Goods Donations Increase Charitable Credit for Low-Warmth Donors

Goods Donations Increase Charitable Credit for Low-Warmth Donors Abstract Low-warmth actors are often assumed to lack communal (or other-oriented) intentions, even when acting generously. Low-warmth donors must therefore send stronger signals of their communal intent when donating to receive the same amount of charitable credit as high-warmth donors. Because goods are linked with communal norms, we find that donating goods allows low-warmth donors to signal communal intent and increase charitable credit received. Study 1 establishes that low-warmth donors receive less credit for unspecified donations than their high-warmth counterparts. Studies 2A and 2B show that goods donations, compared to equally valued monetary or unspecified donations, increase charitable credit for low-warmth donors. Studies 3A and 3B show that donating goods boosts charitable credit for low-warmth donors in particular; high-warmth donors are assumed to have communal intentions, and receive large amounts of credit, regardless of donation type. Finally, study 4 shows that low-warmth donors can increase charitable credit for monetary donations by describing the donation in communal terms—specifically, as a gift. This research has clear practical implications. For example, many corporations are viewed as low warmth, and most corporate donations are monetary, yet companies always have the option to donate goods instead. prosocial behavior, warmth, communal and exchange, donation type, branding, charitable credit, corporate social responsibility In 2012, the devastation of Hurricane Sandy prompted over $380 million in charitable donations (Lawrence et al. 2014). All types of donors gave, often donating in different forms. American Express contributed a $1,000,000 monetary donation. Ikea donated in-kind goods, giving more than 40,000 blankets, pillows, and towels (BCLC 2015). A teacher and her fifth-grade class raised funds to make a monetary donation (Graham 2012). Meanwhile, a hedge fund manager from the Upper West Side donated goods such as canned food and medical supplies (Rampell 2012). While all of these donations provided valuable resources to displaced victims, both the traits of the donors and the types of donations varied substantially. In this research, we propose that such variations in both donor traits and donation type can alter how people award charitable credit to those who give. People who engage in prosocial behavior are often rewarded with higher status and reputation benefits (Berman et al. 2015; Flynn 2003; Flynn et al. 2006); in other words, people grant donors “charitable credit” for their generous acts (Lin-Healy and Small 2012). However, the charitable credit that donors receive does not correspond perfectly with their donation’s impact. Instead, judgments of a donor’s intentions for donating also play a significant role (Barasch et al. 2014; Lin-Healy and Small 2012), and can be even more important than the donation’s actual impact for earning charitable credit (Newman and Cain 2014). Because presumed intentions are a critical component of the credit a donor receives, this research proposes that two factors that influence intention judgments will affect charitable credit as well. First, trait warmth, a central dimension on which people assess others, is closely linked with perceptions of others’ good intentions (Fiske, Cuddy, and Glick 2007). Whereas high-warmth actors are assumed to be well intentioned, previous research suggests that even when low-warmth actors behave prosocially, observers may assume they possess ulterior motives (Cuddy, Glick, and Beninger 2011). Second, donation type has potential to influence intention judgments. Giving goods is often the norm within social or communal relationships (Cheal 1987; Douglas and Isherwood 1979; Heyman and Ariely 2004; Webley, Lea, and Portalska 1983). Accordingly, goods donations (compared to monetary donations) may signal that a donor has donated for communal, or other-focused, reasons. In this research, we predict that donor warmth and donation type will interact to influence the judgments about donors’ communal intentions, ultimately influencing charitable credit received. High-warmth actors are typically assumed to have communal or other-focused intentions (Fiske et al. 2007) and therefore may receive high charitable credit regardless of donation type; any prosocial act reinforces prior beliefs that high-warmth givers act with others’ welfare in mind. By contrast, low-warmth actors are typically assumed to act in accordance with their own interests, and such suspicion surrounding intent may result in lower charitable credit for donating. Therefore, low-warmth donors may need to send additional signals of communal intent to gain equal levels of charitable credit. We show that donations of goods, because of their link with communality, allow low-warmth donors to strengthen their signal of communal intent and earn more credit for their charitable acts. THE LINK BETWEEN TRAIT WARMTH AND COMMUNAL INTENT Warmth refers to the extent to which an individual (or organization; Aaker, Vohs, and Mogilner 2010) is friendly, good-natured, and trustworthy (Fiske et al. 2007), and it is central to the way that people assess one another. People make warmth judgments before judgments of intelligence or competence when forming impressions, and warmth judgments receive greater attention and weight in impression formation than other interpersonal judgments (Wojciszke and Abele 2008; Wojciszke, Bazinska, and Jaworski 1998). Warmth judgments and their consequences also extend beyond individuals; for example, low-warmth companies inspire less customer loyalty than their high-warmth counterparts (Kervyn, Fiske, and Malone 2012). Due to the fundamental importance of warmth in social judgments, warmth has been studied in a wide range of contexts, including romantic partner decisions (Sinclair and Fehr 2005), hiring decisions (Casciaro and Lobo 2008), purchase decisions (Aaker et al. 2010), and customer satisfaction (Andrzejewski and Mooney 2016). Warmth may be particularly relevant to the prosocial domain because people rely on warmth judgments to predict whether or not someone is well intentioned (Fiske et al. 2002). Indeed, judgments of low warmth are linked with competitive and exploitative intentions rather than cooperative and well-meaning intentions (Fiske et al. 2007; Fournier and Alvarez 2012; Kervyn et al. 2012). This association extends to the realm of corporate social responsibility; consumers typically perceive for-profit companies as low in warmth (Aaker et al. 2010) and they tend to be suspicious of companies’ motives for engaging in prosocial behavior (Vlachos et al. 2009). Multiple studies find that the reputational benefits of donating are attenuated, or even reversed, when consumers believe that charitable giving was motivated by profit or was inauthentic in any way (Koschate-Fischer, Stefan, and Hoyer 2012; Wagner, Lutz, and Weitz 2009; Yoon, Gürhan-Canli, and Schwarz 2006). In this research, we hypothesize that low-warmth donors’ intentions are viewed specifically as less communal than high-warmth donors’ intentions. Research in social psychology distinguishes between two categories of social relationships: communal and exchange-based relationships. In communal relationships, benefits are given to others noncontingently and with the recipient’s welfare in mind. Relationships between parents and children or romantic partners typically follow communal norms (Clark and Mills 1979, 2011), such as when parents provide food and shelter for their children without the expectation of being paid back. By contrast, business transactions and most interactions with strangers and acquaintances typically follow exchange norms common in exchange-based relationships; benefits are given to others with the expectation that the giver will receive comparable benefits in return (Clark and Mills 1979, 2011). For example, when selling a car, both the salesperson and the customer give something and expect to receive something of comparable value in return. Although the communal versus exchange distinction was originally conceptualized to describe human relationships (Clark and Mills 1979), consumers also interpret their interactions with companies through the lens of communal or exchange norms, judging some companies to behave more consistently with communal norms and others to behave more consistently with exchange norms (Aggarwal 2004; Aggarwal and Law 2005). We conducted a correlational study as an initial investigation into the relationship between donor warmth and judgments of communal intentions. We asked 96 Mechanical Turk participants about each company on Interbrand’s list of the 10 most valuable global brands from 2016 (Apple, Google, Coca-Cola, Microsoft, Toyota, IBM, Samsung, Amazon, Mercedes-Benz, or General Electric; Interbrand 2017). Participants rated each company on trait warmth (that is, to what extent do the following traits describe the company in general: friendly, well-intentioned, trustworthy, warm, good-natured, sincere; Fiske et al. 2002; α = .96). Next, participants imagined that the company had behaved prosocially in a generalized context (“Imagine that some people needed help and Coca-Cola helped them. How would you interpret this action by Coca-Cola?”). They then rated the company’s intentions on a five-item measure of communal intentions: 1) the company did not expect to receive any benefits from helping, 2) the company helped to respond to others’ needs, 3) the company has a genuine desire to help others, 4) the company helped with hopes of benefiting themselves (reverse-coded), and 5) the company helped in order to get ahead (reverse-coded; α = .80). Results showed that, as predicted, trait warmth and communal intentions were significantly positively correlated (r= .35, p < .001; full details in web appendix A). We propose that such connections between trait warmth and communal, or other-oriented, intentions have the potential to influence the amount of charitable credit that donors receive. Donations that are perceived as communal in nature—that is, motivated by recipients’ needs—tend to be evaluated positively, whereas similar donations that are viewed as self-serving typically are evaluated more negatively (Lin-Healy and Small 2013; Newman and Cain 2014). When judging charitable behavior, people balance the information that the donor has done something good with signals about whether or not the donor truly cared about the recipient’s welfare to arrive at a final judgment of how favorably the good deed should be perceived (Berman et al. 2015). Although people may generally assume that low-warmth donors have self-interested motives, links between goods and communality may allow low-warmth donors who donate goods (compared to money or an unspecified donation) to more effectively signal communal intent, and ultimately, increase charitable credit. GOODS SIGNAL COMMUNAL INTENT The prosocial consumer behavior literature has distinguished between donations of money and time (Liu and Aaker 2008; Macdonnell and White 2015; Reed, Aquino, and Levy 2007); however, there is minimal work investigating goods donations. This gap is notable because donors frequently give in-kind donations of tangible goods such as medical supplies, food, water, and clothing (Charity Navigator 2017). Because of links between giving goods and communality, we propose that donating goods can increase the extent to which low-warmth donors’ contributions are viewed as communally motivated. Giving goods is often the norm within social or communal interactions (Heyman and Ariely 2004), such as in the realm of gift giving (Cheal 1987; Douglas and Isherwood 1979). Although charities themselves tend to prefer monetary donations to other contributions because money is fully fungible and offers flexibility to cover a charity’s most pressing needs (Conan 2011), goods may be more consistent with psychological representations of communality and prosociality. For example, exchanges of goods are considered appropriate within close social relationships, whereas comparable exchanges of money are viewed as taboo within these relationships (Belk and Coon 1993; Webley et al. 1983). The connection between goods and communality is further supported by research about incentives. In some cases, people are less generous when a monetary incentive is offered for generous behavior compared to no incentive at all (Ariely, Bracha, and Meier 2007; Frey and Goette 1999); however, goods incentives, such as T-shirts and lottery tickets, can have neutral or even positive effects on giving (Goette, Stutzer, and Frey 2010; Lacetera, Macis, and Slonim 2013). Research in a similar vein suggests that goods effectively signal communal norms (Gasiorowska et al. 2016; Jiang, Chen, and Wyer 2014; McGraw and Tetlock 2005). When offered a monetary reward, participants report a lower likelihood of helping a friend when the reward is small than when it is large, a pattern consistent with exchange norms whereby effort corresponds with incentive size. When offered a reward of goods, however, participants report an equal (and high) likelihood of helping a friend regardless of reward size, a pattern of noncontingent helping that is consistent with communal norms (Clark and Mills 1979, 2011; Heyman and Ariely 2004). Additional work on gift giving shows that although cash is often more valuable to recipients, gifts of goods can more effectively signal altruistic intent and yield higher esteem for the giver (Ellingsen and Johanneson 2011). Due to differing inferences that arise from goods versus monetary transactions, low-warmth donors may boost signals of communal intent when they donate goods. Further, because goods donations signal positive regard for a recipient (Ellingsen and Johannesson 2011), goods donations may promote communal inferences above and beyond not only monetary donations, but unspecified donations as well. THE PRESENT RESEARCH We predict that low-warmth donors, but not high-warmth donors, will receive greater credit for donating goods compared to money due to differing inferences about communal intent. Because of links between trait warmth and communal intentions (Fiske et al. 2007), and because perceptions of good intentions are critical for gaining credit from charitable giving (Barasch et al. 2014; Newman and Cain 2014), we first hypothesize that trait warmth will play an important role in how much credit a donor receives for giving. We further predict that high-warmth donors may receive high credit for generous acts regardless of the substance of that act; any prosocial act reinforces a prior belief that the high-warmth giver was acting with others’ welfare in mind, making donation type less influential. Low-warmth donors, by contrast, may comparatively struggle to receive credit because people doubt the benevolence of their intent (see figure 1, conceptual model step 1). FIGURE 1 View largeDownload slide CONCEPTUAL MODEL STEP 1: CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION FIGURE 1 View largeDownload slide CONCEPTUAL MODEL STEP 1: CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION Therefore, low-warmth donors may need to send stronger signals of their communal intent compared to their high-warmth counterparts in order to receive similar levels of charitable credit. Due to links between goods and communal norms, we predict that donations of goods (vs. money or unspecified donations) may allow low-warmth donors to strengthen the signal of their communal intent, and increase the amount of credit that they receive (see figure 1, conceptual model step 2). OVERVIEW OF CONSTRUCTS AND STUDIES Throughout our research, we focus on three primary psychological constructs: trait warmth, communal intent, and charitable credit. First, the construct of warmth refers to a character trait regarding how friendly, trustworthy, and good-natured someone is (Fiske 1999). We propose that trait warmth is a key predictor of the influence of donation type on charitable credit. Second, communal (vs. exchange-based) intentions serve as our mediating variable. Judgments of communal, or other-regarding, intentions have been identified as feeding into judgments about generosity (Barasch et al. 2014). Finally, “charitable credit,” or image-related benefits that accrue due to a prosocial act (Lin-Healy and Small 2012; Newman and Cain 2014), serves as our main dependent variable. Charitable credit is conceptualized as the credit received when someone acts benevolently (Lin-Healy and Small 2012). While all of these constructs are related, they also are clearly distinct from one another. Although at times previous work has used notions of warmth and communion interchangeably (Fiske et al. 2002), we delineate important distinctions between them both empirically and conceptually. Warmth refers to a broad and relatively stable trait-level evaluation of a person or organization. Communal intent, the proposed mediating construct, refers to perceptions of the donor’s motives behind a prosocial act, specifically the extent to which the generous act was done with the recipient’s welfare in mind. Charitable credit, our main outcome measure, refers to the amount of credit or esteem a donor receives based on a particular prosocial act (though we note this does not necessarily change global trait perceptions of the donor; Lin-Healy and Small 2012). In the studies where these constructs are measured, we specify in our measurement whether we are asking participants to judge a stable trait (warmth), the motives behind the act (communal intent), or the generosity of the act (charitable credit). We also empirically distinguish these constructs in all studies in which they are measured via confirmatory factor analyses (all reported in web appendix A). Six studies in the main text, as well as eight additional studies reported in the web appendix, test key pieces of our conceptual model. Study 1 establishes the relationship between warmth, communal intent, and the amount of credit a donor receives, showing that consumers grant more charitable credit to high-warmth donors compared to low-warmth donors for unspecified donations, even when they perceive the donations to have equal value. Studies 2A and 2B then demonstrate that low-warmth donors receive more credit for donating goods than equivalent monetary amounts or unspecified donations, and that this is due to differing perceptions of communal intent. Studies 3A and 3B test the full conceptual model from step 2 (figure 1), finding that donor warmth moderates the increased credit for goods donations compared to monetary donations. When low-warmth donors give, they receive more credit for goods donations; when high-warmth donors give, they receive high charitable credit for all donation types. Finally, in study 4, we show one way in which low-warmth donors can increase the charitable credit they receive for monetary donations (which charities typically prefer): low-warmth donors can frame their monetary donations communally—specifically, as a gift. Eight studies in the web appendix replicate these patterns and rule out alternative explanations. STUDY 1: THE IMPACT OF DONOR WARMTH ON CHARITABLE CREDIT Study 1 tests the relationship predicted by step 1 of our conceptual model (figure 1). We predict that low-warmth donors will be judged to have lower communal intent than high-warmth donors, and that this will result in less charitable credit for low-warmth (vs. high-warmth) donors. Previous research finds that for-profit companies tend to be viewed as low warmth (Aaker et al. 2010), and in this study the low-warmth donor is a for-profit company. Nevertheless, judged warmth can vary substantially even among companies (Kervyn et al. 2012). In the high-warmth condition, we therefore describe the same company from the low-warmth condition, but portray the company as particularly high in warmth. Because in this study we are most interested in the general relationship between warmth, communal intent, and charitable credit, we do not specify donation type; however, donation type is a key element that we test in our remaining studies. Pretest Methods For all studies, sample size and exclusion criteria were determined ex ante. Based on predetermined exclusion criteria, we analyzed data only from participants who completed the study and passed an instructional attention check designed