business course. The study was a 2 (preference knowledge: high knowledge vs. low knowledge) × 2 (product type: material vs. experiential) between-subjects design. To manipulate preference knowledge, we asked participants to write the initials of a friend whom they knew very well or whom they did not know very well, depending on condition. Next, we asked participants to imagine that they were buying a birthday gift for their friend. Depending on condition, participants were presented with either five material gifts or five experiential gifts (identical gifts used in study 1a). For each gift, participants rated the extent to which they would like to give it as a birthday gift to their friend (0 = definitely would not give, 10 = definitely would give). Although we specified that the price of the gifts was similar, participants might perceive the general price of certain gifts as more expensive than others, which might explain our effect. To test this possibility, we asked participants to estimate the cost of each gift (open-ended). In addition, participants rated the extent to which each of the gifts was hedonic and utilitarian (0 = definitely not hedonic/utilitarian at all, 10 = definitely hedonic/utilitarian) after reading the definition of a hedonic and a utilitarian purchase (hedonic was defined as something for pleasure, fantasy, and/or fun; utilitarian was defined as something to fulfill a basic need or to accomplish a functional or practical task, Strahilevitz and Myers 1998). Finally, as a manipulation check, we asked participants to indicate 1) how well they know their friend’s preferences and 2) how well they know what their friend likes and dislikes. Manipulation Check Analyzing participants’ averaged responses to two items on perceived knowledge (r = .95), a 2 (preference knowledge: high vs. low) × 2 (gift type: material vs. experiential) ANOVA revealed only a main effect of preference knowledge: the high-knowledge condition perceived greater knowledge of their recipient’s preferences than those in the low-knowledge condition (Mhigh = 7.95 vs. Mlow = 3.53; F(1, 116) = 248.83, p < .001, ηp2= .68). The interaction and the other main effect were not significant (Fs < 1). Gift Preference For our dependent measure, we created a gift preference index by averaging participants’ preference ratings of the five material or experiential gifts. A 2 (social distance: close vs. distant) × 2 (gift type: material vs. experiential) ANOVA found two main effects: participants in the high-knowledge condition indicated greater preferences for all gifts than those in the low-knowledge condition (Mhigh = 4.27 vs. Mlow = 3.38; (F(1, 116) = 8.28, p = .005, ηp2= .07), and preferences were greater for experiential gifts than for material gifts (Mexperiential = 4.15 vs. Mmaterial = 3.51; F(1, 116) = 4.31, p = .04, ηp2= .04). More importantly, and central to our hypothesis, these main effects were qualified by a marginally significant social distance × gift type interaction (F(1, 116) = 2.92, p = .090, ηp2= .02; see figure 3): relative preferences for experiential gifts (vs. material gifts) were greater in the high-knowledge condition (Mexperiential = 4.85 vs. Mmaterial = 3.69; F(1, 116) = 7.16, p = .009, ηp2= .06) than in the low-knowledge condition (Mexperiential = 3.44 vs. Mmaterial = 3.33; F < 1). FIGURE 3 View largeDownload slide STUDY 2A: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error. FIGURE 3 View largeDownload slide STUDY 2A: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error. FIGURE 4 View largeDownload slide STUDY 2B: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. FIGURE 4 View largeDownload slide STUDY 2B: THE EFFECT OF PREFERENCE KNOWLEDGE ON GIFT PREFERENCE NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. Testing Alternative Accounts We conducted a set of analyses to examine three alternative accounts—hedonic, utilitarian, and cost. First, participants did perceive material and experiential purchases differently in terms of cost and their hedonic and utilitarian nature. Compared to material gifts, experiential gifts were perceived as more hedonic (Mexperiential = 7.18 vs. Mmaterial = 4.44; F(1, 116) = 64.98, p < .001, ηp2 = .36),3 less utilitarian (Mexperiential = 2.76 vs. Mmaterial = 5.55; F(1, 116) = 79.83, p < .001, ηp2 = .41), and more expensive (log-transformed price; Mexperiential = 4.38 vs. Mmaterial = 3.89; F(1, 115) = 17.36, p < .001, ηp2 = .13).4 Second, if our key effect is driven by any of these alternative accounts, then we would expect one of these accounts to interact with preference knowledge to predict gift preference. To test these alternatives, we ran three separate models5—one for hedonic, one for utilitarian, and one for cost—that consisted of our basic model (gift type, preference knowledge, and gift type × preference knowledge interaction) plus each alternative account variable and its interaction with preference knowledge (details of each model are in table 1). The hedonic model did not reveal a significant hedonic nature × preference knowledge interaction (F < 1), and our key interaction remained marginally significant (preference knowledge × gift type, F(1, 114) = 3.12, p = .080, ηp2 = .03). The utilitarian model also did not yield a significant utilitarian nature × preference knowledge interaction (F < 1), but our key interaction also became nonsignificant (F(1, 114) = .75, p = .39, ηp2 = .01). Lastly, the cost model also did not find a significant preference knowledge × cost interaction (F < 1), and our key interaction remained marginally significant (F(1, 113) = 2.88, p = .093, ηp2 = .02). In sum, none of these alternative models predicted gift preference; thus, they did not provide evidence of mediation by any of the alternative explanations. TABLE 1 STUDY 2A: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 a Key hypothesized effect. b Potential alternative explanations. TABLE 1 STUDY 2A: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 Model 1 Model 2 Model 3 Model 4 Variable F p ηp2 F p ηp2 F p ηp2 F p ηp2 Preference knowledge 8.28 .005 .07 .70 .41 .01 2.47 .12 .02 1.20 .28 .01 Material–experience 4.31 .04 .04 .00 .97 .00 5.91 .017 .05 3.02 .085 .03 Material–experience × Preference knowledgea 2.92 .09 .02 3.12 .08 .03 .75 .39 .01 2.88 .093 .02 Hedonic 7.79 .006 .06 Hedonic × Preference knowledgeb .00 .99 .00 Utilitarian 1.58 .21 .01 Utilitarian × Preference knowledgeb .30 .58 .00 Cost .00 .95 .00 Cost × Preference knowledgeb .45 .51 .00 a Key hypothesized effect. b Potential alternative explanations. Study 2b Methods and Procedure We recruited 300 MTurk workers and 302 participated (Mage = 35.33, 54% female, US only, 95% approval or higher, previous workers blocked from participation) in a 2 (preference knowledge: high vs. low; between subjects) × 2 (gift type: material vs. experiential; within-subject) mixed design. The procedure was similar to that of study 2a, except that the material-experiential factor was within-subject—in other words, participants rated all 10 gifts (five material goods and five experiences). Participants wrote the initials of a friend either whom they knew very well (high-knowledge condition) or whom they did not know very well (low-knowledge condition). Next, they wrote a short paragraph about how well they knew/did not know their friend. As in study 2a, participants rated the extent to which they would like to give each of the gifts to the friend as a birthday gift (0 = definitely would not give, 10 = definitely would give). Participants estimated the price of each gift and indicated whether their friend was “real, fake, or a little bit of both.” Results Two participants reported that their friend was not a real person but a fake. The results reported below exclude these participants from the subsequence analysis; however, including these participants did not substantially change the findings or significance.6 A repeated-measures ANOVA with preference knowledge (high vs. low) as a between-subjects factor and product type (material vs. experiential) as a within-subject factor yielded a significant effect of preference knowledge (F(1, 298) = 28.23, p < .001, ηp2= .09) and that of gift type (F(1, 298) = 17.02, p < .001, ηp2= .05). Most importantly, these main effects were qualified by a preference knowledge × gift type interaction (F(1, 298) = 4.57, p = .03, ηp2= .02; see figure 4): participants preferred experiential gifts more in the high-knowledge condition (Mexperiential = 4.67 vs. Mmaterial = 3.85; F(1, 298) = 19.62, p < .001, ηp2 = .06) than in the low-knowledge condition (Mexperiential = 3.35 vs. Mmaterial = 3.10; F(1, 298) = 1.98, p > .16, ηp2 = .01). FIGURE 5 View largeDownload slide STUDY 3: GIFT PREFERENCE ACROSS CONDITIONS NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. FIGURE 5 View largeDownload slide STUDY 3: GIFT PREFERENCE ACROSS CONDITIONS NOTE.—Error bars represent +/–1 standard error; not corrected for within-subject variance. Testing an Alternative Account One might argue that gift givers prefer to give more expensive gifts to recipients they have more preference knowledge of and these givers perceive experiential gifts as more expensive. Although we specified that the price of the gifts was similar, participants did perceive the general cost of experiential gifts as higher than material gifts (Mexperiential = $92.52 vs. Mmaterial = $51.23, F(1, 298) = 31.72, p < .001, ηp2 = .10). To test whether cost explains our effect, we calculated a difference score in the estimated price of material and experiential gifts and added this cost difference variable, along with its interaction with knowledge preference, to our basic model (details are in table 2). The results did not show either a significant main effect of cost difference or a cost difference × preference knowledge interaction (Fs < 1). Further, our key interaction between preference knowledge and gift type was still marginally significant (F(1, 296) = 3.50, p = .062, ηp2= .01). TABLE 2 STUDY 2B: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 a Key hypothesized effect. b Potential alternative explanations. TABLE 2 STUDY 2B: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 Model 1 Model 2 Variable F p ηp2 F p ηp2 Preference knowledge 28.23 <.001 .09 22.69 <.001 .07 Material–experience 17.02 <.001 .05 11.03 .001 .04 Material–experience × Preference knowledgea 4.57 .03 .02 3.50 .062 .01 Price difference .12 .72 .00 Price difference × Preference knowledgeb .07 .80 .00 Price difference × Material–experience 1.04 .31 .00 Price difference × Preference knowledge × Material–experience .18 .67 .00 a Key hypothesized effect. b Potential alternative explanations. Study 2c Methods and Procedure We recruited 400 MTurk workers and 403 participated (Mage = 35.03, 51% female, US only, 95% approval or higher, previous workers blocked from participation). Similar to studies 2a and 2b, participants first wrote the initials of a female friend whom they knew very well (high-knowledge condition) or whom they did not know very well (low-knowledge condition) and then described how well they knew or did not know her. Since the product stimuli (i.e., spa-related product) used in this study were female-oriented purchases, we controlled for potential gender effects by asking participants to imagine that they were buying a birthday gift for a female friend. Participants were told that they narrowed a birthday gift to two items—a home spa gift basket (material gift) and an aroma deep facial massage (experiential gift)—and were asked to make a choice between the two. Although we attempted to minimize any possible differences other than the material-experiential dimension by using similar gifts, participants could still perceive these items differently. To rule out this possibility, we again measured participants’ perception of the hedonic and utilitarian nature of each gift by using similar measures as in study 2a (1 = definitely not hedonic/utilitarian at all, 7 = definitely hedonic/utilitarian). Lastly, participants reported the gender of their friend as well as whether their friend was “real, fake, or a little bit of both.” Results Six participants reported that they listed a fake friend rather than a real person, and 13 participants thought of a male friend instead of a female friend. The results reported below exclude them from the subsequence analysis; however, including these participants did not substantially change the findings or significance.7 As predicted, choice of experiential gift (vs. material gift) was significantly greater in the high-knowledge condition (39.27%) compared to the low-knowledge condition (28.50%; b = .24, Wald χ2(1) = 4.94, p = .03). Testing an Alternative Account To examine whether this effect was driven by differences in the hedonic or utilitarian nature of the gifts, we conducted a set of additional analyses. First, compared to material gifts, experiential gifts were indeed rated as more hedonic (Mexperiential = 5.72 vs. Mmaterial = 4.79; F(1, 382) = 134.89, p < .001, ηp2= .26) and as less utilitarian (Mexperiential = 2.64 vs. Mmaterial = 3.81; F(1, 382) = 166.40, p < .001, ηp2= .30).8 Second, if our key effect is driven by any of these alternative accounts, then we would expect one of these accounts to interact with preference knowledge to predict gift choice. To test these alternatives, we created two difference scores (one hedonic and one utilitarian), subtracting the material gift rating from the experiential gift rating in terms of both the hedonic and utilitarian nature of the gifts. Then we conducted two logistic regressions with gift choice (0 = material gift, 1 = experiential gift) as the dependent variable and preference knowledge (–1 = low, 1 = high), hedonic or utilitarian difference score (mean-centered), and their interaction as predictors (details of each model are in table 3). In the hedonic model, we found that the hedonic difference score was marginally significant (b = .13, Wald χ2(1) = 3.22, p = .073), but the hedonic difference score × preference knowledge interaction was not (p = .83), and our key effect of preference knowledge on gift choice remained significant (b = .26, Wald χ2(1) = 5.61, p = .02). In the utilitarian model, the utilitarian difference score was significant (b = .21, Wald χ2(1) = 10.72, p = .001), but the utilitarian difference score × preference knowledge interaction was not (p > .60). Our key effect of preference knowledge decreased slightly to become marginally significant (b = .20, Wald χ2(1) = 3.34, p = .068). In sum, consistent with studies 2a and 2b, none of these alternative models led to a significant mediator that predicted our hypothesized gift preference × preference knowledge interaction, suggesting that they are not viable alternative explanations. TABLE 3 STUDY 2C: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 a Key hypothesized effect. b Potential alternative explanations. TABLE 3 STUDY 2C: ANALYSIS AND ALTERNATIVE MODELS Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 Model 1 Model 2 Model 3 Variable b Wald χ2 p b Wald χ2 p b Wald χ2 p Preference knowledgea .24 4.94 .03 .26 5.61 .02 .20 3.34 .068 Hedonic difference .13 3.22 .073 Hedonic difference × Preference knowledgeb .01 .04 .83 Utilitarian difference .21 10.72 .001 Utilitarian difference × Preference knowledgeb .03 .27 .60 a Key hypothesized effect. b Potential alternative explanations. Discussion By manipulating the knowledge that a giver has of his or her recipient’s preferences, studies 2a–c showed that consumers were more likely to give experiential gifts than material gifts when their knowledge of the recipient’s preference was high compared to low. We consistently observed this shift in gift preference across different samples (i.e., undergraduate and MTurk workers) and different methodologies (i.e., preference rating and choice, between-subjects