to identify inattentive participants (Goodman, Cryder, and Cheema 2013; Oppenheimer, Meyvis, and Davidenko 2009; web appendix B). Following recommendations from Simmons, Nelson, and Simonsohn (2012), we report all data exclusions, all manipulations, and all measures for all studies. We first conducted a pretest in which we recruited 109 Mechanical Turk participants, 91 of whom met our inclusion criteria (MAge = 34.89, 52.75% female); nine participants were removed for failing to complete the study and nine for failing the attention check. In this pretest, we manipulated donor warmth. In a low-warmth condition, participants read “Spades Hardware is a company that sells home improvement goods.” In the high-warmth condition, participants read, “Spades Hardware is a friendly company that sells home improvement goods. Spades Hardware is always warm and welcoming toward visitors.” Participants then rated the company on the six-item perceived warmth scale (see correlational study in the introduction) with items presented in a randomized order. Pretest Results The pretest descriptions successfully manipulated donor warmth. In the high-warmth condition, Spades Hardware was rated as warmer (MHigh-Warmth = 5.98, SD = .81) than in the low-warmth condition (MLow-Warmth = 4.53, SD = 1.36, t(89) = 6.14, p < .001, d = 1.30). Main-Study Methods We recruited 320 Mechanical Turk participants, 274 of whom met our inclusion criteria (MAge = 33.53, 52.31% female); 28 participants were removed for failing to complete the study and 18 for failing the attention check. Using the same descriptions of the low- and high-warmth companies from our pretest, participants in both conditions then read about an unspecified donation, specifically: “This past weekend Spades Hardware made a donation.” In both conditions, participants indicated the charitable credit that they would award to the donating company. Specifically, participants were asked, “How favorably do you view Spades Hardware on the characteristics below as a result of their donation?” Participants rated the extent to which they viewed the company as generous, helpful, and charitable as a result of their donation. Participants also rated how beneficial they believed the company is and to what extent the company makes the world a better place as a result of their donation (1 = Not at all, 7 = Very much so; all items adapted from Lin-Healy and Small 2012; Newman and Cain 2014; see web appendix B for all measures). Although we originally anticipated that the measures of charitable descriptors (generous, helpful, and charitable; Lin-Healy and Small 2012) and charitable benefit (how beneficial they believed the donation was and to what extent the donor made the world a better place; Newman and Cain 2014) would assess distinct constructs, responses to all items loaded onto a single factor that was highly reliable (α = .86). Therefore, we combined all five items to create a single and comprehensive “charitable credit” measure, which we use throughout the article. To assess mediation of communal intent, we asked participants to rate the donor on the extent to which it signaled communal versus exchange-based intentions using the five-item scale of communal intent described in the correlational study from this article’s introduction (α = .80, see also web appendix B). As a manipulation check, participants also rated the donor on trait warmth using the six-item scale of warmth described in the introduction’s correlational study. Importantly, perceptions of both competence and wealth have been shown to correlate negatively with warmth judgments (Cuddy, Fiske, and Glick 2008; Fiske et al. 2002), and we therefore measured both to ensure that they did not account for any findings. We measured competence judgments using Fiske et al.’s (2002) scale, which included the following items: competent, confident, capable, efficient, intelligent, and skillful (α = .93). We measured perceived wealth by asking, “How wealthy do you think Spades Hardware is?” (1 = Not at all wealthy, 7 = Extremely wealthy). We also measured purchase likelihood to gauge correspondence between charitable credit and consumer choices. Participants read: “Please rate how likely you would be to go to Spades Hardware next time you need home improvement goods” (1 = Not at all likely, 7 = Very likely). The order of all dependent measures was counterbalanced. Finally, as a follow-up measure, we asked participants “How much would you estimate Spades Hardware’s donation was worth?” to ensure that differences in donation value estimation did not account for any findings. Participants could enter any value that they wished. Please see web appendix A for factor analysis results for measures in all studies. Main-Study Results Warmth Manipulation Check The manipulation check indicated that our manipulation was successful; the high-warmth company was rated as warmer (MHigh-Warmth = 5.79, SD = .87) than the low-warmth company (MLow-warmth = 4.99, SD = 1.19; t(272) = 6.01, p < .001, d = .73). Charitable Credit Participants awarded the high-warmth company more charitable credit (MHigh-Warmth = 5.54, SD = .84) than the low-warmth company (MLow-warmth = 5.21, SD = 1.07; t(272) = 2.80, p = .005, d = .32). Communal Intent Participants judged the high-warmth company to have greater communal intent (MHigh-Warmth = 4.71, SD = 1.02) than the low-warmth company (MLow-warmth = 4.24, SD = 1.09; t(272) = 3.66, p < .001, d = .44). Purchase Likelihood Participants reported greater purchase likelihood for the high-warmth company (MHigh-warmth = 5.72, SD = 1.04) than for the low-warmth company (MLow-warmth = 5.10, SD = 1.35; t(268) = 4.21, p < .001, d = .51). Estimated Donation Value Using free response, participants estimated that the donation was worth a median value of $2,000 in both conditions (a Mann-Whitney test indicated that there was a nonsignificant difference between the two conditions; U = 9288.50, Z = –.04, p = .97). This implies that the donor warmth manipulation did not influence the perceived value of the donation. Mediation The mediation analysis showed that perceptions of communal intent mediated the effect of manipulated donor warmth on charitable credit. Using methods prescribed by Hayes (2013, model 4) we tested the significance of communal (vs. exchange) intent as the mediator by calculating standardized indirect effects for 5,000 bootstrapped samples and found that communal intent mediates the effect of donor warmth on charitable credit (indirect effect = .25; 95% CI [.11, .39]; direct effect = .08; 95% CI [–.11, .27]). This pattern remains significant when we control for competence and perceived wealth (indirect effect = .12; 95% CI [.02, .23]; direct effect = .02; 95% CI [–.15, .19]; figure 2). We also find that communal intent partially mediated the effect of donor warmth on purchase likelihood (web appendix A). FIGURE 2 View largeDownload slide THE RELATIONSHIP BETWEEN DONOR WARMTH AND CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION, AS MEDIATED BY COMMUNAL INTENT FIGURE 2 View largeDownload slide THE RELATIONSHIP BETWEEN DONOR WARMTH AND CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION, AS MEDIATED BY COMMUNAL INTENT Discussion Consistent with step 1 of our conceptual model, this study finds that people ascribe charitable credit differently based on who a donor is. A high-warmth company received greater charitable credit for a donation than a low-warmth company because donations from high-warmth donors are viewed as more communal. These results set up our following studies, which test step 2 of our conceptual model (figure 1). We first test whether goods donations increase the amount of credit that a low-warmth donor receives, and we then test how donor warmth interacts with donation type to influence the charitable credit a donor receives. STUDY 2A: CORPORATE DONATIONS OF GOODS RECEIVE MORE CREDIT Study 2A begins to test step 2 of our conceptual model (figure 1) by testing whether consumers judge a low-warmth donor more favorably for a goods donation than for a monetary donation. We also include a control condition where the donation type is not specified. We use the same description of a for-profit company from the low-warmth condition in study 1 to operationalize a low-warmth donor in this study. Methods We recruited 450 Mechanical Turk participants, 406 of whom met our inclusion criteria (MAge = 34.51, 40.25% female); 30 participants were removed for failing to complete the study and 14 for failing the attention check. Participants read the same description of the donor from the low-warmth condition in study 1: “Spades Hardware is a large corporation that sells home improvement goods.” Participants were assigned to one of three conditions: 1) control condition: “This past weekend Spades Hardware donated to a food bank,” 2) monetary donation condition: “This past weekend Spades Hardware donated $2,000 to a food bank,” or 3) goods donation condition: “This past weekend Spades Hardware donated boxes of canned food to a food bank. (The donation cost the company $2,000 and it would have cost the food bank the same amount to obtain those goods.)” We did not include a donation value for the control condition, due to a concern that mentioning the value would lead participants to assume the company made a monetary donation. However, we described the donations in the other conditions to be worth $2,000 because that was the median estimated value of the unspecified donations in study 1. We included information about the cost for the company and value of the donation for the food bank in the goods condition to ensure that participants would not assume the actual value for the charity was greater than $2,000 when the company donated goods. In all conditions, participants rated the company on the five-item charitable credit scale (α = .90). As in study 1, we also measured purchase likelihood and asked participants in the control condition, “How much would you estimate Spades Hardware’s donation was worth?” Participants could enter any value that they wished. Finally, we asked all participants “Who would get more canned food for $2,000?” 1) Spades Hardware, 2) the food bank, 3) they can get the same amount. This question was intended to confirm that participants did not assume that the donor could provide greater value by donating goods rather than money. (We note that in this initial test of the preference for goods donations from low-warmth donors, we did not measure communal intent, but do so in studies 2B–4). Results Charitable Credit Planned comparisons showed that participants awarded the company more charitable credit for donating goods (MGoods = 5.84, SD = .80) than money (MMoney = 5.48, SD = .92; t(270) = 3.48, p = .001, d = .42), or for making an unspecified donation (control condition); MControl = 5.52, SD = .92; t(273) = 3.12, p = .002, d = .38). There was no difference in charitable credit between the monetary and unspecified (control) donation condition (t(263) = .31, p = .75; figure 3). FIGURE 3 View largeDownload slide STUDY 2A: CHARITABLE CREDIT AS A FUNCTION OF DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 3 View largeDownload slide STUDY 2A: CHARITABLE CREDIT AS A FUNCTION OF DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Estimated Donation Value Using free response, participants in the control condition estimated that the donation was worth a median value of $1,000. This is lower than the value we stated in the other two conditions, so as an additional test, we looked at only participants who gave estimates in the top 50% (median = $5,000). We found that this group still rated the donating company as marginally less charitable (MControl = 5.62, SD = .86) than those in the goods donation condition (MGoods = 5.84, SD = .80; t(206) = 1.86, p =.065). This suggests that lower assumed donation values in the control (unspecified donation) condition do not fully account for the low levels of charitable credit in that condition. Purchase Likelihood Using planned comparisons, we found that participants indicated a marginally higher likelihood of purchasing from Spades Hardware when it donated goods (MGoods = 5.58, SD = .99) compared to money (MMoney = 5.37, SD = .95; t(268)=1.76, p = .079). There was a nonsignificant difference in purchase likelihood between the goods donation and the control (unspecified donation) conditions (MControl = 5.46, SD = .93; t(271) = .90, p = .37) and between the monetary and control conditions (t(263) = .77, p = .44). Donation Value Inferences In response to the question “Who would be able to get the most canned food for $2,000?” 45.81% of participants responded that the food bank would be able to buy more canned food with $2,000; 48.28% believed that the food bank and hardware store could acquire the same amount of canned food for that amount of cash. Only 5.91% of participants believed that Spades Hardware would get the most bang for their buck. These patterns indicate that participants did not rate charitable credit higher for goods donations because they thought that the donor was able to acquire, and thus donate, more canned food than the charity could buy for the same amount. In fact, participants most frequently indicated that the charity could buy as much or more canned food as the donor, a pattern commonly true in the real world because nonprofits often receive discounts on such goods (Conan 2011; White 2015). Discussion Consistent with our conceptual model, we observe a significant increase in the charitable credit that a low-warmth donor receives when donating goods versus money. Further, we observe that the boost in charitable credit for goods donations not only occurs in comparison to a monetary donation, but also in comparison to an unspecified donation. An additional study (web appendix C, study 1) found that while a goods donation from a low-warmth donor garnered more charitable credit than a monetary donation, both donations received more credit than a no-donation condition. These results suggest that, for low-warmth donors, donating goods results in a unique boost in charitable credit compared to other donation types, but giving money also increases charitable credit compared to making no donation at all. In study 2B, we test whether judgments about the communal nature of a goods donation is responsible for greater charitable credit received by low-warmth donors who give goods. In this study, the vast majority of participants believed the food bank could get the same amount if not more food than the hardware store for the amount of money given, and yet, we still find the company receives greater credit for giving goods rather than their equivalent cash value. As an especially conservative test of the image benefits of goods donations (web appendix C, study 2) we compared three donations in a between-subjects experiment: 1) a monetary donation of $1,000, 2) an equivalent goods donation ($1,000 worth of canned food), and 3) a smaller goods donation ($900 worth of canned food). We found that both goods donations garnered higher charitable credit (and purchase likelihood) than the monetary donation, and the two goods donations were perceived as equally charitable. These results showed that the charitable credit benefits of goods donations still emerge even when the goods donation has a lower value than the monetary donation. We conclude that the benefits of goods donations for low-warmth donors cannot be explained by higher interpreted financial value of goods donations. Moving forward, in all studies, we continue to explicitly state a donation value in the goods condition that matches the donation amount in the monetary donation condition. Even though mentioning a monetary value may dampen communality effects for goods transactions (Heyman and Ariely 2004), it is essential for us to hold donation value constant across conditions to ensure greater donation value is not inferred when donors give goods. Please also see web appendix C, study 3, for a description of an additional study that replicates these findings and compares a monetary donation to two goods donations, one with and one without a monetary value. While both goods donations were given more charitable credit than the monetary donation, this study finds that the effect is indeed stronger when the monetary value of a goods donation is absent. STUDY 2B: COMMUNAL (VS. EXCHANGE) INTENTIONS AS MEDIATOR Study 2B further explores consumer preferences for goods donations by low-warmth donors by testing mediating factors. We test whether goods versus monetary donations trigger inferences of communal versus exchange-based intentions, resulting in different levels of charitable credit. We additionally test two other potential mediators behind the preference for corporate goods donations: perceptions of effort and sacrifice. Consumers may reasonably infer that companies exert more effort, more sacrifice, or both when donating tangible goods compared to simply writing a check to a charity (goods need to be chosen, procured, delivered, etc.). We therefore also measure and test perceived effort and sacrifice as potential contributors to the greater credit for goods donations by low-warmth donors. Methods For study 2B, we recruited 240 Mechanical Turk participants, 212 of whom met our inclusion criteria (MAge = 32.93, 53.7% female); 17 participants were removed for failing to complete the study and 11 for failing the attention check. Participants were randomly assigned to either a monetary donation or goods donation condition and read the following scenario: “Spades Hardware is a large corporation that sells home improvement goods. This past weekend, Spades Hardware made a donation of $1,000,000 ($1,000,000 worth of medical supplies) to humanitarian aid efforts.” Participants next rated Spades Hardware on the charitable credit items (α = .93). To measure mediating factors, participants also rated Spades Hardware on the five-item scale of communal intent described in the correlational study in this article’s introduction and study 1 (α = .85). We additionally measured perceived effort and sacrifice related to the donation. Participants were asked to what extent they agree with the following statements: “Spades Hardware put a lot of effort into this donation,” “Spades Hardware worked hard on this donation,” and “Spades Hardware put thought into this donation” (Bechwati and Xia 2003). We evaluated perceived sacrifice by asking the extent to which participants agreed that “Spades Hardware sacrificed when making this donation” as well as asking “How big was Spades Hardware’s sacrifice when making this donation?” While we planned to test effort and sacrifice as separate constructs, these two potential mediators loaded onto a single factor that was highly reliable (α = .92), and therefore we combined items to create a single effort/sacrifice measure. Finally, we measured purchase likelihood by asking participants, “How likely would you be to go to Spades Hardware next time you need home improvement goods?” (1 = Not at all likely, 7 = Very likely). The order of all dependent measures was counterbalanced. Results Charitable Credit In line with study 2A, participants awarded the company more charitable credit when the company donated goods rather than money (MGoods = 5.95, SD = .84 vs. MMoney = 5.65, SD = .90; t(210) = 2.55, p = .01, d = .35). Purchase Likelihood Participants reported a marginally higher likelihood of purchasing from the company following a goods, rather than monetary, donation (MGoods = 5.49, SD = 1.13; MMoney = 5.23, SD = 1.12; t(208) = 1.70, p = .09, d = .24). Communal Intent Participants believed the company signaled communal (vs. exchange) intentions to a greater extent when it donated goods (in this case, medical supplies; MGoods = 4.66, SD = 1.03) rather than money (MMoney = 4.34, SD = 1.23; t(210) = 2.07, p = .04, d = .29). Effort/Sacrifice There was a nonsignificant difference in perceived effort and sacrifice for donations of goods versus money (MGoods = 5.04, SD = 1.36 vs. MMoney = 4.92, SD = 1.43; t(210) = .64, p = .52). Mediation The mediation analysis showed that perceptions of communal intent mediated the effect of donation type on charitable credit. Using methods prescribed by Hayes (2013, model 4) we simultaneously tested the significance of both mediators by calculating standardized indirect effects for 5,000 bootstrapped samples and found that communal intent mediates the effect of donation type on charitable credit (total indirect effect = .19; 95% CI [.02, .36]; direct effect = .14; 95% CI [–.04, .34]). We found a statistically significant indirect effect of perceived communal intent (.15; 95% CI [.01, .30]). The indirect effect of effort and sacrifice was not significant (.04; 95% CI [–.09, .17]; figure 4). FIGURE 4 View largeDownload slide THE RELATIONSHIP BETWEEN DONATION TYPE AND CHARITABLE CREDIT FOR A LOW-WARMTH DONOR AS MEDIATED BY COMMUNAL INTENT FIGURE 4 View largeDownload slide THE RELATIONSHIP BETWEEN DONATION TYPE AND CHARITABLE CREDIT FOR A LOW-WARMTH DONOR AS MEDIATED BY COMMUNAL INTENT Discussion Results from study 2B show that differences in communal intent perceptions based on goods versus monetary donations explain the effect of donation type on charitable credit for low-warmth donors. These results suggest that consumers judge that low-warmth donors who donate goods have communal intentions for doing so, whereas low-warmth donors who donate money have relatively exchange-based intentions for doing so; this communal versus exchange signaling in turn influences the charitable credit received. Perceived effort and sacrifice, although almost certainly important in assessments of many donations, did not significantly differ due to donation type in this scenario. As a robustness check, we replicated these findings in an additional study (web appendix C, study 4). Participants imagined that a natural disaster impacted a neighboring town and Spades Hardware donated to the relief efforts. Participants read that the store donated either 1) $1,000, 2) $1,000 worth of food, or 3) $1,000 worth of lumber to the cause; the latter two conditions were both included to test whether the “fit” between the donation and the company matters (for Spades Hardware, lumber was pretested as a substantially “higher-fit” donation than food). We tested both donations to determine whether consumers believe the cause may receive greater benefit (more goods) when the donation is relevant to the company (in this case, lumber from the hardware store) rather than irrelevant. Again, companies that donated goods (vs. money) received greater charitable credit. There was a nonsignificant difference in charitable credit for the low-fit and high-fit donations, which leads us to believe that consumers are not incorporating potential increases in benefit to the cause when companies donate wholesale goods. In this study, both perceived communal intent and sacrifice/effort had significant and separate mediating effects. Results from study 2B demonstrate support for the mediating role of communal intent for this pattern; in the follow-up study, effort and sacrifice simultaneously mediated the effect of donation type on charitable credit alongside communal intent. Taken together, we interpret these mediation results to mean that increased perceptions of effort and sacrifice can contribute to increased charitable credit for goods donations; however, even when goods donations do not increase perceptions of effort and sacrifice (as in the main study 2B), judgments of communal intent can drive the overall effect. In addition to providing insight about process, these studies also provide evidence about the robustness of the preference for goods donations by low-warmth donors by testing two new contexts (humanitarian aid efforts and a sudden disaster) and two new donations (medical supplies and lumber). We conclude that the increased credit given to low-warmth donors for donations of goods is not limited to a particular type of cause or specific type of goods donation. STUDIES 3A AND 3B: THE INTERACTION OF DONOR WARMTH AND DONATION TYPE Studies 3A and 3B were designed to test the full conceptual model (figure 1, step 2), including the moderating role of warmth in the preference for goods donations. In an early study for this project (web appendix C, study 5), we measured perceptions of charitable credit for a donation of either money or goods by two types of donors: a low-warmth for-profit company (Spades Hardware) and a high-warmth family (the Joneses). We observed a significant interaction between donor warmth and donation type; the (low-warmth) company received more credit for a donation of goods than a monetary donation, while the (high-warmth) family received more credit for a donation of money. Although these results are consistent with the conceptual model, there are many differences between companies and families besides perceived warmth. In studies 3A and 3B, we build on this initial result and more precisely manipulate warmth. STUDY 3A: THE MODERATING ROLE OF WARMTH—INDIVIDUAL DONORS Thus far we have relied on for-profit companies as the operationalization of a low-warmth donor. In study 3A, we describe individual donors, and manipulate whether they are high or low warmth. Past research shows that certain groups of individuals viewed as subordinate and noncompetitive—for example, elderly individuals and homemakers—tend to be viewed as high warmth (Cuddy, Fiske, and Glick 2004). In contrast, those who are high in status and/or competitive—for example, educated individuals—tend to be viewed as low warmth (Fiske et al. 2002). Previous research examines warmth and competence perceptions of individuals in a variety of professions and finds that people’s perceived warmth varies significantly based on their profession (Imhoff et al. 2013). Building on this work, we operationalized a high-warmth donor as a nursery school teacher and a low-warmth donor as a corporate manager in study 3A. We hypothesized that, consistent with our conceptual model, low-warmth donors will receive more charitable credit for goods donations due to increased perceptions of communal intent, whereas high-warmth donors will receive high charitable credit regardless of donation type (because of high communal intent inferred across donation type). Pretest Methods We recruited 110 Mechanical Turk participants, 105 of whom met our inclusion criteria (MAge = 35.63, 50% female); three participants were removed for failing to complete the study and two for failing the attention check. In this pretest, we manipulated donor warmth. In a high-warmth condition, participants read “Heather is a 41-year-old nursery school teacher.” In the low-warmth condition, participants read, “Heather is a 41-year-old corporate manager.” Participants were then asked to rate the target on the six-item perceived warmth scale (α = .96) with items presented in a randomized order. Pretest Results The descriptions in the pretest successfully manipulated donor warmth. Heather the nursery school teacher was rated as significantly warmer (MHigh-Warmth = 5.70, SD = .90) than Heather the corporate manager (MLow-Warmth = 4.30, SD = 1.09, t(103) = 7.15, p < .001, d = 1.40). Main-Study Methods We recruited 675 Mechanical Turk participants, 599 of whom met our inclusion criteria (MAge =35.75, 60.77% female); 39 participants were removed for failing to complete the study and 37 for failing the attention check. As in the pretest, participants in the high-warmth [low-warmth] condition read the following donor descriptions: “Heather is a 41-year-old nursery school teacher [corporate manager].” Participants were then informed that this past weekend, Heather contributed either a monetary donation (“$100”) or a goods donation (“$100 worth of canned food”) to her local food bank. We then asked participants to respond to the items that measured charitable credit (α = .89), communal intent (α = .82), perceived effort/sacrifice (α = .88), and trait warmth (α = .92, a manipulation check). In addition, due to potential differences in perceived competence and wealth for high-warmth and low-warmth donors, we measured these factors to ensure that they did not account for any findings. The order of all dependent measures was counterbalanced. Main-Study Results Warmth Manipulation Check The high-warmth individual was perceived as significantly warmer (MHigh-Warmth = 6.01, SD = .88) than the low-warmth individual (MLow-Warmth = 5.58, SD = .93; t(596) = 5.78, p < .001, d = .23). Based on a 2 × 2 ANOVA, we found a main effect of the warmth manipulation on perceived warmth (F(1, 598) = 33.63, p < .001); donation type did not exert a significant main effect on warmth (F(1, 598) = .11, p = .74). We observed a marginally significant interaction between donation type and manipulated donor warmth on measured warmth (F(1, 598) = 3.67, p = .06; please also see a similar analysis in study 3B). Charitable Credit We observed a significant main effect of donor type on charitable credit; the high-warmth individual received more charitable credit for her donation than the low-warmth individual (F(1, 598) = 41.70, p < .001, η2p= .07). Donation type also had a significant main effect; donations of goods received more credit than donations of money (F(1, 598) = 5.78, p = .017, η2p= .01). Most importantly, the 2 × 2 ANOVA revealed a significant interaction between donation type (goods/money) and donor warmth (low warmth/high warmth; F(1, 598) = 5.29, p = .022, η2p= .01). Specifically, the low-warmth donor received more charitable credit for a goods donation than a monetary donation (MGoods = 5.75, SD = 1.00 vs. MMoney = 5.37, SD = 1.24; t(304) = –2.96, p = .003 , d = .17). The high-warmth donor, however, received equal (and high) credit for both goods and monetary donations (MGoods = 6.09, SD = .98 vs. MMoney = 6.08, SD = .85; t(291) = –.09, p = .93; figure 5). FIGURE 5 View largeDownload slide STUDY 3A: CHARITABLE CREDIT FOR INDIVIDUAL DONORS AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 5 View largeDownload slide STUDY 3A: CHARITABLE CREDIT FOR INDIVIDUAL DONORS AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Communal Intent The donor warmth manipulation had a significant main effect on communal intent; the high-warmth donor’s intent was viewed as more communal (F(1, 598) = 32.35, p < .001, η2p= .05). Donation type had a nonsignificant main effect (F(1, 598) = 2.50, p = .11). A 2 × 2 ANOVA revealed a marginally significant interaction between donation type (goods/money) and donor warmth (low warmth/high warmth; F(1, 598) = 3.39, p = .066, η2p= .01) on communal intent. For the low-warmth donor, a goods donation produced significantly higher perceptions of communal intent than a monetary donation (MGoods = 5.70, SD = .98 vs. MMoney = 5.42, SD = 1.25; t(304) = –2.20 , p = .027, d = .12). There was no difference in perceptions of communal intent by donation type for the high-warmth donor (MGoods = 6.05, SD = .90 vs. MMoney = 6.02, SD = .90; t(291) = .21, p = .84). Effort/Sacrifice The donor warmth manipulation had a significant main effect on perceived effort and sacrifice; the high-warmth individual was seen as putting in more effort than the low-warmth individual (F(1, 598) = 77.09, p < .001, η2p= .12). Donation type, however, showed no main effect (F(1, 598) = .26, p = .61). There was a nonsignificant interaction between donation type and donor warmth on effort/sacrifice (F(1, 598) = 1.97, p = .16). There was no significant difference in perceptions of effort and sacrifice by donation type for the low-warmth individual (MGoods = 4.43, SD = 1.29 vs. MMoney = 4.25, SD = 1.23; t(304) = –1.26, p = .21), nor was there a significant difference in effort perceptions for the high-warmth individual (MGoods =5.13, SD = 1.10 vs. MMoney = 5.22, SD = 1.00; t(291) = .70, p = .49). Moderated Mediation We conducted a moderated mediation analysis (Hayes 2013, model 8) to test the predicted relationship of donation type by low- or high-warmth donors on charitable credit, simultaneously testing both communal intent and effort/sacrifice as mediators. We tested the significance of both mediators by calculating standardized indirect effects for 5,000 bootstrapped samples and found that the model mediates the effect of donation type on charitable credit (direct effect = .07; 95% CI [–.01, .26]). More specifically, we found that donation type produced an indirect effect of communal intent on charitable credit that was conditional on individual warmth, but found no indirect effect of effort/sacrifice. As hypothesized, inferences about communal intent mediated the effect of donation type on charitable credit for the low-warmth individual (indirect effect = .12; 95% CI [.02, .26]), but not for the high-warmth individual (indirect effect = –.01; 95% CI [–.10, .08]). Effort and sacrifice did not mediate the effect for the low-warmth individual (indirect effect = .05; 95% CI [–.03, .14]), nor for the high-warmth individual (indirect effect = .02; 95% CI [–.09, .04]). The index of moderated mediation was not significant at the 95% level of confidence. However, at the 90% level of confidence, the index of moderated mediation was significant for communal intent (index = –.13; 90% CI [–.27, –.02]) but not for effort/sacrifice (index = –.07; 90% CI [–.16, .01]). We find a similar and significant pattern of results when controlling for both competence and perceived wealth (web appendix A). Please see table 1 for descriptive results for all dependent variables measured in studies 3A and 3B. Table 1 Study 3A and 3B Results Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      † p < .10, *p < .05, **p < .01; these significance notations refer to differences in mean evaluations for monetary donations compared to goods donations with standard deviations in parentheses. A° symbol next to the variable name indicates that there is a significant interaction between donor warmth and donation type on this variable at a p < .05 level. Table 1 Study 3A and 3B Results Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      † p < .10, *p < .05, **p < .01; these significance notations refer to differences in mean evaluations for monetary donations compared to goods donations with standard deviations in parentheses. A° symbol next to the variable name indicates that there is a significant interaction between donor warmth and donation type on this variable at a p < .05 level. Discussion Results from study 3A replicate the pattern of preferences from studies 2A and 2B for donations of goods, but only when the donor is perceived as low warmth. In contrast to low-warmth donors who receive more credit for goods donations, high-warmth donors receive the same amount of (high) charitable credit regardless of donation type. This pattern provides insight into the role of warmth in the preference for donations of goods (vs. money); when perceived warmth is high, communal intentions are assumed and are less likely to fluctuate based on the donation’s substance. When perceived warmth is low, however, judgments of communal intent are more sensitive to donation type; specifically, low-warmth donors receive a boost in charitable credit for goods donations. STUDY 3B: THE MODERATING ROLE OF WARMTH—CORPORATE DONORS In study 3B, we aim to replicate the patterns from study 3A when studying corporate donors and using a new, more direct, manipulation of warmth. Pretest Methods We first conducted a pretest in which we recruited 120 Mechanical Turk participants, 102 of whom met our inclusion criteria (MAge = 34.96, 56% female); 13 participants were removed for failing to complete the study and five for failing the attention check. In this pretest, we manipulated company warmth using a manipulation similar to the one used in study 1. In a high-warmth condition, participants read “Spades Hardware is a small, friendly hardware store that is always competent and also warm and welcoming toward visitors.” In the low-warmth condition, participants read, “Spades Hardware is a small, corporate hardware store that is always competent though also cold and indifferent toward visitors.” Participants then rated the company on the six items from the perceived warmth scale from previous studies, presented in a randomized order. Pretest Results The descriptions used in the pretest successfully manipulated donor warmth. The high-warmth donor was rated as warmer (MHigh-Warmth = 6.07, SD = .94) than the low-warmth donor (MLow-Warmth = 3.42, SD = 1.46, t(99) = 10.69, p < .001, d = 2.15). Main-Study Methods For study 3B, we recruited 615 Mechanical Turk participants, 565 of whom met our inclusion criteria (MAge = 35.14, 58% female); 34 participants were removed for failing to complete the study and 16 for failing the attention check. The study used a 2 (donor warmth: high, low) × 2 (donation type: $1,000, canned food worth $1,000) between-subjects experimental design. For the low-warmth and high-warmth conditions, Spades Hardware was described in the same way as in the pretest. Then, participants were told that the company donated either “$1,000” or “canned food worth $1,000” to their local food bank. We then asked participants to respond to the items that measured charitable credit (α = .93), communal intent (α = .88), perceived effort/sacrifice (α = .92), and trait warmth (α = .96, a manipulation check). Main-Study Results Warmth Manipulation Check The high-warmth company was perceived as warmer (MHigh-Warmth = 6.02, SD = .99) than the low-warmth company (MLow-Warmth = 3.99, SD = 1.39; t(563) = 20.06, p < .001, d = 1.69). When testing a 2 × 2 ANOVA, we observed a main effect of the warmth manipulation on perceived warmth (F(1, 561) = 402.33, p < .001, η2p= .42). Looking at additional patterns, we did not observe a main effect of donation type on measured warmth (F(1, 561) = 1.82, p = .18), nor did we observe an interaction between donation type and manipulated donor warmth on measured warmth (F(1, 561) = .06, p = .80). These patterns suggest that the warmth construct is a relatively stable trait that is minimally influenced by donation type from any one donation, exhibiting distinct patterns compared to the other constructs under study, such as communal intent and charitable credit. Charitable Credit The warmth manipulation had a significant main effect; the high-warmth company received more charitable credit (F(1, 563) = 102.2 , p < .001, η2p= .15). Donation type did not show a significant main effect (F(1, 561) = .66 , p = .42). Most importantly, the 2 × 2 ANOVA revealed a significant interaction between donation type (money/goods) and donor image (low warmth/high warmth; F(1, 561) = 4.65, p = .03, η2p= .01). In line with the previous findings, when the company was described as low warmth, it received a marginally significant increase in charitable credit for a goods donation compared to a monetary donation (MGoods = 5.05, SD = 1.23; MMoney = 4.77, SD = 1.21; t(274) = 1.92, p = .056, d = .23). However, when the company was described as high warmth, there was no significant difference in charitable credit between donation types (MGoods = 5.77, SD = 1.00 vs. MMoney = 5.89, SD = .86; t(287) = –1.08, p = .28; figure 6). FIGURE 6 View largeDownload slide EXPERIMENT 3B: CHARITABLE CREDIT AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 6 View largeDownload slide EXPERIMENT 3B: CHARITABLE CREDIT AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Communal Intent The donor manipulation had a significant main effect on communal intent; the high-warmth company’s donation was perceived as more communal (F(1, 561) = 93.86, p < .001, η2p= .14). Donation type also had a significant main effect; goods donations were seen as more communal than donations of money (F(1, 561) = 5.03, p = .025, η2p= .01). In this study we found no significant interaction between donation type (money/goods) and donor warmth (low warmth/high warmth; F(1, 561) = 1.66, p = .198). However, for the low-warmth company, a donation of goods produced significantly higher perceptions of communal intent (MGoods = 4.37, SD = 1.33 vs. MMoney = 4.02, SD = 1.31; t(274) = 2.21, p = .028, d = .27) and for the high-warmth company, there was no difference in perceptions of communal intent by donation type (MGoods = 5.25, SD = 1.19 vs. MMoney = 5.15, SD = 1.12; t(287) = .75, p = .45). Effort/Sacrifice The donor warmth manipulation had a significant main effect on perceived effort and sacrifice; the high-warmth company was seen as putting in more effort than the low-warmth company (F(1, 561) = 108.70, p < .001, η2p= .16). Furthermore, there was a marginally significant main effect of donation type; the donation of goods was viewed as marginally more effortful than the donation of money (F(1, 561) = 2.80, p = .095). We also observed a significant interaction between donation type and donor warmth (F(1, 561) = 5.90, p < .02, η2p= .01). There was a significant difference in perceptions of effort and sacrifice