and within-subject). Studies 2a–c also provide evidence suggesting that our effect cannot be fully explained by cost and the differential utilitarian or hedonic dimensions of these gifts. In sum, the results provide support for our theoretical process that consumers are likely to prefer material gifts relatively more than experiential gifts when they do not have sufficient knowledge of recipients’ preferences. STUDY 3: SHARED CONSUMPTION Our goal for study 3 was twofold. First, we sought to capture more direct evidence on our proposed mechanism of preference knowledge. Thus, we measured gift givers’ perceived knowledge of their recipients’ preferences and tested whether this knowledge mediates the effect of social distance on gift type. Second, we aimed to rule out the alternative explanation that the effect of social distance is driven by the shared nature of experiential gifts. Experiences are often consumed with others (Caprariello and Reis 2013), and participants might have preferred to give gifts that they could experience with their recipient (shared consumption) only when they were socially close to the recipient. To examine this alternative account, we manipulated whether the experiential gifts were jointly consumed by givers and recipients or consumed alone. If our findings are due to the shared consumption of experiences, then gift givers should prefer to give experiential gifts for a close recipient only when they can participate in the experiences with the recipient. Alternatively, if social distance affects gift preference independent of shared consumption, then the effect of social distance should not interact with shared consumption. Further, to distinguish sharing between the giver and recipient from sharing in general, we added a third condition where experiential gifts were described as to be shared with someone else (but not with the giver). Method and Procedure We recruited 300 participants via MTurk and 291 responded (Mage = 32.47, 47% female, US only, 95% approval or higher, previous workers blocked from participation). Sample size was set based on a power analysis indicating that we would need 139 participants to replicate the effect with 80% power, but given that we were testing a potential moderator as an alternative explanation, we more than doubled our sample size to be conservative. The study was a 2 (social distance: close vs. distant; between subjects) × 3 (shared consumption of experience: nonshared vs. shared but not with giver vs. shared with giver; between subjects) × 2 (gift type: material vs. experiential; within-subject) mixed design. The procedure of study 3 was similar to that of study 1a. Participants first provided the initials of either a close or distant friend (depending on condition) and then were presented with a similar set of gifts to those used in study 1a (see appendix B). In the nonshared condition, the experiential gifts were described as for a single person’s use (e.g., individual massage/spa pass for one person), and participants were told that they would not be able to join the experiences. In the shared-but-not-with-giver condition, the experiential gifts were described as for two people, but participants were told that they would not be able to participate in these experiences, suggesting that the recipient would share the experiences with someone other than the giver. In the shared-with-giver condition, the experiential gifts were described as for two people (e.g., massage/spa day pass with the friend), and participants were asked to assume that they would participate in these experiences with the recipient. For each gift, participants reported their preferences for the gift by using the same scale as in study 2b. In addition, participants reported their knowledge of their friend by indicating 1) how well they knew their friend’s preferences and 2) how well they knew what their friend likes and dislikes (r = .98). Results Gift Preference Our key dependent measure was relative gift preference for experiential gifts compared to material gifts. To calculate this measure, we averaged participants’ ratings for the five experiential gifts and for the five material gifts, respectively, and then created a difference score by subtracting the average rating of material gifts from that of experiential gifts. Thus, a larger number indicates a greater relative preference for experiences over material gifts. Consistent with our previous findings, relative preference for experiential gifts compared to material gifts depended on social distance (F(1, 285) = 21.18, p < .001, ηp2= .07; see table 4 and figure 5). Participants’ relative preference for experiential gifts over material gifts was greater when selecting a gift for a close friend (M = .57) than for a distant friend (M = –.83). TABLE 4 STUDY 3: GIFT PREFERENCE MEAN ACROSS CONDITIONS Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 TABLE 4 STUDY 3: GIFT PREFERENCE MEAN ACROSS CONDITIONS Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 Shared consumption of experience Social distance Material gifts Experiential gifts Relative gift preference (Experiential – Material) Nonshared Distant 3.62 2.45 –1.18 Close 3.96 3.74 –.22 Shared but not with giver Distant 3.06 2.56 –.50 Close 3.21 3.69 .48 Shared with giver Distant 3.11 2.32 –.78 Close 3.43 4.84 1.41 Supporting the notion that experiences are often associated with shared consumption (Caprariello and Reis 2013), we also found a significant main effect of shared consumption on preference for experiences over material gifts (F(2, 285) = 4.07, p = .02, ηp2= .03). Participants preferred experiences more (relative to material goods) when experiential gifts were described as to be shared with the giver than not shared (Mshared-with-giver = .31 vs. Mnonshared = –.71; F(1, 285) = 7.83, p = .006). We found a similar pattern between the shared-but-not-with-giver condition and nonshared condition, but the difference was marginally significant (Mshared-but-not-with-giver = –.03 vs. Mnonshared = –.71; F(1, 285) = 3.49, p = .064). We did not find a significant difference between the two shared conditions (Mshared-with-giver = .31 vs. Mshared-but-not-with-giver = –.03; F < 1). For our purpose, we were interested in whether shared consumption could explain the effects of social distance—that is, whether our key effect (social distance on relative gift preference) was moderated by shared consumption. While this effect was not significantly moderated by a shared consumption (F(2, 285) = 1.91, p = .15), to be thorough we examined two planned contrasts and the simple effect of social distance in each condition. The effect of social distance in the shared-with-giver condition was marginally stronger than the nonshared condition (F(1, 285) = 2.93, p = .088, ηp2= .01) and marginally stronger than the shared-but-not-with-giver condition (F(1, 285) = 2.76, p = .098, ηp2= .01). Next, we tested the simple effect of social distance in each condition. We found a significant social distance effect in the shared-with-giver condition: participants preferred experiential gifts more when selecting a gift for a socially close friend rather than a socially distant friend (Mclose = 1.41 vs. Mdistant = –.78; F(1, 285) = 18.57, p < .001, ηp2= .06). More importantly, and consistent with our hypotheses, the same pattern was found in the shared-but-not-with-giver condition (Mclose = .48 vs. Mdistant = –.50; F(1, 285) = 3.35, p = .068, ηp2= .01) and in the nonshared condition (Mclose = –.22 vs. Mdistant = –1.18; F(1, 285) = 3.51, p = .062, ηp2= .01), although the effect became marginal in these conditions. Mediation Analysis We proposed that the effect of social distance on relative gift preference is driven by perceived knowledge of recipients’ preferences. To test this hypothesis, we first examined the effect of social distance on preference knowledge. Consistent with our prediction, the analysis revealed a main effect of social distance: participants reported better knowledge about close friends than distant friends (Mclose = 9.22 vs. Mdistant = 5.67, F(1, 284) = 212.56, p < .001, ηp2= .43).9 There was no significant main effect of shared consumption or interaction of shared consumption and social distance on preference knowledge (Fs < 1). Next, we tested whether knowledge of recipients’ preferences mediates the effect of social distance on gift preference. A mediation analysis on the relative preference difference score using the PROCESS macro (Hayes 2013, model 4, with shared consumption as a covariate) revealed a significant indirect effect of social distance on relative gift preference through preference knowledge (95% CI = [.1956, 1.1798]; 5,000 samples). Discussion Study 3 provides more compelling evidence for our theory. Consistent with our previous results, participants exhibited stronger relative preferences for experiential gifts over material gifts when they were socially close to their recipients rather than distant. In addition, the mediation analysis revealed that this preference pattern was driven by the increase in knowledge about recipients’ preferences. Although the effect of social distance increases when experiences are shared (with the gift giver), this effect still holds even when experiential gifts are not shared. Taken together, these results suggest that our key effect of social distance cannot be solely explained by the social nature of experiences and this effect is at least partly driven by preference knowledge of recipients. STUDIES 4A AND 4B: SOCIAL RISK AS A MODERATOR We proposed that a gift giver’s relative preference for experiences over material gifts is greater for a close recipient rather than for a distant one because choosing experiential gifts requires a deeper knowledge of a recipient’s individual tastes and preferences. The mediating process of preference knowledge suggests that social risk is an important moderator—specifically, the social risk of choosing a gift that may not match a recipient’s individual preferences. Given that experiential gifts are perceived as more unique (Carter and Gilovich 2010; Dai et al. 2017), we proposed that material goods would be a safer, less socially risky alternative than experiences when givers are socially distant. To test this social risk component, in studies 4a and 4b, we manipulated the social risk associated with experiential gifts by varying the specificity of experiential gifts. Compared to specific experiences (e.g., aroma deep facial massage), general experiences (e.g., spa gift certificate) should require less knowledge of the recipient, involving relatively little social risk. Accordingly, although givers tend to shy away from giving experiential gifts to a distant recipient (vs. close recipient), they are more likely to give experiences when these gifts carry relatively low social risk. Study 4a Method and Procedure We recruited 400 MTurk workers and 399 participated (Mage = 40.17, 50% female, US only, 95% approval or higher, previous workers blocked from participation). The study was a 2 (social distance: close vs. distant; between subjects) × 2 (gift type: low-risk experiential vs. high-risk experiential; between subjects) × 2 (product replicate: spa domain vs. glass accessories; within-subject) mixed design. Similar to the previous studies, participants typed the initials either of a close or distant female friend. Since the product stimuli used in this study (i.e., spa-related product and glass accessories) were female-oriented purchases, we asked participants to imagine that they were buying a birthday gift for a female friend. All participants were presented with two pairwise choices that consisted of one material gift and one experiential gift in each choice. To manipulate the social risk associated with experiential gifts, we varied the specificity of experiential gifts. In the low-risk condition, participants chose between (1a) a spa gift basket (material) and (1b) a gift certificate for a massage, facial, or any other spa treatment (low-risk experience), and between (2a) a glass pendant necklace (material) and (2b) a glass ornament–making class, other art class, or special exhibit entrance at a local museum (low-risk experience). In the high-risk condition, participants chose between (1a) a spa gift basket (material) and (1b) an aroma deep facial massage (high-risk experience), and between (2a) a glass pendant necklace (material) and (2b) a glass pendant–making class (high-risk experience). To test whether participants followed instructions, we asked whether their friend was (1) male or female and (2) “real, fake, or a little bit of both.” Results Three participants did not follow instructions and generated a male friend instead of a female friend, and six participants admitted to listing a fake friend rather than a real person. Consistent with our previous studies, the results reported below exclude these participants from the subsequence analysis, but including these responses leads to similar results.10 Using a generalized estimating equation (GEE) model, we examined participants’ choices between material and experiential gifts. The analysis yielded a significant main effect of social distance (Wald χ2(1) = 13.49, p < .001) and of social risk (Wald χ2(1) = 12.57, p < .001). More importantly, these main effects were qualified by a marginally significant social distance and social risk interaction (Wald χ2 (1) = 2.72, p = .099; see figure 6). Consistent with the previous findings, when experiential gifts carried high social risk, participants were more likely to choose experiential gifts for a close friend rather than for a distant friend (Mclose = 44.4 % vs. Mdistant = 25.5%; Wald χ2(1) = 14.80, p < .001). However, when experiential gifts involved low social risk, the choice of experiential gifts did not differ for a close and distant friend (Mclose = 52.1% vs. Mdistant = 43.9%; Wald χ2 (1) = 2.08, p > .15). The product replicate factor did not have a significant effect, nor did it interact with other factors (ps > .1; see appendix C for the choice distributions across product replicates). FIGURE 6 View largeDownload slide STUDY 4A: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE FIGURE 6 View largeDownload slide STUDY 4A: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE Study 4b Method and Procedure We recruited 500 MTurk workers and 501 participated (Mage = 34.80, 47% female, US only, 95% approval or higher, previous workers blocked from participation). This study was a 2 (social distance: close vs. distant; between subjects) × 2 (gift type: low-risk experiential vs. high-risk experiential; between subjects) × 3 (product replicate: spa domain vs. glass accessories vs. coffee domain; within-subject) mixed design. The procedure was identical to study 4a, except that participants were presented with three pairwise choices instead of two. The third pairwise choice was between (3a) a tea and coffee mug gift basket (material) and (3b) a local café gift card (low-risk experience), or between (3a) a tea and coffee mug gift basket (material) and (3b) a local coffee tasting (high-risk experience). To ensure that our social risk manipulation was effective, we conducted a separate pretest in which 101 MTurk workers rated how risky they thought each gift was to give an average female. These ratings confirmed that high-risk experiences were perceived as more risky than low-risk experiences (Mhigh = 3.47 vs. Mlow = 2.45; F(1, 100) = 73.34, p < .001, ηp2= .42; see appendix D for details). Results Ten participants did not think of a female friend, and seven participants admitted to listing a fake friend rather than a real person. Consistent with our previous studies, the results reported below exclude these participants from the subsequence analysis, and including them does not change the results.11 A GEE model analyzing participants’ choices between