by donation type for the low-warmth company (MGoods = 3.94, SD = 1.44 vs. MMoney = 3.48, SD = 1.39; t(274) = 2.71, p = .007, d = .33), but not for the high-warmth company (MGoods = 4.81, SD = 1.17 vs. MMoney = 4.89, SD = 1.21; t(287) = .55, p = .58). Moderated Mediation We conducted a moderated mediation analysis (Hayes 2013, model 8) to test the predicted relationship of donation type by low- or high-warmth companies on charitable credit received, with communal intent and effort/sacrifice simultaneously tested as mediators. By calculating standardized indirect effects for 5,000 bootstrapped samples, we found an indirect effect of both communal intent and effort/sacrifice conditional on company warmth when tested simultaneously. As hypothesized, inferences about communal intent mediated the effect of donation type on charitable credit for the low-warmth company (indirect effect = .12; 95% CI [.02, .22]), but not the high-warmth company (indirect effect = .03; 95% CI [–.05, .12]). Likewise, effort and sacrifice mediated the effect for the low-warmth company (indirect effect = .16; 95% CI [.05, .29]), but not for the high-warmth company (indirect effect = –.03; 95% CI [–.13, .07]). The index of moderated mediation was not significant at the 95% level of confidence. However, at the 90% level of confidence, the index of moderated mediation was significant for both communal intent (index = –.08; 90% CI [–.19, –.02]) and for effort/sacrifice (index = –.19; 90% CI [–.32, –.06]). Purchase Likelihood A similar, although nonsignificant pattern emerged for purchase likelihood. For the low-warmth company, participants reported a marginally significant increase in purchase likelihood when they read about a goods donation (MGoods = 4.54, SD = 1.49) compared to a monetary donation (MMoney = 4.23, SD = 1.59; t(274) = 1.66, p = .097). However, donation type had no effect on purchase likelihood for the high-warmth company (MGoods = 5.82, SD = 1.08 vs. MMoney= 5.81, SD = 1.11; t(287) = .04, p = .97). While the pattern was consistent with the charitable credit dependent variable, the interaction between donation type and donor warmth was not significant (F(1, 561) = 1.80, p = .18). Discussion Results from study 3B replicate the pattern observed in study 3A; people prefer donations of goods from low-warmth donors, but do not show this preference for high-warmth donors. Judgments of communal intentions mediate this pattern, and in this study, effort and sacrifice also simultaneously mediated this effect. In two additional studies (for full results see web appendix C, studies 6 and 7), we replicate these patterns. In appendix study 6, we manipulate perceived warmth by using two real-world companies: a company that is perceived as relatively low in warmth, Pepsi Co., and a company selling similar products that is perceived as relatively high in warmth, Bolthouse Farms. There was a significant interaction between donor warmth and donation type on charitable credit; Pepsi (low warmth) received more charitable credit when donating goods compared to money, whereas Bolthouse Farms (high warmth) received equal (high) amounts of charitable credit for both donation types. We found the same pattern for purchase likelihood. Due to the many potential differences besides perceived warmth between the individuals in study 3A (nursery school teacher and corporate manager), in appendix study 7, we also tested an additional manipulation of warmth that described an individual donor as either warm or cold (in a manipulation similar to study 3B). We found the same pattern of results using this manipulation. In this study, we also found that when low-warmth donors donate goods, they are viewed as having greater communal intent and receive greater charitable credit; however, goods donations do not similarly improve judgments of warmth. This pattern fits with the notion that warmth is a stable construct that varies minimally due to subtle changes or contextual factors (Kenworthy and Tausch 2008). We do predict, however, that consistently donating goods and/or sending other sustained signals of communal intent may improve perceptions of a donor’s trait warmth over time. STUDY 4: MONETARY DONATIONS CAN BE COMMUNAL The image benefits of goods donations from low-warmth donors documented thus far is somewhat unfortunate because for many charities, monetary donations are superior to equivalent in-kind or goods donations. Monetary contributions are typically preferred because cash donations provide charities with the flexibility to purchase exactly what they need when they need it, reducing waste from unneeded or untimely goods donations (Charity Navigator 2017; Conan 2011; USAID 2017). In study 4, we test whether framing a monetary donation as communal increases charitable credit for low-warmth donors in an analogous way to goods donations. We specifically test whether framing a monetary donation as a gift increases perceptions of communal intent, and subsequently, charitable credit. As opposed to other commodity exchanges, gifting is valued as a symbolic gesture of caring and commitment (Belk and Coon 1993). Therefore, if money can be described as a gift in a compelling way, it has the potential to be seen as more communal and creditworthy. Methods We recruited 650 Mechanical Turk participants, 567 of whom met our inclusion criteria (MAge = 34.24, 61.85% female); 56 participants were removed for failing to complete the study and 27 for failing the attention check. The study included a 2 (donation type: goods vs. money) × 2 (communal frame: control vs. communal [gift]) between-subjects experimental design. In all conditions, participants read about the low-warmth donor Spades Hardware. They read that “Spades Hardware is a large corporation that sells home improvement goods.” They then read either the 1) money condition: “This past weekend, Spades Hardware donated $10,000 to a humanitarian aid charity. (The charity purchased medical supplies with the $10,000.)” or 2) goods condition: “This past weekend, Spades Hardware donated medical supplies (worth a total of $10,000) to a humanitarian aid charity.” In the communal frame conditions, participants also read that “The company carefully packaged the gift of $10,000 [medical supplies] and hand-delivered it.” (We note that in this study, even in the monetary donation conditions we explained that the donation was ultimately used for medical supplies to hold information about donation use constant across conditions.) As in previous studies, we measured charitable credit (α = .89), communal intent (α = .80), perceived effort/sacrifice (α = .85), and trait warmth (α = .94).The order of all measures was counterbalanced. Results Charitable Credit The framing manipulation had a significant main effect; the company whose donation was framed as a gift received more charitable credit than did the company whose donation was not framed as a gift (F(1, 562) = 7.65, p < .01, η2p= .01). There was also a significant main effect for donation type; goods donations increased charitable credit (F(1, 562) = 6.81 , p < .01, η2p= .01). Most importantly, the 2 × 2 analysis of variance revealed a significant interaction between donation type (goods/money) and the communal frame (F(1, 562) = 18.01, p < .001, η2p= .03). In the control condition, the low-warmth donor received more charitable credit for a goods donation (MGoods = 5.85, SD = .65) than an equivalent monetary donation (MMoney = 5.36, SD = .85; t(277) = 5.33, p < .001, d = .64). However, when the donation was framed communally, the low-warmth donor received equal credit for both donation types (MGoods = 5.73, SD = .95 vs. MMoney = 5.86, SD = .84; t(285) = 1.20, p = .23; figure 7). FIGURE 7 View largeDownload slide EXPERIMENT 4: CHARITABLE CREDIT AS A FUNCTION OF COMMUNAL FRAMING AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 7 View largeDownload slide EXPERIMENT 4: CHARITABLE CREDIT AS A FUNCTION OF COMMUNAL FRAMING AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Communal Intent The communal frame had a significant main effect on communal intent; overall the donation described as a gift was viewed as more communal than the control condition counterpart (F(1, 561) = 23.36, p < .001, η2p= .04). Donation type had a marginally significant effect; goods donations were seen as marginally more communal than donations of money (F(1, 561) = 2.91, p = .09, η2p= .01). There was also a significant interaction between donation type (goods/money) and the communal frame (F(1, 561) = 4.05, p < .05, η2p= .01). In the control frame, a donation of goods versus money produced significantly higher perceptions of communal intent (MGoods = 4.72, SD = .88 vs. MMoney = 4.39, SD = 1.02; t(275) = 2.94, p < .01, d = .35). There was no difference in perceptions of communal intent in the communal frame condition (MGoods = 4.97, SD = 1.19 vs. MMoney = 4.99, SD = 1.13; t(286) = –.18, p = .86). Effort/Sacrifice The gift-giving manipulation had a significant main effect on perceived effort and sacrifice; the communal frame increased perceptions of effort and sacrifice compared to the control frame (F(1, 560) = 18.78, p < .001, η2p= .03). Donation type also showed a significant main effect; donations of goods were viewed as more effortful (F(1, 560) = 5.28, p = .022, η2p= .01). However, we did not observe a significant interaction between donation type and the communal frame (F(1, 560) = .13, p = .72). There was no difference in perceptions of effort and sacrifice by donation type for the control condition (MGoods = 4.67, SD = .99 vs. MMoney = 4.49, SD = 1.11; t(275) = 1.44, p = .15), and there was a marginally significant difference in effort perceptions for the communal-frame (gift) condition (MGoods = 5.10, SD = 1.16 vs. MMoney = 4.86, SD = 1.21; t(285) = 1.71, p = .09). Purchase Likelihood There was a significant main effect of the donor manipulation on purchase likelihood; overall participants were more likely to purchase from the company whose donation was framed communally (F(1, 555) = 4.60, p =.03, η2p= .01). Donation type also had a significant main effect; donations of goods led to higher purchase likelihood (F(1, 555) = 7.24, p < .01, η2p= .01). There was a marginally significant interaction between donation type and the gift-giving manipulation on purchase likelihood (F(1, 555) = 3.53, p = .06, η2p= .01). In the control frame condition, participants were more likely to purchase from Spades Hardware after a donation of goods (MGoods = 5.60, SD = .91) versus money (MMoney = 5.20, SD = 1.05; t(271) = 3.40, p < .001, d = .41). By contrast, in the communal frame condition, we found no difference in purchase intentions based on donation type (MGoods = 5.60, SD = 1.06 vs. MMoney = 5.56, SD = 1.11; t(284) = .30, p = .76). Measured Warmth We found a main effect of the communal frame on perceived warmth (F(1, 566) = 15.02, p < .001, η2p= .03), but a nonsignificant main effect of donation type on warmth (F(1, 566) = .167, p = .68). There was no significant interaction between communal framing and donation type on measured warmth (F(1, 566) = 1.96, p = .16). Moderated Mediation We conducted a moderated mediation analysis (Hayes 2013, model 8) to test the predicted relationship of donation type by companies who do or do not hand-deliver their gifts on charitable credit received, with communal intent and effort/sacrifice as mediators. We simultaneously tested the significance of both mediators by calculating standardized indirect effects for 5,000 bootstrapped samples and found that the model mediates the effect of donation type on charitable credit (direct effect = .08; 95% CI [–.04, .20]). We found that donation type produced an indirect effect of communal intent conditional on communal frame. As hypothesized, inferences about communal intent mediated the effect of donation type on charitable credit for the control condition (indirect effect = .09; 95% CI [.03, .16]), but not for the communal-frame (gift) condition (indirect effect = –.01; 95% CI [–.09, .06]). Effort and sacrifice did not mediate the effect for the control condition (indirect effect = .05; 95% CI [–.02, .12]), nor for the gift condition (indirect effect = .07; 95% CI [–.004, .14]). The index of moderated mediation was significant for communal intent (index = –.10; 95% CI [–.21, –.01]), but not significant for effort/sacrifice (index = .02; 95% CI [–.09, .11]). Discussion In study 4, we found once again that a low-warmth donor (Spades Hardware) received more credit for a donation of goods than a donation of money. However, when the donations were framed as communal (i.e., a gift), both donations were perceived to be motivated by communal intentions and therefore creditworthy. By increasing the perceived communality of the monetary donation, the low-warmth donor was able to receive high levels of charitable credit for making a cash donation. GENERAL DISCUSSION This research finds that donors receive charitable credit based on who they are and what they give. High-warmth donors receive more credit for being generous than low-warmth donors because people assume high-warmth donors have better intentions. However, low-warmth donors can boost the charitable credit they receive by donating goods. This article documents a novel interaction between donor traits (high vs. low warmth) and donation types (goods vs. money) and is among the first to study goods donations. The conceptualization that trait warmth triggers strong assumptions about the nature of a donor’s intent generates numerous predictions about the level of credit donors will receive. For example, although we did not find a consistent effect of donation type (goods vs. money) on effort perceptions, effort likely has independent effects on charitable credit when varied through other means. If people assume that high-warmth donors have good intentions, then, similar to donation type, donation effort may minimally impact charitable credit for high-warmth donors; high-warmth donors are likely to receive high levels of charitable credit for any magnitude of donation effort. In contrast, low-warmth donors, whose intentions are more subject to skepticism, may receive minimal credit for a low-effort donation, but receive more credit for sending a stronger signal of good intentions via a high-effort donation. These predictions are analogous to other findings in the literature in which individuals with different traits are judged differently for the same act. For example, prior research finds that donors with good reputations (i.e., social workers) are viewed as equally altruistic whether they brag about their good deed or not, whereas donors with less altruistic reputations (i.e., investment bankers) are viewed as even less altruistic when they brag about a good deed (Berman et al. 2015, study 2). Future research should further examine these and related patterns regarding donor traits and assumptions about communal intent. Although little research exists that examines outcomes associated with goods donations, there is some that distinguishes donations of money versus time (Liu and Aaker 2008; Reed et al. 2007; Macdonnell and White 2015). For example, research finds that people are happier, and therefore likely to make larger charitable donations, when they think about time rather than money (Liu and Aaker 2008), and that when the value of the two resources is equivalent, individuals view donating time as more moral than donating money (Reed et al. 2007). Assuming that time, like goods, signals communal rather than market-exchange intent, we would expect donations of time from low-warmth donors to appear more charitable than equivalent donations of money. One interesting, and perplexing, implication of the current findings is that for image reasons, low-warmth donors, such as most corporations, may benefit from donating goods rather than money, even though charities can often do more good with monetary donations (Conan 2011; USAID 2017). In fact, in web appendix C, study 8, we find that participants appear aware that nonprofits generally prefer monetary donations, and yet still show more favorable evaluations of donors who make goods donations. For low-warmth donors, such as most for-profit companies, who wish to maximize both their impact and credit received, we propose two potential solutions. First, as in study 4, low-warmth donors can describe a monetary donation as a gift, and thus frame it more communally. Alternatively, low-warmth donors can coordinate closely with charities to identify their most pressing tangible needs, and fund those needs quickly and directly via goods donations. Indeed, working closely with nonprofits to identify and meet their greatest needs may serve as an additional signal to consumers of a donor’s communal intentions. In summary, despite the fact that monetary donations are the most frequent donation type (USA.gov 2012), and the type that charities often prefer, we find that low-warmth donors who donate goods receive more credit for their generosity than those who make equivalent monetary donations. We conclude that low-warmth donors, including many corporations (Aaker et al. 2010), aiming to maximize both charitable credit and actual impact may benefit from spending their philanthropic funds on donations of goods that are coordinated with a charity’s needs, or alternatively, describing a monetary donation communally. Future research should continue to explore our differing judgments of donors based on both donation type and donor traits and the implications for charitable credit and sustained giving. DATA COLLECTION INFORMATION The first and second authors conducted the study in the article’s introduction on Amazon Mechanical Turk in March 2017; study 1 on Amazon Mechanical Turk in August 2017; study 2A on Amazon Mechanical Turk in May 2017; study 2B on Amazon Mechanical Turk in February 2017; study 3A on Amazon Mechanical Turk in October 2017; study 3B on Amazon Mechanical Turk in December 2016; and study 4 on Amazon Mechanical Turk in June 2017. The first author analyzed these data with help from the second author. Web appendix C studies: the first and second authors conducted appendix study 1 on Amazon Mechanical Turk in September 2016; appendix study 2 on Amazon Mechanical Turk in September 2017; appendix study 3 on Amazon Mechanical Turk in February 2015; appendix study 4 on Amazon Mechanical Turk in October 2016; appendix study 5 on Amazon Mechanical Turk in January 2015; appendix study 6 on Amazon Mechanical Turk in November 2015; and appendix studies 7 and 8 on Amazon Mechanical Turk in January 2017. The first author analyzed these data with help from the second author. The authors wish to thank Lucy Gao and Nicole Cooper for research assistance. They would also like to thank Alix Barasch and the Washington University Journal Club for helpful feedback on an earlier draft of the article. REFERENCES Aaker Jennifer L., Vohs Kathleen D., Mogilner Cassie ( 2010), “Non-Profits Are Seen as Warm and For-Profits as Competent: Firm Stereotypes Matter,” Journal of Consumer Research , 37 ( 2), 224– 37. Google Scholar CrossRef Search ADS   Aggarwal Pankaj ( 2004), “The Effects of Brand Relationship Norms on Consumer Attitudes and Behavior,” Journal of Consumer Research , 31 ( 1), 87– 101. Google Scholar CrossRef Search ADS   Aggarwal Pankaj, Law Sharmistha ( 2005), “Role of Relationship Norms in Processing Brand Information,” Journal of Consumer Research , 32 ( 3), 453– 64. Google Scholar CrossRef Search ADS   Andrzejewski Susan A., Mooney Emily C. ( 2016), “Service with a Smile: Does the Type of Smile Matter?” Journal of Retailing and Consumer Services , 29 ( March), 135– 41. Google Scholar CrossRef Search ADS   Ariely Dan, Bracha Anat, Meier Stephan ( 2007), “Doing Good or Doing Well? Image Motivation and Monetary Incentives in Behaving Prosocially,” American Economic Review , 99 ( 1), 544– 55. Google Scholar CrossRef Search ADS   Barasch Alixandra, Levine Emma E., Berman Jonathan Z., Small Deborah A. ( 2014), “Selfish or Selfless? On the Signal Value of Emotion in Altruistic Behavior,” Journal of Personality and Social Psychology , 107 ( 3), 393– 413. Google Scholar CrossRef Search ADS PubMed  BCLC, “Hurricane Sandy—Corporate Aid Tracker,” U.S. Chamber of Commerce Foundation, https://www.uschamberfoundation.org/corporate-citizenship-center/hurricane-sandy-corporate-aid-tracker. Bechwati Nada N., Xia Lan ( 2003), “Do Computers Sweat? The Impact of Perceived Effort of Online Decision Aids