material and experiential gifts revealed main effects of social distance (Wald χ2(1) = 4.76, p = .03), social risk (Wald χ2 (1) = 16.42, p < .001), and product replicate (Wald χ2(2) = 40.64, p < .001). Additionally, a social distance × product replicate interaction emerged (Wald χ2(2) = 7.10, p = .03), but a social risk × product replicate interaction was not significant (Wald χ2(2) = 2.83, p > .24). Most importantly, replicating the results of study 4a, a social distance × social risk interaction was marginally significant (Wald χ2(1) = 3.67, p = .055; see figure 7). When experiential gifts carried high social risk, participants were more likely to choose experiential gifts for a close friend rather than for a distant friend (Mclose = 44.1 % vs. Mdistant = 31.6%; Wald χ2(1) = 7.93, p = .005). However, when experiential gifts were associated with low social risk, social distance did not affect the choice of experiential gifts (Mclose = 50.8% vs. Mdistant = 49.9%; Wald χ2(1) = .04, p > .84). The three-way interaction was not significant, which means the social distance × social risk interaction did not vary by product replicate (Wald χ2(2) = 1.27, p > .53; see appendix C for the choice distributions across product replicates). FIGURE 7 View largeDownload slide STUDY 4B: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE FIGURE 7 View largeDownload slide STUDY 4B: MODERATING EFFECT OF SOCIAL RISK ON GIFT PREFERENCE Single-Paper Meta-Analysis, Replication, and Discussion While the simple effects in studies 4a and 4b were significant at traditional levels, we should note that our interactions were both marginally significant at traditional alpha levels. Thus, for reliability we conducted both a single-paper meta-analysis (SPM; McShane and Böckenholt 2017) and a conceptual replication. An SPM of our studies estimates the social distance × social risk interaction at –.11 (95% CI: –.03, –.20). The confidence interval did not contain zero, demonstrating the robustness of our findings. Examining the simple effects also showed robust results. When experiential gifts carried high social risk, the SPM showed that participants were more likely to choose experiential gifts for a close friend rather than for a distant friend (.14, 95% CI: .09, .21); however, when experiential gifts were associated with low social risk, social distance did not affect the choice of experiential gifts (.03, 95% CI: –.02, .09). For our conceptual replication of studies 4a and 4b, we used 10 replicates in a within-subject design similar to studies 1a and 3 (see web appendix C for study details). We again found that social risk moderated the effect of social distance on gift type preference (F(1, 364) = 15.19, p < .001, ηp2= .04). When experiences were high risk, relative preference for experiential gifts (vs. material gifts) was greater for close recipients (Mexperiential = 4.57 vs. Mmaterial = 3.72) than for distant recipients (Mexperiential = 2.04 vs. Mmaterial = 2.80; F(1, 364) = 23.75, p < .001). But when experiences were low risk, there was no relative preference difference for close recipients (Mexperiential = 3.78 vs. Mmaterial = 3.24) compared to distant recipients (Mexperiential = 2.64 vs. Mmaterial = 1.90; F < 1). In sum, we replicated the effect of social distance when the risk associated with experiential gifts was high, and the effect of social distance was mitigated when the risk was low. In sum, studies 4a and 4b provide further support for our theorization by showing the moderating effect of the social risk associated with experiential gifts. Namely, when the social risk associated with experiences was high, participants’ relative preference for experiential gifts over material gifts was greater for close friends than for distant friends (consistent with our previous studies). However, when experiential gifts did not involve such risk, the effect of social distance on relative preference was mitigated. In other words, when experiential gifts carry the higher social risk of giving a gift that poorly matches a recipient’s preference, consumers shy away from giving these experiential gifts to distant recipients whom they do not know well; however, when this social risk is reduced, socially distant recipients are more likely to be given experiential gifts. GENERAL DISCUSSION Despite evidence suggesting that givers should prefer to give experiential gifts because experiences tend to be more individualized and unique, we found that on average gift givers prefer to give material gifts instead of experiences. The current research explored this mismatched preference for material gifts by focusing on the social distance between givers and recipients, a key component in gift exchange. We proposed that gift givers’ preferences shift toward experiential gifts when they are socially close to their recipients compared to socially distant from their recipients. Further, we hypothesized that this effect of social distance on gift type preference is due to givers’ perceived knowledge of the recipients’ individual preferences—knowledge that is necessary for avoiding the social risk of giving a gift that poorly matches recipients’ preferences. We provided evidence for our hypotheses across eight studies. Study 1a compared preferences for various material and experiential gifts and found that relative preferences for experiential gifts were greater when the recipients were socially close compared to socially distant. Study 1b used open-ended responses that allowed participants to generate a gift that they were willing to give. Consistent with the findings of study 1a, we again found that gifts chosen for close recipients were more experiential than those chosen for distant recipients. By manipulating (studies 2a–c) and measuring (study 3) perceived preference knowledge, we tested whether the effect of social distance was driven by a giver’s perceived knowledge of the recipient’s preferences. Studies 2a–c provide converging evidence that participants’ relative preferences for experiential gifts over material gifts were greater when their knowledge of recipient was high compared to low. In these studies, we also ruled out an alternative explanation based on cost and the differential utilitarian-hedonic dimensions of these gifts. Further supporting the preference knowledge process, study 3 showed that the effect of social distance was mediated by givers’ knowledge of their recipients. Additionally, study 3 ruled out joint/shared consumption as an alternative explanation by demonstrating that the mediating effect of preference knowledge remained constant regardless of shared consumption of experiential gifts. Studies 4a and 4b identified a moderator—the social risk of experiences—that further tested our proposed process. While we found the effect of social distance with relatively high-social-risk experiential gifts, this effect was significantly mitigated with low-social-risk experiential gifts. In other words, when experiential gifts were less socially risky to give, the difference in gift preference due to social distance was reduced. In sum, these results suggest that consumers are willing to give experiences, but only when they feel close enough to the recipient to choose appropriate experiential gifts that avoid the social risk of giving experiences that might not match the recipient’s individual preferences. Alternative Explanations and Future Research We should note that, in addition to social distance, there might be other independent factors influencing consumers’ choices of material versus experiential gifts. In the current research, we focused on social distance because it is a key component to the social exchange of gift giving. To choose unique and individualized gifts, consumers must have knowledge about their recipients, which made social distance a natural and critical variable to study. Nonetheless, given other differences between material and experiential gifts, we did address and attempt to rule out several alternative accounts as to why social distance affects preference for material and experiential gifts. For example, experiential gifts can be consumed with others (Caprariello and Reis 2013), and givers might prefer to share an experience only with close recipients, but not with distant recipients. We tested this explanation in study 3 by manipulating whether the experience was shared with others or consumed alone. We found that the effect of social distance replicated regardless of joint consumption, suggesting that shared consumption does not drive our findings. Another possible explanation is that material goods better adhere to social norms or are more memorable because of their tangible and physical nature (Goodman, Malkoc, and Stephenson 2016). We ran a follow-up study to rule out these two alternatives by examining whether socially distant givers were more concerned about social norms and/or giving a physical reminder (and thus less likely to give experiences). The follow-up study used the same procedures as study 1a but with additional measures. At the end of the survey, we asked participants, “When you were evaluating the gifts, how much did you consider the following dimensions?” (1 = not at all, 10 = very much). We collected social norms and physical reminder data using two measures each (social norms: “Whether the gift is socially acceptable,” and “Whether the gift would be appropriate for our friendship”; physical reminder: “Whether the gift is a physical reminder of me or the relationship,” and “Whether the gift physically persists over time.”). As expected, we replicated our previous effect of social distance on gift type preference such that the relative preference to give experiential gifts (vs. material gifts) was significantly greater when the recipient was socially close (Mexperiential = 4.36 vs. Mmaterial = 3.48) compared to socially distant (Mexperiential = 2.65 vs. Mmaterial = 2.86; F(1, 398) = 24.08, p < .001, ηp2= .06). More importantly, in terms of a physical reminder, the alternative explanation did not pan out: we found that socially distant gift givers actually indicated less of a need for a physical reminder (M = 2.38), compared to close givers (M = 3.07; F(1, 398) = 7.79, p = .006, ηp2= .02). In terms of social norms, we did find that participants were more concerned with social norms when they were selecting a gift for a socially distant friend (M = 5.80) compared to a socially close friend (M = 5.18; F(1, 398) = 4.61, p = .03, ηp2= .01); however, a concern for social norms did not predict preference for material versus experiential gifts (p > .6). Thus, neither a need for a physical reminder nor a concern for social norms is a viable alternative explanation for our findings. Implications and Contribution This research has important contributions to both the gift giving and happiness literatures. First, in terms of happiness, the current findings highlight that an experiential gift might not always be a viable option for givers, despite evidence suggesting that experiences contribute more to recipients’ happiness and strengthen relationships between givers and recipients (Chan and Mogilner 2017; Goodman and Lim 2017; Wallman 2015). Given that money spent on others leads to more happiness than money spent on oneself (Dunn, Aknin, and Norton 2008), it seems that the optimal strategy would be to double down—not only spend money on others, but also buy them experiential gifts. However, based on the current findings, buying experiential gifts seems to come with a tradeoff: gift givers believe that choosing an experiential gift requires more knowledge about recipients’ individual preferences, and thus giving an experiential gift carries greater social risk. Of course, gift givers could be wrong and unnecessarily concerned, and distant receivers may still be better off with an experiential gift. Future research may want to investigate the accuracy and consequences of such choices. To give experiences, gift givers believe that they need the proper knowledge of what their recipients like and dislike, and acquiring this knowledge may require a large investment (in time or social resources). If givers do not have such information, they will try to avoid the social risks in giving an experiential gift that does not match their recipient’s preference. However, if this is merely a perception, then it is possible that simply thinking about all the ways that a giver feels more knowledgeable about the recipient could increase the gifting of experiential gifts. Future research needs to investigate other methods to reduce social risks associated with giving experiential gifts. Second, the studies extend the social relevance theories to experiential purchases. Previous literature has highlighted that experiential consumption promotes a sense of social connectedness and therefore brings greater happiness to consumers (Caprariello and Reis 2013; Chan and Mogilner 2017). Adding to this line of research, the current research suggests that social factors can affect people’s experiential purchases. Our research demonstrates that greater social closeness increases the rate at which consumers give experiential gifts, highlighting social closeness as an important antecedent of experiential giving. The present findings, coupled with the recent finding that experiential gifts can increase social connectedness between givers and receivers (Chan and Mogilner 2017), suggest that further research is needed to explore experiential gift giving and consumption. There is a vast literature exploring the nuances of gift giving, but most of this literature has assumed (perhaps indirectly) that gifts are material in nature—that is, they are, and should be, physically given in a box wrapped neatly with a bow. Our research challenges this assumption and identifies situations where experiential gifts may be preferred. The results also suggest that there is an opportunity for firms to provide more experiential gifts, particularly those types of gifts that entail low social risk. Websites that facilitate giving experiences (such as Excitations, Cloud9 Living, and Blueboard) seem to have already noticed the value of experiential giving. Further, they seem to be aware that experiential gifts are still not the norm, as Excitations’ tagline is “Give the unexpected.” Yet giving experiences does not have to be unexpected or a niche business, and these perceptions may change over time. Just as retailers have adapted strategies to be friendlier to giving material gifts—offering exchanges, gift receipts, gift boxes, and gift cards—perhaps marketers of experiential purchases (e.g., Ticketmaster, Travelocity, or the local theater) should be doing the same: finding ways to lower the risk of experiential gifts by offering exchanges, gift cards, and gift receipts. If firms can provide better offerings that lower the social risk of experiential gift giving, thus lowering the knowledge needed to give an experience, then we may find a future where giving experiences becomes mainstream. DATA COLLECTION INFORMATION Two pilot studies in the introduction were conducted on Amazon Mechanical Turk. The first one was run in 2013 and the second in 2017. Studies 1a and 1b were collected at Washington University in St. Louis in 2015 and 2014, respectively. The conceptual replication in the discussion of study 1a was conducted on MTurk in 2015. Study 2a was collected at The Ohio State University in 2017. Studies 2b, 2c, 3, 4a, and 4b were collected on MTurk, along with the conceptual replication of study 4 and the follow-up study in the General Discussion. The authors conducted studies 2b and 2c in 2017, study 3 in 2016, studies 4a and 4b in 2017, the replication of study 4 in 2016, and the follow-up study in 2015. The data were analyzed by the second author with oversight from the first author. The authors wish to thank Selin Malkoc, Kathleen Vohs, Cindy Cryder, the editorial team, and attendees of the OSU Marketing Seminar series, GAP, and CB Research Lab. Supplementary materials are included in the web appendix accompanying the online version of this article. APPENDIX A: STIMULI AND PRETEST RESULTS (STUDY 1A) MEAN RATING OF EXPERIENTIAL AND MATERIAL GIFTS Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* NOTE.