on Consumer’s Satisfaction with the Decision Process,” Journal of Consumer Psychology , 13 ( 1), 139– 48. Google Scholar CrossRef Search ADS   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   Berman Jonathan Z., Levine Emma E., Barasch Alixandra, Small Deborah A. ( 2015), “The Braggart’s Dilemma: On the Social Rewards and Penalties of Advertising Prosocial Behavior,” Journal of Marketing Research , 52 ( 1), 90– 104. Google Scholar CrossRef Search ADS   Casciaro Tiziana, Lobo Miguel Sousa ( 2008), “When Competence Is Irrelevant: The Role of Interpersonal Affect in Task-Related Ties,” Administrative Science Quarterly , 53 ( 4), 655– 84. Google Scholar CrossRef Search ADS   Charity Navigator ( 2017), “Guide to Donating Noncash Items,” https://www.charitynavigator.org/index.cfm?bay=content.view&cpid=335. Cheal David ( 1987), “Showing Them You Love Them: Gift Giving and the Dialectic of Intimacy,” Sociological Review , 35 ( 1), 150– 69. Google Scholar CrossRef Search ADS   Clark Margaret S., Mills Judson R. ( 1979), “Interpersonal Attraction in Exchange and Communal Relationships,” Journal of Personality and Social Psychology , 37 ( 1), 12– 24. Google Scholar CrossRef Search ADS   Clark Margaret S., Mills Judson R. ( 2011), “A Theory of Communal (and Exchange) Relationships,” in Handbook of Theories of Social Psychology , Vol. 2, ed. Van Lange Paul A. M., Kruglanski Arie W., Higgins E. Tory, Thousand Oaks, CA: Sage, 232– 50. Conan Neal ( 2011), “A Case for Cash Donations, Instead of Cans,” NPR, http://www.npr.org/2011/11/22/142661882/a-case-for-cash-donations-instead-of-cans. Cuddy Amy J. C., Fiske Susan T., Glick Peter ( 2004), “When Professionals Become Mothers, Warmth Doesn’t Cut the Ice,” Journal of Social Issues , 60 ( 4), 701– 18. Google Scholar CrossRef Search ADS   Cuddy Amy J. C., Fiske Susan T., Glick Peter. ( 2008), “Warmth and Competence as Universal Dimensions of Social Perception: The Stereotype Content Model and the BIAS Map,” Advances in Experimental Social Psychology , 40, 61– 149. Google Scholar CrossRef Search ADS   Cuddy Amy J. C., Glick Peter, Beninger Anna ( 2011), “The Dynamics of Warmth and Competenc Judgments, and their Outcomes in Organizations,” Research in Organizational Behavior , 31, 73– 98. Google Scholar CrossRef Search ADS   Douglas Mary, Isherwood Baron ( 1979), “The World of Goods,”  New York: Basic. Google Scholar CrossRef Search ADS   Ellingsen Tore, Johannesson Magnus ( 2011), “Conspicuous Generosity,” Journal of Public Economics , 95 ( 9–10), 1131– 43. Google Scholar CrossRef Search ADS   Fiske Susan T., Cuddy Amy J. C., Glick Peter ( 2007), “Universal Dimensions of Social Cognition: Warmth and Competence,” Trends in Cognitive Sciences , 11 ( 2), 77– 83. Google Scholar CrossRef Search ADS PubMed  Fiske Susan T., Cuddy Amy J. C., Glick Peter, Xu Jun ( 2002), “A Model of (Often Mixed) Stereotype Content: Competence and Warmth Respectively Follow from Perceived Status and Competition,” Journal of Personality and Social Psychology , 82 ( 6), 878– 902. Google Scholar CrossRef Search ADS PubMed  Flynn Francis J. ( 2003), “How Much Should I Give and How Often? The Effects of Generosity and Frequency of Favor Exchange on Social Status and Productivity,” Academy of Management Journal , 46 ( 5), 539– 53. Google Scholar CrossRef Search ADS   Flynn Francis J., Reagans Ray E., Amanatullah Emily T., Ames Daniel R. ( 2006), “Helping One’s Way to the Top: Self-Monitors Achieve Status by Helping Others and Knowing Who Helps Whom,” Journal of Personality and Social Psychology , 91 ( 6), 1123– 37. Google Scholar CrossRef Search ADS PubMed  Fournier Susan, Alvarez Claudio ( 2012), “Brands as Relationship Partners: Warmth, Competence, and In-Between,” Journal of Consumer Psychology , 22 ( 2), 177– 85. Google Scholar CrossRef Search ADS   Frey Bruno S., Goette Lorenz ( 1999), Does Pay Motivate Volunteers ? Vol. 7, Zurich, Switzerland: Institute for Empirical Research in Economics, University of Zurich. Gasiorowska Agata, Chaplin Lan N., Zaleskiewicz Tomasz, Wygrab Sandra, Vohs Kathleen D. ( 2016), “Money Cues Increase Agency and Decrease Prosociality among Children: Early Signs of Market-Mode Behaviors,” Psychological Science , 27 ( 3), 331– 44. Google Scholar CrossRef Search ADS PubMed  Goette Lorenz, Stutzer Alois, Frey Beat M. ( 2010), “Prosocial Motivation and Blood Donations: A Survey of the Empirical Literature,” Transfusion Medicine and Hemotherapy , 37 ( 3), 149– 54. Google Scholar CrossRef Search ADS PubMed  Goodman Joseph K., Cryder Cynthia E., Cheema Amar ( 2013), “Data Collection in a Flat World: The Strengths and Weaknesses of Mechanical Turk Samples,” Journal of Behavioral Decision Making , 26 ( 3), 213– 24. Google Scholar CrossRef Search ADS   Graham Melanie ( 2012), “Williams Elementary 5th Graders Help Hurricane Sandy Relief Efforts,” Newton Patch , December 9, https://patch.com/massachusetts/newton/wlliams-elementary-5th-graders-help-hurricane-sandy-rbfba38878d. Hayes Andrew F. ( 2013), Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-based Approach , New York: Guilford. Heyman James, Ariely Dan ( 2004), “Effort for Payment: A Tale of Two Markets,” Psychological Science , 15 ( 11), 787– 93. Google Scholar CrossRef Search ADS PubMed  Imhoff Roland, Woelki Jonas, Hanke Sebastian, Dotsch Ron ( 2013), “Warmth and Competence in Your Face! Visual Encoding of Stereotype Content,” Frontiers in Psychology , 4, 386. Google Scholar CrossRef Search ADS PubMed  Interbrand ( 2017), “Best Global Brands 2016 Rankings,” http://interbrand.com/best-brands/best-global-brands/2016/ranking/. Jiang Yuwei, Chen Zhansheng, Wyer Robert S.Jr. ( 2014), “Impact of Money on Emotional Expression,” Journal of Experimental Psychology , 55 ( November), 228– 33. Kenworthy Jared B., Tausch Nicole ( 2008), “Expectations about the Accuracy and Stability of Warmth versus Competence Traits: An Intergroup Analysis,” European Journal of Social Psychology , 38 ( 7), 1121– 29. Google Scholar CrossRef Search ADS   Kervyn N., Fiske Susan T., Malone Chris ( 2012), “Brands as Intentional Agents Framework: How Perceived Intentions and Ability Can Map Brand Perception,” Journal of Consumer Psychology , 22 ( 2), 166– 76. Google Scholar CrossRef Search ADS   Koschate-Fischer Nicole, Stefan Isabel V., Hoyer Wayne D. ( 2012), “Willingness to Pay for Cause-Related Marketing: The Impact of Donation Amount and Moderating Effects,” Journal of Marketing Research , 49 ( 6), 910– 27. Google Scholar CrossRef Search ADS   Lacetera Nicola, Macis Mario, Slonim Robert ( 2013), “Economic Rewards to Motivate Blood Donations,” Science , 340 ( 6135), 927– 8. Google Scholar CrossRef Search ADS PubMed  Lawrence Steven, Ronna Brown, Michael Hamill Remaley, Nina Stack ( 2014), Philanthropy and Hurricane Sandy: A Report on the Foundation and Corporate Response , New York: Foundation Center, Center for Disaster Philanthropy, Philanthropy New York, and Council of New Jersey Grantmakers. Lin-Healy Fern, Small Deborah A. ( 2012), “Cheapened Altruism: Discounting Personally Affected Prosocial Actors,” Organizational Behavior and Human Decision Processes , 117 ( 2), 269– 74. Google Scholar CrossRef Search ADS   Lin-Healy Fern, Small Deborah A. ( 2013), “Nice Guys Finish Last and Guys in Last Are Nice: The Clash between Doing Well and Doing Good,” Social Psychological and Personality Science , 4 ( 6), 692– 98. Google Scholar CrossRef Search ADS   Liu Wendy, Aaker Jennifer ( 2008), “The Happiness of Giving: The Time-Ask Effect,” Journal of Consumer Research , 35 ( 3), 543– 57. Google Scholar CrossRef Search ADS   Macdonnell Rhiannon, White Katherine ( 2015), “How Construals of Money versus Time Impact Consumer Charitable Giving,” Journal of Consumer Research , 42 ( 4), 551– 63. McGraw Peter A., Tetlock Philip E. ( 2005), “Taboo Trade-Offs, Relational Framing, and the Acceptability of Exchanges,” Journal of Consumer Psychology , 15 ( 1), 2– 15. Google Scholar CrossRef Search ADS   Newman George E., Cain Daylian M. ( 2014), “Tainted Altruism When Doing Some Good Is Evaluated as Worse than Doing No Good at All,” Psychological Science , 25 ( 3), 648– 55. Google Scholar CrossRef Search ADS PubMed  Oppenheimer Daniel M., Meyvis Tom, Davidenko Nicolas ( 2009), “Instructional Manipulation Checks: Detecting Satisficing to Increase Statistical Power,” Journal of Experimental Social Psychology , 45 ( 4), 867– 72. Google Scholar CrossRef Search ADS   Rampell Catherine ( 2012), “Volunteers Flock to Disaster Areas, Overwhelming City Relief Centers,” New York Times , November 4, http://www.nytimes.com/2012/11/05/nyregion/volunteers-and-donations-flock-to-areas-affected-by-hurricane-sandy.html. Reed AmericusII, Aquino Karl, Levy Eric ( 2007), “Moral Identity and Judgments of Charitable Behaviors,” Journal of Marketing , 71 ( January), 178– 93. Google Scholar CrossRef Search ADS   Simmons Joseph P., Nelson Leif D., Simonsohn Uri ( 2012), “A 21-Word Solution,” Dialogue , 26, 4– 12. USAID ( 2017), “Why Cash Is Best,” USAID Center for International Disaster Information CIDI, https://www.cidi.org/how-disaster-relief-works/monetary-contributions-workbestwhy-cash-is-best/#.Wh2rOGhSxPY. USA.gov ( 2012), “Donating to Charity,” USA.gov, www.usa.gov/donate-to-charity. Vlachos Pavlos A., Tsamakos Argiris, Vrechopoulos Adam P., Avramidis Panagiotis K. ( 2009), “Corporate Social Responsibility: Attributions, Loyalty, and the Mediating Role of Trust,” Journal of the Academy of Marketing Science , 37 ( 2), 170– 80. Google Scholar CrossRef Search ADS   Wagner Tillmann, Lutz Richard J., Weitz Barton A. ( 2009), “Corporate Hypocrisy: Overcoming the Threat of Inconsistent Corporate Social Responsibility Perceptions,” Journal of Marketing , 73 ( 6), 77– 91. Google Scholar CrossRef Search ADS   Webley Paul, Lea S. E., Portalska R. ( 1983), “The Unacceptability of Money as a Gift,” Journal of Economic Psychology , 4 ( 3), 223– 38. Google Scholar CrossRef Search ADS   White Martha C. ( 2015), “Is It Better to Donate Food or Money around the Holidays?” Time , http://time.com/money/4123363/donations-food-money/. Wojciszke Bogdan, Abele Andrea E. ( 2008), “The Primacy of Communion over Agency and Its Reversals in Evaluations,” European Journal of Social Psychology , 38 ( 7), 1139– 47. Google Scholar CrossRef Search ADS   Wojciszke Bogdan, Bazinska Rosa, Jaworski Marcin ( 1998), “On the Dominance of Moral Categories in Impression Formation,” Personality and Social Psychology Bulletin , 24 ( 12), 1251– 63. Google Scholar CrossRef Search ADS   Yoon Yeosun, Gürhan-Canli Zeynep, Schwarz Norbert ( 2006), “The Effect of Corporate Social Responsibility (CSR) Activities on Companies with Bad Reputations,” Journal of Consumer Psychology , 16 ( 4), 377– 90. Google Scholar CrossRef Search ADS   © The Author 2017. 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 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Consumer Research Oxford University Press

Goods Donations Increase Charitable Credit for Low-Warmth Donors

Loading next page...
 
/lp/ou_press/goods-donations-increase-charitable-credit-for-low-warmth-donors-ShDaYBNR8H
Publisher
University of Chicago Press
Copyright
© The Author 2017. 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/ucx126
Publisher site
See Article on Publisher Site

Abstract

Abstract Low-warmth actors are often assumed to lack communal (or other-oriented) intentions, even when acting generously. Low-warmth donors must therefore send stronger signals of their communal intent when donating to receive the same amount of charitable credit as high-warmth donors. Because goods are linked with communal norms, we find that donating goods allows low-warmth donors to signal communal intent and increase charitable credit received. Study 1 establishes that low-warmth donors receive less credit for unspecified donations than their high-warmth counterparts. Studies 2A and 2B show that goods donations, compared to equally valued monetary or unspecified donations, increase charitable credit for low-warmth donors. Studies 3A and 3B show that donating goods boosts charitable credit for low-warmth donors in particular; high-warmth donors are assumed to have communal intentions, and receive large amounts of credit, regardless of donation type. Finally, study 4 shows that low-warmth donors can increase charitable credit for monetary donations by describing the donation in communal terms—specifically, as a gift. This research has clear practical implications. For example, many corporations are viewed as low warmth, and most corporate donations are monetary, yet companies always have the option to donate goods instead. prosocial behavior, warmth, communal and exchange, donation type, branding, charitable credit, corporate social responsibility In 2012, the devastation of Hurricane Sandy prompted over $380 million in charitable donations (Lawrence et al. 2014). All types of donors gave, often donating in different forms. American Express contributed a $1,000,000 monetary donation. Ikea donated in-kind goods, giving more than 40,000 blankets, pillows, and towels (BCLC 2015). A teacher and her fifth-grade class raised funds to make a monetary donation (Graham 2012). Meanwhile, a hedge fund manager from the Upper West Side donated goods such as canned food and medical supplies (Rampell 2012). While all of these donations provided valuable resources to displaced victims, both the traits of the donors and the types of donations varied substantially. In this research, we propose that such variations in both donor traits and donation type can alter how people award charitable credit to those who give. People who engage in prosocial behavior are often rewarded with higher status and reputation benefits (Berman et al. 2015; Flynn 2003; Flynn et al. 2006); in other words, people grant donors “charitable credit” for their generous acts (Lin-Healy and Small 2012). However, the charitable credit that donors receive does not correspond perfectly with their donation’s impact. Instead, judgments of a donor’s intentions for donating also play a significant role (Barasch et al. 2014; Lin-Healy and Small 2012), and can be even more important than the donation’s actual impact for earning charitable credit (Newman and Cain 2014). Because presumed intentions are a critical component of the credit a donor receives, this research proposes that two factors that influence intention judgments will affect charitable credit as well. First, trait warmth, a central dimension on which people assess others, is closely linked with perceptions of others’ good intentions (Fiske, Cuddy, and Glick 2007). Whereas high-warmth actors are assumed to be well intentioned, previous research suggests that even when low-warmth actors behave prosocially, observers may assume they possess ulterior motives (Cuddy, Glick, and Beninger 2011). Second, donation type has potential to influence intention judgments. Giving goods is often the norm within social or communal relationships (Cheal 1987; Douglas and Isherwood 1979; Heyman and Ariely 2004; Webley, Lea, and Portalska 1983). Accordingly, goods donations (compared to monetary donations) may signal that a donor has donated for communal, or other-focused, reasons. In this research, we predict that donor warmth and donation type will interact to influence the judgments about donors’ communal intentions, ultimately influencing charitable credit received. High-warmth actors are typically assumed to have communal or other-focused intentions (Fiske et al. 2007) and therefore may receive high charitable credit regardless of donation type; any prosocial act reinforces prior beliefs that high-warmth givers act with others’ welfare in mind. By contrast, low-warmth actors are typically assumed to act in accordance with their own interests, and such suspicion surrounding intent may result in lower charitable credit for donating. Therefore, low-warmth donors may need to send additional signals of communal intent to gain equal levels of charitable credit. We show that donations of goods, because of their link with communality, allow low-warmth donors to strengthen their signal of communal intent and earn more credit for their charitable acts. THE LINK BETWEEN TRAIT WARMTH AND COMMUNAL INTENT Warmth refers to the extent to which an individual (or organization; Aaker, Vohs, and Mogilner 2010) is friendly, good-natured, and trustworthy (Fiske et al. 2007), and it is central to the way that people assess one another. People make warmth judgments before judgments of intelligence or competence when forming impressions, and warmth judgments receive greater attention and weight in impression formation than other interpersonal judgments (Wojciszke and Abele 2008; Wojciszke, Bazinska, and Jaworski 1998). Warmth judgments and their consequences also extend beyond individuals; for example, low-warmth companies inspire less customer loyalty than their high-warmth counterparts (Kervyn, Fiske, and Malone 2012). Due to the fundamental importance of warmth in social judgments, warmth has been studied in a wide range of contexts, including romantic partner decisions (Sinclair and Fehr 2005), hiring decisions (Casciaro and Lobo 2008), purchase decisions (Aaker et al. 2010), and customer satisfaction (Andrzejewski and Mooney 2016). Warmth may be particularly relevant to the prosocial domain because people rely on warmth judgments to predict whether or not someone is well intentioned (Fiske et al. 2002). Indeed, judgments of low warmth are linked with competitive and exploitative intentions rather than cooperative and well-meaning intentions (Fiske et al. 2007; Fournier and Alvarez 2012; Kervyn et al. 2012). This association extends to the realm of corporate social responsibility; consumers typically perceive for-profit companies as low in warmth (Aaker et al. 2010) and they tend to be suspicious of companies’ motives for engaging in prosocial behavior (Vlachos et al. 2009). Multiple studies find that the reputational benefits of donating are attenuated, or even reversed, when consumers believe that charitable giving was motivated by profit or was inauthentic in any way (Koschate-Fischer, Stefan, and Hoyer 2012; Wagner, Lutz, and Weitz 2009; Yoon, Gürhan-Canli, and Schwarz 2006). In this research, we hypothesize that low-warmth donors’ intentions are viewed specifically as less communal than high-warmth donors’ intentions. Research in social psychology distinguishes between two categories of social relationships: communal and exchange-based relationships. In communal relationships, benefits are given to others noncontingently and with the recipient’s welfare in mind. Relationships between parents and children or romantic partners typically follow communal norms (Clark and Mills 1979, 2011), such as when parents provide food and shelter for their children without the expectation of being paid back. By contrast, business transactions and most interactions with strangers and acquaintances typically follow exchange norms common in exchange-based relationships; benefits are given to others with the expectation that the giver will receive comparable benefits in return (Clark and Mills 1979, 2011). For example, when selling a car, both the salesperson and the customer give something and expect to receive something of comparable value in return. Although the communal versus exchange distinction was originally conceptualized to describe human relationships (Clark and Mills 1979), consumers also interpret their interactions with companies through the lens of communal or exchange norms, judging some companies to behave more consistently with communal norms and others to behave more consistently with exchange norms (Aggarwal 2004; Aggarwal and Law 2005). We conducted a correlational study as an initial investigation into the relationship between donor warmth and judgments of communal intentions. We asked 96 Mechanical Turk participants about each company on Interbrand’s list of the 10 most valuable global brands from 2016 (Apple, Google, Coca-Cola, Microsoft, Toyota, IBM, Samsung, Amazon, Mercedes-Benz, or General Electric; Interbrand 2017). Participants rated each company on trait warmth (that is, to what extent do the following traits describe the company in general: friendly, well-intentioned, trustworthy, warm, good-natured, sincere; Fiske et al. 2002; α = .96). Next, participants imagined that the company had behaved prosocially in a generalized context (“Imagine that some people needed help and Coca-Cola helped them. How would you interpret this action