—11-point Likert scale (0 = Material, 10 = Experiential); one-sample t-test; *p < .001. Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* Gift type Items Mean SD t Material gifts Handmade pen 1.61 2.43 –13.86* Sunglasses 1.29 2.17 –16.99* Headphones 2.27 2.78 –9.76* Leather gloves 1.21 2.03 –18.56* Wallet/purse 1.20 2.24 –16.86* Experiential gifts Concert ticket 8.91 1.71 22.76* Sightseeing boat ride (available in over 50 cities) 9.22 1.33 31.49* Ticket for comedy night 8.89 1.75 22.12* Wine tour and tasting 9.00 1.43 27.90* Cooking classes (four sessions) 8.68 1.85 19.79* NOTE.—11-point Likert scale (0 = Material, 10 = Experiential); one-sample t-test; *p < .001. APPENDIX B: STIMULI (STUDY 3) Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) View Large Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) Material gifts Nonshared experience Experience shared but not with a giver Experience shared with a giver Leather gloves Massage/spa day pass for one person (you wouldn’t be able to go) Massage/spa day pass with a friend (you wouldn’t be able to go) Massage/spa day pass with the friend (you would go together) Sunglasses Sightseeing boat ride for one person (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with a friend (available in over 50 cities; you wouldn’t be able to go) Sightseeing boat ride with the friend (available in over 50 cities; you would go together) Headphones Wine tour and tasting for one person (you wouldn’t be able to go) Wine tour and tasting with a friend (you wouldn’t be able to go) Wine tour and tasting with the friend (you would go together) Handmade pen Cooking classes (private class with four sessions; you wouldn’t be able to go) Cooking classes with a friend (private class with four sessions; you wouldn’t be able to go) Cooking classes with the friend (private class with four sessions; you would go together) Wallet/purse A ticket for a music concert (you wouldn’t be able to go) Tickets for a music concert with a friend (you wouldn’t be able to go) Tickets for a music concert with the friend (you would go together) View Large APPENDIX C: GIFT CHOICE ACROSS CONDITIONS (STUDIES 4A AND 4B) Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% View Large Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% Study Product replicate Social risk Social distance Gift choice Material Experiential Study 4a Spa domain Low risk Distant 53.5% 46.5% Close 48.9% 51.1% High risk Distant 79.6% 20.4% Close 57.6% 42.4% Glass accessories Low risk Distant 58.6% 41.4% Close 46.8% 53.2% High risk Distant 69.4% 30.6% Close 53.5% 46.5% Study 4b Spa domain Low risk Distant 49.6% 50.4% Close 44.4% 55.6% High risk Distant 72.4% 27.6% Close 52.1% 47.9% Glass accessories Low risk Distant 47.1% 52.9% Close 40.5% 59.5% High risk Distant 56.9% 43.1% Close 45.5% 54.5% Coffee domain Low risk Distant 53.7% 46.3% Close 62.7% 37.3% High risk Distant 75.9% 24.1% Close 70.2% 29.8% View Large APPENDIX D: SOCIAL RISK PRETEST (STUDY 4B) Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) View Large Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) Product categories Low-risk experience Mean (SD) High-risk experience Mean (SD) Spa domain 1.97 (2.14) 2.51 (2.33) Glass accessories 3.69 (2.52) 4.29 (2.77) Coffee domain 1.68 (1.97) 3.59 (2.45) View Large Footnotes 1 We asked 100 Mechanical Turk (MTurk) workers (US only, 95% approval or higher) to list up to five gifts they recently received for the holiday season and their last birthday. After reading definitions of material and experiential gifts, they rated each gift as more material (1) or more experiential (7) on a seven-point scale: 81% of holiday gifts and 66% of birthday gifts were rated below the midpoint of 4. 2 To test this assumption (i.e., whether gift givers perceive experiential gifts as involving greater social risk), 188 MTurk workers (US only, 95% approval rating or higher) listed one material gift and one experiential gift and then rated “Which gift would be more risky to give your friend?” (on a 1–9 scale). Confirming our assumption, participants perceived experiential gifts as carrying higher social risk relative to material gifts (t(187) = 5.93, p < .001; see web appendix A for details). 3 Unexpectedly, we also found a marginal interaction between preference knowledge and gift type (F(1, 116) = 3.64, p = .059, ηp2 = .03) on the hedonic nature of the gift. Differences (experience vs. material) in the hedonic nature were slightly smaller for participants in the high-knowledge condition compared to the low-knowledge condition. 4 One respondent did not answer the price estimation measures, leaving 119 usable responses. 5 We ran three separate models, instead of one large model, due to our relatively modest sample size and potential collinearity between the potential alternative factors. 6 The results of study 2b do not change substantially when we include the two participants who imagined a fake friend rather than thought of a real friend. Results revealed a main effect of preference knowledge (F(1, 300) = 29.29, p < .001, ηp2= .09), a main effect of gift type (F(1, 300) = 17.48, p < .001, ηp2= .06), and the hypothesized interaction between gift type and social distance (F(1, 300) = 4.38, p = .037, ηp2= .01). 7 The results of study 2c do not change substantially when we include the 19 participants that did not follow instructions. The hypothesized effect of preference knowledge remains significant (b = .24, Wald χ2(1) = 5.19, p = .02). 8 A preference knowledge × product type interaction was marginally significant on hedonic rating (F(1, 382) = 3.42, p = .065, ηp2 = .01) as well as on utilitarian rating (F(1, 382) = 4.47, p = .04, ηp2 = .01). Compared to low knowledge, high-knowledge participants showed smaller differences for both hedonic and utilitarian natures. 9 One respondent did not answer both preference knowledge measures, leaving 290 usable responses. 10 When we include the nine participants that did not follow instructions, we found a significant main effect of social distance (Wald χ2(1) = 13.00, p < .001) and of social risk (Wald χ2(1) = 11.15, p < .001). The hypothesized interaction between social distance and social risk was marginally significant and virtually unchanged (Wald χ2(1) = 2.58, p = .108). 11 When we include those 17 participants that did not follow instructions, we found a significant main effect of social distance (Wald χ2(1) = 4.53, p = .03), of social risk (Wald χ2 (1) = 16.24, p < .001), and of product replicate (Wald χ2 (2) = 40.65, p < .001). Additionally, a social distance × product replicate interaction was significant (Wald χ2(2) = 8.66, p = .01). Most importantly, the hypothesized interaction between gift type and social distance was marginally significant and virtually unchanged (Wald χ2(1) = 3.77, p = .052). References Aron Arthur , Aron Elaine N. , Smollan Danny ( 1992 ), “Inclusion of Other in the Self Scale and the Structure of Interpersonal Closeness,” Journal of Personality and Social Psychology , 63 4 , 596 – 612 . Google Scholar CrossRef Search ADS Baskin Ernest , Wakslak Cheryl J. , Trope Yaacov , Novemsky Nathan ( 2014 ), “Why Feasibility Matters More to Gift Receivers than to Givers: A Construal-Level Approach to Gift Giving,” Journal of Consumer Research , 41 1 , 169 – 82 . Google Scholar CrossRef Search ADS Belk Russell W. ( 1996 ), “The Perfect Gift,” in Gift Giving: A Research Anthology , ed. Otnes Cele , Beltramini Richard F. , Bowling Green, OH : Bowling Green University Popular Press , 59 – 84 . Belk Russell W. , Coon Gregory S. ( 1993 ), “Gift Giving as Agapic Love: An Alternative to the Exchange Paradigm Based on Dating Experiences,” Journal of Consumer Research , 20 3 , 393 – 417 . Google Scholar CrossRef Search ADS Berscheid Ellen , Snyder Mark , Omoto Allen M. ( 1989 ), “The Relationship Closeness Inventory: Assessing the Closeness of Interpersonal Relationships,” Journal of Personality and Social Psychology , 57 5 , 792 – 807 . Google Scholar CrossRef Search ADS Camerer Colin ( 1988 ), “Gifts as Economic Signals and Social Symbols,” American Journal of Sociology , 94 , S180 – 214 . Google Scholar CrossRef Search ADS Caprariello Peter A. , Reis Harry T. ( 2013 ), “To Do, to Have, or to Share? Valuing Experiences over Material Possessions Depends on the Involvement of Others,” Journal of Personality and Social Psychology , 104 2 , 199 – 215 . Google Scholar CrossRef Search ADS PubMed Carter Travis J. , Gilovich Thomas ( 2010 ), “The Relative Relativity of Experiential and Material Purchases,” Journal of Personality and Social Psychology , 98 1 , 146 – 59 . Google Scholar CrossRef Search ADS PubMed Carter Travis J. , Gilovich Thomas ( 2012 ), “I Am What I Do, Not What I Have: The Centrality of Experiential Purchases to the Self-Concept,” Journal of Personality and Social Psychology , 102 6 , 1304 – 17 . Google Scholar CrossRef Search ADS PubMed Chan Cindy , Mogilner Cassie ( 2017 ), “Experiential Gifts Foster Stronger Social Relationships than Material Gifts,” Journal of Consumer Research , 43 6 , 913 – 31 . Dai, Hengchen, Cindy Chan, and Cassie Mogilner (2017), “People Rely Less on Consumer Reviews for Experiential Purchases than for Material Purchases,” working paper, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2493536. Dunn Elizabeth W. , Aknin Lara B. , Norton Michael I. ( 2008 ), “Spending Money on Others Promotes Happiness,” Science , 319 5870 , 1687 – 8 . Google Scholar CrossRef Search ADS PubMed Foa Edna , Foa Uriel G. ( 1980 ), “Resource Theory: Interpersonal Behavior as Exchange,” in Social Exchange: Advances in Theory and Research , ed. Gergen Kenneth , Greenberg Martin , Willis Richard , New York : Plenum Press , 215 – 37 . Gino Francesca , Flynn Francis J. ( 2011 ), “Give Them What They Want: The Benefits of Explicitness in Gift Exchange,” Journal of Experimental Social Psychology , 47 5 , 915 – 22 . Google Scholar CrossRef Search ADS Goodman Joseph K. , Lim Sarah ( 2017 ), “Experiential Gifts Lead to More Happiness than Material Gifts,” working paper, Fisher College of Business, The Ohio State University, Columbus, OH 43210. Goodman Joseph K. , Malkoc Selin A. , Stephenson Brittney ( 2016 ), “Celebrate or Commemorate? A Material Purchase Advantage When Honoring Special Life Events,” Journal of the Association for Consumer Research , 1 4 , 497 – 508 . Google Scholar CrossRef Search ADS Hayes Andrew F. ( 2013 ), Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach , New York : Guilford Press . Huang Ming-Hui , Yu Shihti ( 2000 ), “Gifts in Romantic Relationships: A Survival Analysis,” Journal of Consumer Psychology , 9 3 , 179 – 88 . Google Scholar CrossRef Search ADS Hoyer Wayne D. , MacInnis Deborah J. , Pieters Rik ( 2012 ), Consumer Behavior , New York : Cengage Learning . Kelley Harold H. , Berscheid Ellen , Christensen Andrew , Harvey John H. , Huston Ted L. , Levinger George ( 1983 ), “Analyzing Close Relationships,” in Close Relationships , ed. Kelley Harold H. , Berscheid Ellen , Christensen Andrew , Harvey John H. , Huston Ted L. , Levinger George , New York : Freeman , 20 – 67 . Komter Aafke E. ( 1996 ), The Gift. An Interdisciplinary Perspective , Amsterdam : Amsterdam University Press . Mauss Marcel ( 1925 ), The Gift: Forms and Functions of Exchange in Archaic Societies , tr. Cunnison Ian , New York : W. W. Norton . McShane Blakeley B. , Böckenholt Ulf ( 2017 ), “Single Paper Meta-Analysis: Benefits for Study Summary, Theory Testing, and Replicability,” Journal of Consumer Research , 43 6 , 1048 – 63 . New York Magazine ( 2017 ), “Gifts, the Ultimate Catalogue,” http://nymag.com/guides/holidays/gifts/2017/. Nicolao Leonardo , Irwin Julie R. , Goodman Joseph K. ( 2009 ), “Happiness for Sale: Do Experiential Purchases Make Consumers Happier than Material Purchases?” Journal of Consumer Research , 36 2 , 188 – 98 . Google Scholar CrossRef Search ADS Otnes Cele , Lowrey Tina M. , Kim Young Chan ( 1993 ), “Gift Selection for Easy and Difficult Recipients: A Social Roles Interpretation,” Journal of Consumer Research , 20 2 , 229 – 44 . Google Scholar CrossRef Search ADS Parks Malcolm R. , Floyd Kory ( 1996 ), “Meanings for Closeness and Intimacy in Friendship,” Journal of Social and Personal Relationships , 13 1 , 85 – 107 . Google Scholar CrossRef Search ADS Robben Henry S. J. , Verhallen Theo M. M. ( 1994 ), “Behavioral Costs as Determinants of Cost Perception and Preference Formation for Gifts to Receive and Gifts to Give,” Journal of Economic Psychology , 15 2 , 333 – 50 . Google Scholar CrossRef Search ADS Rosenzweig Emily , Gilovich Thomas ( 2012 ), “Buyers Remorse or Missed Opportunity? Differential Regrets for Material and Experiential Purchases,” Journal of Personality and Social Psychology , 102 2 , 215 – 23 . Google Scholar CrossRef Search ADS PubMed Schwartz Barry ( 1967 ), “The Social Psychology of the Gift,” American Journal of Sociology , 73 1 , 1 – 11 . Google Scholar CrossRef Search ADS Steffel Mary , Le Boeuf Robyn A. ( 2014 ), “Overindividuation in Gift Giving: Shopping for Multiple Recipients Leads Givers to Choose Unique but Less Preferred Gifts,” Journal of Consumer Research , 40 6 , 1167 – 80 . Google Scholar CrossRef Search ADS Stinson Linda , Ickes William ( 1992 ), “Empathic Accuracy in the Interactions of Male Friends versus Male Strangers,” Journal of Personality and Social Psychology , 62 5 , 787 – 97 . Google Scholar CrossRef Search ADS PubMed Strahilevitz Michal , Myers John G. ( 1998 ), “Donations to Charity as Purchase Incentives: How Well They Work May Depend on What You Are Trying to Sell,” Journal of Consumer Research , 24 4 , 434 – 46 . Google Scholar CrossRef Search ADS Teigen Karl Halvor , Olsen Marina V. G. , Solås Odd Egil ( 2005 ), “Giver–Receiver Asymmetries in Gift Preferences,” British Journal of Social Psychology , 44 1 , 125 – 44 . Google Scholar CrossRef Search ADS PubMed Tully Stephanie M. , Hershfield Hal E. , Meyvis Tom ( 2015 ), “Seeking Lasting Enjoyment with Limited Money: Financial Constraints Increase Preference for Material Goods over Experiences,” Journal of Consumer Research , 42 1 , 59 – 75 . Google Scholar CrossRef Search ADS Van Boven Leaf , Campbell Margaret C. , Gilovich Thomas ( 2010 ), “Stigmatizing Materialism: On Stereotypes and Impressions of Materialistic and Experiential Pursuits,” Personality and Social Psychology Bulletin , 36 4 , 551 – 63 . Google Scholar CrossRef Search ADS PubMed Van Boven Leaf , Gilovich Thomas ( 2003 ), “To Do or to Have? That Is the Question,” Journal of Personality and Social Psychology , 85 6 , 1193 – 202 . Google Scholar CrossRef Search ADS PubMed Wallman James ( 2015 ), Stuffocation: Living More with Less , London : Penguin Books . Ward Morgan K. , Broniarczyk Susan M. ( 2011 ), “It’s Not Me, It’s You: How Gift Giving Creates Giver Identity Threat as a Function of Social Closeness,” Journal of Consumer Research , 38 1 , 164 – 81 . Google Scholar CrossRef Search ADS Ward Morgan K. , Broniarczyk Susan M. ( 2016 ), “Ask and You Shall (Not) Receive: Relationship Signaling Results in Registry Rejection and Worse Gift Choices,” Journal of Marketing Research , 53 6 , 1001 – 18 . Google Scholar CrossRef Search ADS Wooten David B. ( 2000 ), “Qualitative Steps toward an Expanded Model of Anxiety in Gift Giving,” Journal of Consumer Research , 27 1 , 84 – 95 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Consumer ResearchOxford University Press

Published: Feb 21, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off