by Coca-Cola?”). They then rated the company’s intentions on a five-item measure of communal intentions: 1) the company did not expect to receive any benefits from helping, 2) the company helped to respond to others’ needs, 3) the company has a genuine desire to help others, 4) the company helped with hopes of benefiting themselves (reverse-coded), and 5) the company helped in order to get ahead (reverse-coded; α = .80). Results showed that, as predicted, trait warmth and communal intentions were significantly positively correlated (r= .35, p < .001; full details in web appendix A). We propose that such connections between trait warmth and communal, or other-oriented, intentions have the potential to influence the amount of charitable credit that donors receive. Donations that are perceived as communal in nature—that is, motivated by recipients’ needs—tend to be evaluated positively, whereas similar donations that are viewed as self-serving typically are evaluated more negatively (Lin-Healy and Small 2013; Newman and Cain 2014). When judging charitable behavior, people balance the information that the donor has done something good with signals about whether or not the donor truly cared about the recipient’s welfare to arrive at a final judgment of how favorably the good deed should be perceived (Berman et al. 2015). Although people may generally assume that low-warmth donors have self-interested motives, links between goods and communality may allow low-warmth donors who donate goods (compared to money or an unspecified donation) to more effectively signal communal intent, and ultimately, increase charitable credit. GOODS SIGNAL COMMUNAL INTENT The prosocial consumer behavior literature has distinguished between donations of money and time (Liu and Aaker 2008; Macdonnell and White 2015; Reed, Aquino, and Levy 2007); however, there is minimal work investigating goods donations. This gap is notable because donors frequently give in-kind donations of tangible goods such as medical supplies, food, water, and clothing (Charity Navigator 2017). Because of links between giving goods and communality, we propose that donating goods can increase the extent to which low-warmth donors’ contributions are viewed as communally motivated. Giving goods is often the norm within social or communal interactions (Heyman and Ariely 2004), such as in the realm of gift giving (Cheal 1987; Douglas and Isherwood 1979). Although charities themselves tend to prefer monetary donations to other contributions because money is fully fungible and offers flexibility to cover a charity’s most pressing needs (Conan 2011), goods may be more consistent with psychological representations of communality and prosociality. For example, exchanges of goods are considered appropriate within close social relationships, whereas comparable exchanges of money are viewed as taboo within these relationships (Belk and Coon 1993; Webley et al. 1983). The connection between goods and communality is further supported by research about incentives. In some cases, people are less generous when a monetary incentive is offered for generous behavior compared to no incentive at all (Ariely, Bracha, and Meier 2007; Frey and Goette 1999); however, goods incentives, such as T-shirts and lottery tickets, can have neutral or even positive effects on giving (Goette, Stutzer, and Frey 2010; Lacetera, Macis, and Slonim 2013). Research in a similar vein suggests that goods effectively signal communal norms (Gasiorowska et al. 2016; Jiang, Chen, and Wyer 2014; McGraw and Tetlock 2005). When offered a monetary reward, participants report a lower likelihood of helping a friend when the reward is small than when it is large, a pattern consistent with exchange norms whereby effort corresponds with incentive size. When offered a reward of goods, however, participants report an equal (and high) likelihood of helping a friend regardless of reward size, a pattern of noncontingent helping that is consistent with communal norms (Clark and Mills 1979, 2011; Heyman and Ariely 2004). Additional work on gift giving shows that although cash is often more valuable to recipients, gifts of goods can more effectively signal altruistic intent and yield higher esteem for the giver (Ellingsen and Johanneson 2011). Due to differing inferences that arise from goods versus monetary transactions, low-warmth donors may boost signals of communal intent when they donate goods. Further, because goods donations signal positive regard for a recipient (Ellingsen and Johannesson 2011), goods donations may promote communal inferences above and beyond not only monetary donations, but unspecified donations as well. THE PRESENT RESEARCH We predict that low-warmth donors, but not high-warmth donors, will receive greater credit for donating goods compared to money due to differing inferences about communal intent. Because of links between trait warmth and communal intentions (Fiske et al. 2007), and because perceptions of good intentions are critical for gaining credit from charitable giving (Barasch et al. 2014; Newman and Cain 2014), we first hypothesize that trait warmth will play an important role in how much credit a donor receives for giving. We further predict that high-warmth donors may receive high credit for generous acts regardless of the substance of that act; any prosocial act reinforces a prior belief that the high-warmth giver was acting with others’ welfare in mind, making donation type less influential. Low-warmth donors, by contrast, may comparatively struggle to receive credit because people doubt the benevolence of their intent (see figure 1, conceptual model step 1). FIGURE 1 View largeDownload slide CONCEPTUAL MODEL STEP 1: CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION FIGURE 1 View largeDownload slide CONCEPTUAL MODEL STEP 1: CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION Therefore, low-warmth donors may need to send stronger signals of their communal intent compared to their high-warmth counterparts in order to receive similar levels of charitable credit. Due to links between goods and communal norms, we predict that donations of goods (vs. money or unspecified donations) may allow low-warmth donors to strengthen the signal of their communal intent, and increase the amount of credit that they receive (see figure 1, conceptual model step 2). OVERVIEW OF CONSTRUCTS AND STUDIES Throughout our research, we focus on three primary psychological constructs: trait warmth, communal intent, and charitable credit. First, the construct of warmth refers to a character trait regarding how friendly, trustworthy, and good-natured someone is (Fiske 1999). We propose that trait warmth is a key predictor of the influence of donation type on charitable credit. Second, communal (vs. exchange-based) intentions serve as our mediating variable. Judgments of communal, or other-regarding, intentions have been identified as feeding into judgments about generosity (Barasch et al. 2014). Finally, “charitable credit,” or image-related benefits that accrue due to a prosocial act (Lin-Healy and Small 2012; Newman and Cain 2014), serves as our main dependent variable. Charitable credit is conceptualized as the credit received when someone acts benevolently (Lin-Healy and Small 2012). While all of these constructs are related, they also are clearly distinct from one another. Although at times previous work has used notions of warmth and communion interchangeably (Fiske et al. 2002), we delineate important distinctions between them both empirically and conceptually. Warmth refers to a broad and relatively stable trait-level evaluation of a person or organization. Communal intent, the proposed mediating construct, refers to perceptions of the donor’s motives behind a prosocial act, specifically the extent to which the generous act was done with the recipient’s welfare in mind. Charitable credit, our main outcome measure, refers to the amount of credit or esteem a donor receives based on a particular prosocial act (though we note this does not necessarily change global trait perceptions of the donor; Lin-Healy and Small 2012). In the studies where these constructs are measured, we specify in our measurement whether we are asking participants to judge a stable trait (warmth), the motives behind the act (communal intent), or the generosity of the act (charitable credit). We also empirically distinguish these constructs in all studies in which they are measured via confirmatory factor analyses (all reported in web appendix A). Six studies in the main text, as well as eight additional studies reported in the web appendix, test key pieces of our conceptual model. Study 1 establishes the relationship between warmth, communal intent, and the amount of credit a donor receives, showing that consumers grant more charitable credit to high-warmth donors compared to low-warmth donors for unspecified donations, even when they perceive the donations to have equal value. Studies 2A and 2B then demonstrate that low-warmth donors receive more credit for donating goods than equivalent monetary amounts or unspecified donations, and that this is due to differing perceptions of communal intent. Studies 3A and 3B test the full conceptual model from step 2 (figure 1), finding that donor warmth moderates the increased credit for goods donations compared to monetary donations. When low-warmth donors give, they receive more credit for goods donations; when high-warmth donors give, they receive high charitable credit for all donation types. Finally, in study 4, we show one way in which low-warmth donors can increase the charitable credit they receive for monetary donations (which charities typically prefer): low-warmth donors can frame their monetary donations communally—specifically, as a gift. Eight studies in the web appendix replicate these patterns and rule out alternative explanations. STUDY 1: THE IMPACT OF DONOR WARMTH ON CHARITABLE CREDIT Study 1 tests the relationship predicted by step 1 of our conceptual model (figure 1). We predict that low-warmth donors will be judged to have lower communal intent than high-warmth donors, and that this will result in less charitable credit for low-warmth (vs. high-warmth) donors. Previous research finds that for-profit companies tend to be viewed as low warmth (Aaker et al. 2010), and in this study the low-warmth donor is a for-profit company. Nevertheless, judged warmth can vary substantially even among companies (Kervyn et al. 2012). In the high-warmth condition, we therefore describe the same company from the low-warmth condition, but portray the company as particularly high in warmth. Because in this study we are most interested in the general relationship between warmth, communal intent, and charitable credit, we do not specify donation type; however, donation type is a key element that we test in our remaining studies. Pretest Methods For all studies, sample size and exclusion criteria were determined ex ante. Based on predetermined exclusion criteria, we analyzed data only from participants who completed the study and passed an instructional attention check designed to identify inattentive participants (Goodman, Cryder, and Cheema 2013; Oppenheimer, Meyvis, and Davidenko 2009; web appendix B). Following recommendations from Simmons, Nelson, and Simonsohn (2012), we report all data exclusions, all manipulations, and all measures for all studies. We first conducted a pretest in which we recruited 109 Mechanical Turk participants, 91 of whom met our inclusion criteria (MAge = 34.89, 52.75% female); nine participants were removed for failing to complete the study and nine for failing the attention check. In this pretest, we manipulated donor warmth. In a low-warmth condition, participants read “Spades Hardware is a company that sells home improvement goods.” In the high-warmth condition, participants read, “Spades Hardware is a friendly company that sells home improvement goods. Spades Hardware is always warm and welcoming toward visitors.” Participants then rated the company on the six-item perceived warmth scale (see correlational study in the introduction) with items presented in a randomized order. Pretest Results The pretest descriptions successfully manipulated donor warmth. In the high-warmth condition, Spades Hardware was rated as warmer (MHigh-Warmth = 5.98, SD = .81) than in the low-warmth condition (MLow-Warmth = 4.53, SD = 1.36, t(89) = 6.14, p < .001, d = 1.30). Main-Study Methods We recruited 320 Mechanical Turk participants, 274 of whom met our inclusion criteria (MAge = 33.53, 52.31% female); 28 participants were removed for failing to complete the study and 18 for failing the attention check. Using the same descriptions of the low- and high-warmth companies from our pretest, participants in both conditions then read about an unspecified donation, specifically: “This past weekend Spades Hardware made a donation.” In both conditions, participants indicated the charitable credit that they would award to the donating company. Specifically, participants were asked, “How favorably do you view Spades Hardware on the characteristics below as a result of their donation?” Participants rated the extent to which they viewed the company as generous, helpful, and charitable as a result of their donation. Participants also rated how beneficial they believed the company is and to what extent the company makes the world a better place as a result of their donation (1 = Not at all, 7 = Very much so; all items adapted from Lin-Healy and Small 2012; Newman and Cain 2014; see web appendix B for all measures). Although we originally anticipated that the measures of charitable descriptors (generous, helpful, and charitable; Lin-Healy and Small 2012) and charitable benefit (how beneficial they believed the donation was and to what extent the donor made the world a better place; Newman and Cain 2014) would assess distinct constructs, responses to all items loaded onto a single factor that was highly reliable (α = .86). Therefore, we combined all five items to create a single and comprehensive “charitable credit” measure, which we use throughout the article. To assess mediation of communal intent, we asked participants to rate the donor on the extent to which it signaled communal versus exchange-based intentions using the five-item scale of communal intent described in the correlational study from this article’s introduction (α = .80, see also web appendix B). As a manipulation check, participants also rated the donor on trait warmth using the six-item scale of warmth described in the introduction’s correlational study. Importantly, perceptions of both competence and wealth have been shown to correlate negatively with warmth judgments (Cuddy, Fiske, and Glick 2008; Fiske et al. 2002), and we therefore measured both to ensure that they did not account for any findings. We measured competence judgments using Fiske et al.’s (2002) scale, which included the following items: competent, confident, capable, efficient, intelligent, and skillful (α = .93). We measured perceived wealth by asking, “How wealthy do you think Spades Hardware is?” (1 = Not at all wealthy, 7 = Extremely wealthy). We also measured purchase likelihood to gauge correspondence between charitable credit and consumer choices. Participants read: “Please rate how likely you would be to go to Spades Hardware next time you need home improvement goods” (1 = Not at all likely, 7 = Very likely). The order of all dependent measures was counterbalanced. Finally, as a follow-up measure, we asked participants “How much would you estimate Spades Hardware’s donation was worth?” to ensure that differences in donation value estimation did not account for any findings. Participants could enter any value that they wished. Please see web appendix A for factor analysis results for measures in all studies. Main-Study Results Warmth Manipulation Check The manipulation check indicated that our manipulation was successful; the high-warmth company was rated as warmer (MHigh-Warmth = 5.79, SD = .87) than the low-warmth company (MLow-warmth = 4.99, SD = 1.19; t(272) = 6.01, p < .001, d = .73). Charitable Credit Participants awarded the high-warmth company more charitable credit (MHigh-Warmth = 5.54, SD = .84) than the low-warmth company (MLow-warmth = 5.21, SD = 1.07; t(272) = 2.80, p = .005, d = .32). Communal Intent Participants judged the high-warmth company to have greater communal intent (MHigh-Warmth = 4.71, SD = 1.02) than the low-warmth company (MLow-warmth = 4.24, SD = 1.09; t(272) = 3.66, p < .001, d = .44). Purchase Likelihood Participants reported greater purchase likelihood for the high-warmth company (MHigh-warmth = 5.72, SD = 1.04) than for the low-warmth company (MLow-warmth = 5.10, SD = 1.35; t(268) = 4.21, p < .001, d = .51). Estimated Donation Value Using free response, participants estimated that the donation was worth a median value of $2,000 in both conditions (a Mann-Whitney test indicated that there was a nonsignificant difference between the two conditions; U = 9288.50, Z = –.04, p = .97). This implies that the donor warmth manipulation did not influence the perceived value of the donation. Mediation The mediation analysis showed that perceptions of communal intent mediated the effect of manipulated donor warmth on charitable credit. Using methods prescribed by Hayes (2013, model 4) we tested the significance of communal (vs. exchange) intent as the mediator by calculating standardized indirect effects for 5,000 bootstrapped samples and found that communal intent mediates the effect of donor warmth on charitable credit (indirect effect = .25; 95% CI [.11, .39]; direct effect = .08; 95% CI [–.11, .27]). This pattern remains significant when we control for competence and perceived wealth (indirect effect = .12; 95% CI [.02, .23]; direct effect = .02; 95% CI [–.15, .19]; figure 2). We also find that communal intent partially mediated the effect of donor warmth on purchase likelihood (web appendix A). FIGURE 2 View largeDownload slide THE RELATIONSHIP BETWEEN DONOR WARMTH AND CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION, AS MEDIATED BY COMMUNAL INTENT FIGURE 2 View largeDownload slide THE RELATIONSHIP BETWEEN DONOR WARMTH AND CHARITABLE CREDIT FOR AN UNSPECIFIED DONATION, AS MEDIATED BY COMMUNAL INTENT Discussion Consistent with step 1 of our conceptual model, this study finds that people ascribe charitable credit differently based on who a donor is. A high-warmth company received greater charitable credit for a donation than a low-warmth company because donations from high-warmth donors are viewed as more communal. These results set up our following studies, which test step 2 of our conceptual model (figure 1). We first test whether goods donations increase the amount of credit that a low-warmth donor receives, and we then test how donor warmth interacts with donation type to influence the charitable credit a donor receives. STUDY 2A: CORPORATE DONATIONS OF GOODS RECEIVE MORE CREDIT Study 2A begins to test step 2 of our conceptual model (figure 1) by testing whether consumers judge a low-warmth donor more favorably for a goods donation than for a monetary donation. We also include a control condition where the donation type is not specified. We use the same description of a for-profit company from the low-warmth condition in study 1 to operationalize a low-warmth donor in this study. Methods We recruited 450 Mechanical Turk participants, 406 of whom met our inclusion criteria (MAge = 34.51, 40.25% female); 30 participants were removed for failing to complete the study and 14 for failing the attention check. Participants read the same description of the donor from the low-warmth condition in study 1: “Spades Hardware is a large corporation that sells home improvement goods.” Participants were assigned to one of three conditions: 1) control condition: “This past weekend Spades Hardware donated to a food bank,” 2) monetary donation condition: “This past weekend Spades Hardware donated $2,000 to a food bank,” or 3) goods donation condition: “This past weekend Spades Hardware donated boxes of canned food to a food bank. (The donation cost the company $2,000 and it would have cost the food bank the same amount to obtain those goods.)” We did not include a donation value for the control condition, due to a concern that mentioning the value would lead participants to assume the company made a monetary donation. However, we described the donations in the other conditions to be worth $2,000 because that was the median estimated value of the unspecified donations in study 1. We included information about the cost for the company and value of the donation for the food bank in the goods condition to ensure that participants would not assume the actual value for the charity was greater than $2,000 when the company donated goods. In all conditions, participants rated the company on the five-item charitable credit scale (α = .90). As in study 1, we also measured purchase likelihood and asked participants in the control condition, “How much would you estimate Spades Hardware’s donation was worth?” Participants could enter any value that they wished. Finally, we asked all participants “Who would get more canned food for $2,000?” 1) Spades Hardware, 2) the food bank, 3) they can get the same amount. This question was intended to confirm that participants did not assume that the donor could provide greater value by donating goods rather than money. (We note that in this initial test of the preference for goods donations from low-warmth donors, we did not measure communal intent, but do so in studies 2B–4). Results Charitable Credit Planned comparisons showed that participants awarded the company more charitable credit for donating goods (MGoods = 5.84, SD = .80) than money (MMoney = 5.48, SD = .92; t(270) = 3.48, p = .001, d = .42), or for making an unspecified donation (control condition); MControl = 5.52, SD = .92; t(273) = 3.12, p = .002, d = .38). There was no difference in charitable credit between the monetary and unspecified (control) donation condition (t(263) = .31, p = .75; figure 3). FIGURE 3 View largeDownload slide STUDY 2A: CHARITABLE CREDIT AS A FUNCTION OF DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 3 View largeDownload slide STUDY 2A: CHARITABLE CREDIT AS A FUNCTION OF DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Estimated Donation Value Using free response, participants in the control condition estimated that the donation was worth a median value of $1,000. This is lower than the value we stated in the other two conditions, so as an additional test, we looked at only participants who gave estimates in the top 50% (median = $5,000). We found that this group still rated the donating company as marginally less charitable (MControl = 5.62, SD = .86) than those in the goods donation condition (MGoods = 5.84, SD = .80; t(206) = 1.86, p =.065). This suggests that lower assumed donation values in the control (unspecified donation) condition do not fully account for the low levels of charitable credit in that condition. Purchase Likelihood Using planned comparisons, we found that participants indicated a marginally higher likelihood of purchasing from Spades Hardware when it donated goods (MGoods = 5.58, SD = .99) compared to money (MMoney = 5.37, SD = .95; t(268)=1.76, p = .079). There was a nonsignificant difference in purchase likelihood between the goods donation and the control (unspecified donation) conditions (MControl = 5.46, SD = .93; t(271) = .90, p = .37) and between the monetary and control conditions (t(263) = .77, p = .44). Donation Value Inferences In response to the question “Who would be able to get the most canned food for $2,000?” 45.81% of participants responded that the food bank would be able to buy more canned food with $2,000; 48.28% believed that the food bank and hardware store could acquire the same amount of canned food for that amount of cash. Only 5.91% of participants believed that Spades Hardware would get the most bang for their buck. These patterns indicate that participants did not rate charitable credit higher for goods donations because they thought that the donor was able to acquire, and thus donate, more canned food than the charity could buy for the same amount. In fact, participants most frequently indicated that the charity could buy as much or more canned food as the donor, a pattern commonly true in the real world because nonprofits often receive discounts on such goods (Conan 2011; White 2015). Discussion Consistent with our conceptual model, we observe a significant increase in the charitable credit that a low-warmth donor receives when donating goods versus money. Further, we observe that the boost in charitable credit for goods donations not only occurs in comparison to a monetary donation, but also in comparison to an unspecified donation. An additional study (web appendix C, study 1) found that while a goods donation from a low-warmth donor garnered more charitable credit than a monetary donation, both donations received more credit than a no-donation condition. These results suggest that, for low-warmth donors, donating goods results in a unique boost in charitable credit compared to other donation types, but giving money also increases charitable credit compared to making no donation at all. In study 2B, we test whether judgments about the communal nature of a goods donation is responsible for greater charitable credit received by low-warmth donors who give goods. In this study, the vast majority of participants believed the food bank could get the same amount if not more food than the hardware store for the amount of money given, and yet, we still find the company receives greater credit for giving goods rather than their equivalent cash value. As an especially conservative test of the image benefits of goods donations (web appendix C, study 2) we compared three donations in a between-subjects experiment: 1) a monetary donation of $1,000, 2) an equivalent goods donation ($1,000 worth of canned food), and 3) a smaller goods donation ($900 worth of canned food). We found that both goods donations garnered higher charitable credit (and purchase likelihood) than the monetary donation, and the two goods donations were perceived as equally charitable. These results showed that the charitable credit benefits of goods donations still emerge even when the goods donation has a lower value than the monetary donation. We conclude that the benefits of goods donations for low-warmth donors cannot be explained by higher interpreted financial value of goods donations. Moving forward, in all studies, we continue to explicitly state a donation value in the goods condition that matches the donation amount in the monetary donation condition. Even though mentioning a monetary value may dampen communality effects for goods transactions (Heyman and Ariely 2004), it is essential for us to hold donation value constant across conditions to ensure greater donation value is not inferred when donors give goods. Please also see web appendix C, study 3, for a description of an additional study that replicates these findings and compares a monetary donation to two goods donations, one with and one without a monetary value. While both goods donations were given more charitable credit than the monetary donation, this study finds that the effect is indeed stronger when the monetary value of a goods donation is absent. STUDY 2B: COMMUNAL (VS. EXCHANGE) INTENTIONS AS MEDIATOR Study 2B further explores consumer preferences for goods donations by low-warmth donors by testing mediating factors. We test whether goods versus monetary donations trigger inferences of communal versus exchange-based intentions, resulting in different levels of charitable credit. We additionally test two other potential mediators behind the preference for corporate goods donations: perceptions of effort and sacrifice. Consumers may reasonably infer that companies exert more effort, more sacrifice, or both when donating tangible goods compared to simply writing a check to a charity (goods need to be chosen, procured, delivered, etc.). We therefore also measure and test perceived effort and sacrifice as potential contributors to the greater credit for goods donations by low-warmth donors. Methods For study 2B, we recruited 240 Mechanical Turk participants, 212 of whom met our inclusion criteria (MAge = 32.93, 53.7% female); 17 participants were removed for failing to complete the study and 11 for failing the attention check. Participants were randomly assigned to either a monetary donation or goods donation condition and read the following scenario: “Spades Hardware is a large corporation that sells home improvement goods. This past weekend, Spades Hardware made a donation of $1,000,000 ($1,000,000 worth of medical supplies) to humanitarian aid efforts.” Participants next rated Spades Hardware on the charitable credit items (α = .93). To measure mediating factors, participants also rated Spades Hardware on the five-item scale of communal intent described in the correlational study in this article’s introduction and study 1 (α = .85). We additionally measured perceived effort and sacrifice related to the donation. Participants were asked to what extent they agree with the following statements: “Spades Hardware put a lot of effort into this donation,” “Spades Hardware worked hard on this donation,” and “Spades Hardware put thought into this donation” (Bechwati and Xia 2003). We evaluated perceived sacrifice by asking the extent to which participants agreed that “Spades Hardware sacrificed when making this donation” as well as asking “How big was Spades Hardware’s sacrifice when making this donation?” While we planned to test effort and sacrifice as separate constructs, these two potential mediators loaded onto a single factor that was highly reliable (α = .92), and therefore we combined items to create a single effort/sacrifice measure. Finally, we measured purchase likelihood by asking participants, “How likely would you be to go to Spades Hardware next time you need home improvement goods?” (1 = Not at all likely, 7 = Very likely). The order of all dependent measures was counterbalanced. Results Charitable Credit In line with study 2A, participants awarded the company more charitable credit when the company donated goods rather than money (MGoods = 5.95, SD = .84 vs. MMoney = 5.65, SD = .90; t(210) = 2.55, p = .01, d = .35). Purchase Likelihood Participants reported a marginally higher likelihood of purchasing from the company following a goods, rather than monetary, donation (MGoods = 5.49, SD = 1.13; MMoney = 5.23, SD = 1.12; t(208) = 1.70, p = .09, d = .24). Communal Intent Participants believed the company signaled communal (vs. exchange) intentions to a greater extent when it donated goods (in this case, medical supplies; MGoods = 4.66, SD = 1.03) rather than money (MMoney = 4.34, SD = 1.23; t(210) = 2.07, p = .04, d = .29). Effort/Sacrifice There was a nonsignificant difference in perceived effort and sacrifice for donations of goods versus money (MGoods = 5.04, SD = 1.36 vs. MMoney = 4.92, SD = 1.43; t(210) = .64, p = .52). Mediation The mediation analysis showed that perceptions of communal intent mediated the effect of donation type on charitable credit. Using methods prescribed by Hayes (2013, model 4) we simultaneously tested the significance of both mediators by calculating standardized indirect effects for 5,000 bootstrapped samples and found that communal intent mediates the effect of donation type on charitable credit (total indirect effect = .19; 95% CI [.02, .36]; direct effect = .14; 95% CI [–.04, .34]). We found a statistically significant indirect effect of perceived communal intent (.15; 95% CI [.01, .30]). The indirect effect of effort and sacrifice was not significant (.04; 95% CI [–.09, .17]; figure 4). FIGURE 4 View largeDownload slide THE RELATIONSHIP BETWEEN DONATION TYPE AND CHARITABLE CREDIT FOR A LOW-WARMTH DONOR AS MEDIATED BY COMMUNAL INTENT FIGURE 4 View largeDownload slide THE RELATIONSHIP BETWEEN DONATION TYPE AND CHARITABLE CREDIT FOR A LOW-WARMTH DONOR AS MEDIATED BY COMMUNAL INTENT Discussion Results from study 2B show that differences in communal intent perceptions based on goods versus monetary donations explain the effect of donation type on charitable credit for low-warmth donors. These results suggest that consumers judge that low-warmth donors who donate goods have communal intentions for doing so, whereas low-warmth donors who donate money have relatively exchange-based intentions for doing so; this communal versus exchange signaling in turn influences the charitable credit received. Perceived effort and sacrifice, although almost certainly important in assessments of many donations, did not significantly differ due to donation type in this scenario. As a robustness check, we replicated these findings in an additional study (web appendix C, study 4). Participants imagined that a natural disaster impacted a neighboring town and Spades Hardware donated to the relief efforts. Participants read that the store donated either 1) $1,000, 2) $1,000 worth of food, or 3) $1,000 worth of lumber to the cause; the latter two conditions were both included to test whether the “fit” between the donation and the company matters (for Spades Hardware, lumber was pretested as a substantially “higher-fit” donation than food). We tested both donations to determine whether consumers believe the cause may receive greater benefit (more goods) when the donation is relevant to the company (in this case, lumber from the hardware store) rather than irrelevant. Again, companies that donated goods (vs. money) received greater charitable credit. There was a nonsignificant difference in charitable credit for the low-fit and high-fit donations, which leads us to believe that consumers are not incorporating potential increases in benefit to the cause when companies donate wholesale goods. In this study, both perceived communal intent and sacrifice/effort had significant and separate mediating effects. Results from study 2B demonstrate support for the mediating role of communal intent for this pattern; in the follow-up study, effort and sacrifice simultaneously mediated the effect of donation type on charitable credit alongside communal intent. Taken together, we interpret these mediation results to mean that increased perceptions of effort and sacrifice can contribute to increased charitable credit for goods donations; however, even when goods donations do not increase perceptions of effort and sacrifice (as in the main study 2B), judgments of communal intent can drive the overall effect. In addition to providing insight about process, these studies also provide evidence about the robustness of the preference for goods donations by low-warmth donors by testing two new contexts (humanitarian aid efforts and a sudden disaster) and two new donations (medical supplies and lumber). We conclude that the increased credit given to low-warmth donors for donations of goods is not limited to a particular type of cause or specific type of goods donation. STUDIES 3A AND 3B: THE INTERACTION OF DONOR WARMTH AND DONATION TYPE Studies 3A and 3B were designed to test the full conceptual model (figure 1, step 2), including the moderating role of warmth in the preference for goods donations. In an early study for this project (web appendix C, study 5), we measured perceptions of charitable credit for a donation of either money or goods by two types of donors: a low-warmth for-profit company (Spades Hardware) and a high-warmth family (the Joneses). We observed a significant interaction between donor warmth and donation type; the (low-warmth) company received more credit for a donation of goods than a monetary donation, while the (high-warmth) family received more credit for a donation of money. Although these results are consistent with the conceptual model, there are many differences between companies and families besides perceived warmth. In studies 3A and 3B, we build on this initial result and more precisely manipulate warmth. STUDY 3A: THE MODERATING ROLE OF WARMTH—INDIVIDUAL DONORS Thus far we have relied on for-profit companies as the operationalization of a low-warmth donor. In study 3A, we describe individual donors, and manipulate whether they are high or low warmth. Past research shows that certain groups of individuals viewed as subordinate and noncompetitive—for example, elderly individuals and homemakers—tend to be viewed as high warmth (Cuddy, Fiske, and Glick 2004). In contrast, those who are high in status and/or competitive—for example, educated individuals—tend to be viewed as low warmth (Fiske et al. 2002). Previous research examines warmth and competence perceptions of individuals in a variety of professions and finds that people’s perceived warmth varies significantly based on their profession (Imhoff et al. 2013). Building on this work, we operationalized a high-warmth donor as a nursery school teacher and a low-warmth donor as a corporate manager in study 3A. We hypothesized that, consistent with our conceptual model, low-warmth donors will receive more charitable credit for goods donations due to increased perceptions of communal intent, whereas high-warmth donors will receive high charitable credit regardless of donation type (because of high communal intent inferred across donation type). Pretest Methods We recruited 110 Mechanical Turk participants, 105 of whom met our inclusion criteria (MAge = 35.63, 50% female); three participants were removed for failing to complete the study and two for failing the attention check. In this pretest, we manipulated donor warmth. In a high-warmth condition, participants read “Heather is a 41-year-old nursery school teacher.” In the low-warmth condition, participants read, “Heather is a 41-year-old corporate manager.” Participants were then asked to rate the target on the six-item perceived warmth scale (α = .96) with items presented in a randomized order. Pretest Results The descriptions in the pretest successfully manipulated donor warmth. Heather the nursery school teacher was rated as significantly warmer (MHigh-Warmth = 5.70, SD = .90) than Heather the corporate manager (MLow-Warmth = 4.30, SD = 1.09, t(103) = 7.15, p < .001, d = 1.40). Main-Study Methods We recruited 675 Mechanical Turk participants, 599 of whom met our inclusion criteria (MAge =35.75, 60.77% female); 39 participants were removed for failing to complete the study and 37 for failing the attention check. As in the pretest, participants in the high-warmth [low-warmth] condition read the following donor descriptions: “Heather is a 41-year-old nursery school teacher [corporate manager].” Participants were then informed that this past weekend, Heather contributed either a monetary donation (“$100”) or a goods donation (“$100 worth of canned food”) to her local food bank. We then asked participants to respond to the items that measured charitable credit (α = .89), communal intent (α = .82), perceived effort/sacrifice (α = .88), and trait warmth (α = .92, a manipulation check). In addition, due to potential differences in perceived competence and wealth for high-warmth and low-warmth donors, we measured these factors to ensure that they did not account for any findings. The order of all dependent measures was counterbalanced. Main-Study Results Warmth Manipulation Check The high-warmth individual was perceived as significantly warmer (MHigh-Warmth = 6.01, SD = .88) than the low-warmth individual (MLow-Warmth = 5.58, SD = .93; t(596) = 5.78, p < .001, d = .23). Based on a 2 × 2 ANOVA, we found a main effect of the warmth manipulation on perceived warmth (F(1, 598) = 33.63, p < .001); donation type did not exert a significant main effect on warmth (F(1, 598) = .11, p = .74). We observed a marginally significant interaction between donation type and manipulated donor warmth on measured warmth (F(1, 598) = 3.67, p = .06; please also see a similar analysis in study 3B). Charitable Credit We observed a significant main effect of donor type on charitable credit; the high-warmth individual received more charitable credit for her donation than the low-warmth individual (F(1, 598) = 41.70, p < .001, η2p= .07). Donation type also had a significant main effect; donations of goods received more credit than donations of money (F(1, 598) = 5.78, p = .017, η2p= .01). Most importantly, the 2 × 2 ANOVA revealed a significant interaction between donation type (goods/money) and donor warmth (low warmth/high warmth; F(1, 598) = 5.29, p = .022, η2p= .01). Specifically, the low-warmth donor received more charitable credit for a goods donation than a monetary donation (MGoods = 5.75, SD = 1.00 vs. MMoney = 5.37, SD = 1.24; t(304) = –2.96, p = .003 , d = .17). The high-warmth donor, however, received equal (and high) credit for both goods and monetary donations (MGoods = 6.09, SD = .98 vs. MMoney = 6.08, SD = .85; t(291) = –.09, p = .93; figure 5). FIGURE 5 View largeDownload slide STUDY 3A: CHARITABLE CREDIT FOR INDIVIDUAL DONORS AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 5 View largeDownload slide STUDY 3A: CHARITABLE CREDIT FOR INDIVIDUAL DONORS AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Communal Intent The donor warmth manipulation had a significant main effect on communal intent; the high-warmth donor’s intent was viewed as more communal (F(1, 598) = 32.35, p < .001, η2p= .05). Donation type had a nonsignificant main effect (F(1, 598) = 2.50, p = .11). A 2 × 2 ANOVA revealed a marginally significant interaction between donation type (goods/money) and donor warmth (low warmth/high warmth; F(1, 598) = 3.39, p = .066, η2p= .01) on communal intent. For the low-warmth donor, a goods donation produced significantly higher perceptions of communal intent than a monetary donation (MGoods = 5.70, SD = .98 vs. MMoney = 5.42, SD = 1.25; t(304) = –2.20 , p = .027, d = .12). There was no difference in perceptions of communal intent by donation type for the high-warmth donor (MGoods = 6.05, SD = .90 vs. MMoney = 6.02, SD = .90; t(291) = .21, p = .84). Effort/Sacrifice The donor warmth manipulation had a significant main effect on perceived effort and sacrifice; the high-warmth individual was seen as putting in more effort than the low-warmth individual (F(1, 598) = 77.09, p < .001, η2p= .12). Donation type, however, showed no main effect (F(1, 598) = .26, p = .61). There was a nonsignificant interaction between donation type and donor warmth on effort/sacrifice (F(1, 598) = 1.97, p = .16). There was no significant difference in perceptions of effort and sacrifice by donation type for the low-warmth individual (MGoods = 4.43, SD = 1.29 vs. MMoney = 4.25, SD = 1.23; t(304) = –1.26, p = .21), nor was there a significant difference in effort perceptions for the high-warmth individual (MGoods =5.13, SD = 1.10 vs. MMoney = 5.22, SD = 1.00; t(291) = .70, p = .49). Moderated Mediation We conducted a moderated mediation analysis (Hayes 2013, model 8) to test the predicted relationship of donation type by low- or high-warmth donors on charitable credit, simultaneously testing both communal intent and effort/sacrifice as mediators. We tested the significance of both mediators by calculating standardized indirect effects for 5,000 bootstrapped samples and found that the model mediates the effect of donation type on charitable credit (direct effect = .07; 95% CI [–.01, .26]). More specifically, we found that donation type produced an indirect effect of communal intent on charitable credit that was conditional on individual warmth, but found no indirect effect of effort/sacrifice. As hypothesized, inferences about communal intent mediated the effect of donation type on charitable credit for the low-warmth individual (indirect effect = .12; 95% CI [.02, .26]), but not for the high-warmth individual (indirect effect = –.01; 95% CI [–.10, .08]). Effort and sacrifice did not mediate the effect for the low-warmth individual (indirect effect = .05; 95% CI [–.03, .14]), nor for the high-warmth individual (indirect effect = .02; 95% CI [–.09, .04]). The index of moderated mediation was not significant at the 95% level of confidence. However, at the 90% level of confidence, the index of moderated mediation was significant for communal intent (index = –.13; 90% CI [–.27, –.02]) but not for effort/sacrifice (index = –.07; 90% CI [–.16, .01]). We find a similar and significant pattern of results when controlling for both competence and perceived wealth (web appendix A). Please see table 1 for descriptive results for all dependent variables measured in studies 3A and 3B. Table 1 Study 3A and 3B Results Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      † p < .10, *p < .05, **p < .01; these significance notations refer to differences in mean evaluations for monetary donations compared to goods donations with standard deviations in parentheses. A° symbol next to the variable name indicates that there is a significant interaction between donor warmth and donation type on this variable at a p < .05 level. Table 1 Study 3A and 3B Results Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      Study 3A  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  5.37 (1.24)  5.42 (1.25)  4.25 (1.23)  5.52 (.95)    5.32 (.95)  4.83 (.85)  Goods  5.75** (1.00)  5.70* (.98)  4.43 (1.29)  5.64 (.91)    5.26 (1.12)  4.94 (.92)  High warmth  Money  6.08 (.85)  6.05 (.90)  5.22 (1.00)  6.09 (.83)    5.16 (1.06)  3.52 (1.22)  Goods  6.09 (.98)  6.02 (.90)  5.13 (1.10)  5.92 (.92)    5.14 (1.16)  3.84* (1.11)  Study 3B  Donor warmth condition  Donation type  Charitable credit°  Communal intent°  Effort/sacrifice°  Warmth  Purchase likelihood  Competence  Perceived wealth    Low warmth  Money  4.77 (1.21)  4.02 (1.31)  3.48 (1.39)  3.92 (1.41)  4.23 (1.59)      Goods  5.05† (1.23)  4.37* (1.33)  3.94** (1.44)  4.07 (1.36)  4.54† (1.49)      High warmth  Money  5.77 (1.00)  5.15 (1.12)  4.81 (1.17)  5.97 (1.09)  5.81 (1.11)      Goods  5.89 (.86)  5.25 (1.19)  4.89 (1.21)  6.08 (.87)  5.82 (1.08)      † p < .10, *p < .05, **p < .01; these significance notations refer to differences in mean evaluations for monetary donations compared to goods donations with standard deviations in parentheses. A° symbol next to the variable name indicates that there is a significant interaction between donor warmth and donation type on this variable at a p < .05 level. Discussion Results from study 3A replicate the pattern of preferences from studies 2A and 2B for donations of goods, but only when the donor is perceived as low warmth. In contrast to low-warmth donors who receive more credit for goods donations, high-warmth donors receive the same amount of (high) charitable credit regardless of donation type. This pattern provides insight into the role of warmth in the preference for donations of goods (vs. money); when perceived warmth is high, communal intentions are assumed and are less likely to fluctuate based on the donation’s substance. When perceived warmth is low, however, judgments of communal intent are more sensitive to donation type; specifically, low-warmth donors receive a boost in charitable credit for goods donations. STUDY 3B: THE MODERATING ROLE OF WARMTH—CORPORATE DONORS In study 3B, we aim to replicate the patterns from study 3A when studying corporate donors and using a new, more direct, manipulation of warmth. Pretest Methods We first conducted a pretest in which we recruited 120 Mechanical Turk participants, 102 of whom met our inclusion criteria (MAge = 34.96, 56% female); 13 participants were removed for failing to complete the study and five for failing the attention check. In this pretest, we manipulated company warmth using a manipulation similar to the one used in study 1. In a high-warmth condition, participants read “Spades Hardware is a small, friendly hardware store that is always competent and also warm and welcoming toward visitors.” In the low-warmth condition, participants read, “Spades Hardware is a small, corporate hardware store that is always competent though also cold and indifferent toward visitors.” Participants then rated the company on the six items from the perceived warmth scale from previous studies, presented in a randomized order. Pretest Results The descriptions used in the pretest successfully manipulated donor warmth. The high-warmth donor was rated as warmer (MHigh-Warmth = 6.07, SD = .94) than the low-warmth donor (MLow-Warmth = 3.42, SD = 1.46, t(99) = 10.69, p < .001, d = 2.15). Main-Study Methods For study 3B, we recruited 615 Mechanical Turk participants, 565 of whom met our inclusion criteria (MAge = 35.14, 58% female); 34 participants were removed for failing to complete the study and 16 for failing the attention check. The study used a 2 (donor warmth: high, low) × 2 (donation type: $1,000, canned food worth $1,000) between-subjects experimental design. For the low-warmth and high-warmth conditions, Spades Hardware was described in the same way as in the pretest. Then, participants were told that the company donated either “$1,000” or “canned food worth $1,000” to their local food bank. We then asked participants to respond to the items that measured charitable credit (α = .93), communal intent (α = .88), perceived effort/sacrifice (α = .92), and trait warmth (α = .96, a manipulation check). Main-Study Results Warmth Manipulation Check The high-warmth company was perceived as warmer (MHigh-Warmth = 6.02, SD = .99) than the low-warmth company (MLow-Warmth = 3.99, SD = 1.39; t(563) = 20.06, p < .001, d = 1.69). When testing a 2 × 2 ANOVA, we observed a main effect of the warmth manipulation on perceived warmth (F(1, 561) = 402.33, p < .001, η2p= .42). Looking at additional patterns, we did not observe a main effect of donation type on measured warmth (F(1, 561) = 1.82, p = .18), nor did we observe an interaction between donation type and manipulated donor warmth on measured warmth (F(1, 561) = .06, p = .80). These patterns suggest that the warmth construct is a relatively stable trait that is minimally influenced by donation type from any one donation, exhibiting distinct patterns compared to the other constructs under study, such as communal intent and charitable credit. Charitable Credit The warmth manipulation had a significant main effect; the high-warmth company received more charitable credit (F(1, 563) = 102.2 , p < .001, η2p= .15). Donation type did not show a significant main effect (F(1, 561) = .66 , p = .42). Most importantly, the 2 × 2 ANOVA revealed a significant interaction between donation type (money/goods) and donor image (low warmth/high warmth; F(1, 561) = 4.65, p = .03, η2p= .01). In line with the previous findings, when the company was described as low warmth, it received a marginally significant increase in charitable credit for a goods donation compared to a monetary donation (MGoods = 5.05, SD = 1.23; MMoney = 4.77, SD = 1.21; t(274) = 1.92, p = .056, d = .23). However, when the company was described as high warmth, there was no significant difference in charitable credit between donation types (MGoods = 5.77, SD = 1.00 vs. MMoney = 5.89, SD = .86; t(287) = –1.08, p = .28; figure 6). FIGURE 6 View largeDownload slide EXPERIMENT 3B: CHARITABLE CREDIT AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 6 View largeDownload slide EXPERIMENT 3B: CHARITABLE CREDIT AS A FUNCTION OF DONOR WARMTH AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Communal Intent The donor manipulation had a significant main effect on communal intent; the high-warmth company’s donation was perceived as more communal (F(1, 561) = 93.86, p < .001, η2p= .14). Donation type also had a significant main effect; goods donations were seen as more communal than donations of money (F(1, 561) = 5.03, p = .025, η2p= .01). In this study we found no significant interaction between donation type (money/goods) and donor warmth (low warmth/high warmth; F(1, 561) = 1.66, p = .198). However, for the low-warmth company, a donation of goods produced significantly higher perceptions of communal intent (MGoods = 4.37, SD = 1.33 vs. MMoney = 4.02, SD = 1.31; t(274) = 2.21, p = .028, d = .27) and for the high-warmth company, there was no difference in perceptions of communal intent by donation type (MGoods = 5.25, SD = 1.19 vs. MMoney = 5.15, SD = 1.12; t(287) = .75, p = .45). Effort/Sacrifice The donor warmth manipulation had a significant main effect on perceived effort and sacrifice; the high-warmth company was seen as putting in more effort than the low-warmth company (F(1, 561) = 108.70, p < .001, η2p= .16). Furthermore, there was a marginally significant main effect of donation type; the donation of goods was viewed as marginally more effortful than the donation of money (F(1, 561) = 2.80, p = .095). We also observed a significant interaction between donation type and donor warmth (F(1, 561) = 5.90, p < .02, η2p= .01). There was a significant difference in perceptions of effort and sacrifice by donation type for the low-warmth company (MGoods = 3.94, SD = 1.44 vs. MMoney = 3.48, SD = 1.39; t(274) = 2.71, p = .007, d = .33), but not for the high-warmth company (MGoods = 4.81, SD = 1.17 vs. MMoney = 4.89, SD = 1.21; t(287) = .55, p = .58). Moderated Mediation We conducted a moderated mediation analysis (Hayes 2013, model 8) to test the predicted relationship of donation type by low- or high-warmth companies on charitable credit received, with communal intent and effort/sacrifice simultaneously tested as mediators. By calculating standardized indirect effects for 5,000 bootstrapped samples, we found an indirect effect of both communal intent and effort/sacrifice conditional on company warmth when tested simultaneously. As hypothesized, inferences about communal intent mediated the effect of donation type on charitable credit for the low-warmth company (indirect effect = .12; 95% CI [.02, .22]), but not the high-warmth company (indirect effect = .03; 95% CI [–.05, .12]). Likewise, effort and sacrifice mediated the effect for the low-warmth company (indirect effect = .16; 95% CI [.05, .29]), but not for the high-warmth company (indirect effect = –.03; 95% CI [–.13, .07]). The index of moderated mediation was not significant at the 95% level of confidence. However, at the 90% level of confidence, the index of moderated mediation was significant for both communal intent (index = –.08; 90% CI [–.19, –.02]) and for effort/sacrifice (index = –.19; 90% CI [–.32, –.06]). Purchase Likelihood A similar, although nonsignificant pattern emerged for purchase likelihood. For the low-warmth company, participants reported a marginally significant increase in purchase likelihood when they read about a goods donation (MGoods = 4.54, SD = 1.49) compared to a monetary donation (MMoney = 4.23, SD = 1.59; t(274) = 1.66, p = .097). However, donation type had no effect on purchase likelihood for the high-warmth company (MGoods = 5.82, SD = 1.08 vs. MMoney= 5.81, SD = 1.11; t(287) = .04, p = .97). While the pattern was consistent with the charitable credit dependent variable, the interaction between donation type and donor warmth was not significant (F(1, 561) = 1.80, p = .18). Discussion Results from study 3B replicate the pattern observed in study 3A; people prefer donations of goods from low-warmth donors, but do not show this preference for high-warmth donors. Judgments of communal intentions mediate this pattern, and in this study, effort and sacrifice also simultaneously mediated this effect. In two additional studies (for full results see web appendix C, studies 6 and 7), we replicate these patterns. In appendix study 6, we manipulate perceived warmth by using two real-world companies: a company that is perceived as relatively low in warmth, Pepsi Co., and a company selling similar products that is perceived as relatively high in warmth, Bolthouse Farms. There was a significant interaction between donor warmth and donation type on charitable credit; Pepsi (low warmth) received more charitable credit when donating goods compared to money, whereas Bolthouse Farms (high warmth) received equal (high) amounts of charitable credit for both donation types. We found the same pattern for purchase likelihood. Due to the many potential differences besides perceived warmth between the individuals in study 3A (nursery school teacher and corporate manager), in appendix study 7, we also tested an additional manipulation of warmth that described an individual donor as either warm or cold (in a manipulation similar to study 3B). We found the same pattern of results using this manipulation. In this study, we also found that when low-warmth donors donate goods, they are viewed as having greater communal intent and receive greater charitable credit; however, goods donations do not similarly improve judgments of warmth. This pattern fits with the notion that warmth is a stable construct that varies minimally due to subtle changes or contextual factors (Kenworthy and Tausch 2008). We do predict, however, that consistently donating goods and/or sending other sustained signals of communal intent may improve perceptions of a donor’s trait warmth over time. STUDY 4: MONETARY DONATIONS CAN BE COMMUNAL The image benefits of goods donations from low-warmth donors documented thus far is somewhat unfortunate because for many charities, monetary donations are superior to equivalent in-kind or goods donations. Monetary contributions are typically preferred because cash donations provide charities with the flexibility to purchase exactly what they need when they need it, reducing waste from unneeded or untimely goods donations (Charity Navigator 2017; Conan 2011; USAID 2017). In study 4, we test whether framing a monetary donation as communal increases charitable credit for low-warmth donors in an analogous way to goods donations. We specifically test whether framing a monetary donation as a gift increases perceptions of communal intent, and subsequently, charitable credit. As opposed to other commodity exchanges, gifting is valued as a symbolic gesture of caring and commitment (Belk and Coon 1993). Therefore, if money can be described as a gift in a compelling way, it has the potential to be seen as more communal and creditworthy. Methods We recruited 650 Mechanical Turk participants, 567 of whom met our inclusion criteria (MAge = 34.24, 61.85% female); 56 participants were removed for failing to complete the study and 27 for failing the attention check. The study included a 2 (donation type: goods vs. money) × 2 (communal frame: control vs. communal [gift]) between-subjects experimental design. In all conditions, participants read about the low-warmth donor Spades Hardware. They read that “Spades Hardware is a large corporation that sells home improvement goods.” They then read either the 1) money condition: “This past weekend, Spades Hardware donated $10,000 to a humanitarian aid charity. (The charity purchased medical supplies with the $10,000.)” or 2) goods condition: “This past weekend, Spades Hardware donated medical supplies (worth a total of $10,000) to a humanitarian aid charity.” In the communal frame conditions, participants also read that “The company carefully packaged the gift of $10,000 [medical supplies] and hand-delivered it.” (We note that in this study, even in the monetary donation conditions we explained that the donation was ultimately used for medical supplies to hold information about donation use constant across conditions.) As in previous studies, we measured charitable credit (α = .89), communal intent (α = .80), perceived effort/sacrifice (α = .85), and trait warmth (α = .94).The order of all measures was counterbalanced. Results Charitable Credit The framing manipulation had a significant main effect; the company whose donation was framed as a gift received more charitable credit than did the company whose donation was not framed as a gift (F(1, 562) = 7.65, p < .01, η2p= .01). There was also a significant main effect for donation type; goods donations increased charitable credit (F(1, 562) = 6.81 , p < .01, η2p= .01). Most importantly, the 2 × 2 analysis of variance revealed a significant interaction between donation type (goods/money) and the communal frame (F(1, 562) = 18.01, p < .001, η2p= .03). In the control condition, the low-warmth donor received more charitable credit for a goods donation (MGoods = 5.85, SD = .65) than an equivalent monetary donation (MMoney = 5.36, SD = .85; t(277) = 5.33, p < .001, d = .64). However, when the donation was framed communally, the low-warmth donor received equal credit for both donation types (MGoods = 5.73, SD = .95 vs. MMoney = 5.86, SD = .84; t(285) = 1.20, p = .23; figure 7). FIGURE 7 View largeDownload slide EXPERIMENT 4: CHARITABLE CREDIT AS A FUNCTION OF COMMUNAL FRAMING AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. FIGURE 7 View largeDownload slide EXPERIMENT 4: CHARITABLE CREDIT AS A FUNCTION OF COMMUNAL FRAMING AND DONATION TYPE NOTE.—Error bars represent standard errors of the mean. Communal Intent The communal frame had a significant main effect on communal intent; overall the donation described as a gift was viewed as more communal than the control condition counterpart (F(1, 561) = 23.36, p < .001, η2p= .04). Donation type had a marginally significant effect; goods donations were seen as marginally more communal than donations of money (F(1, 561) = 2.91, p = .09, η2p= .01). There was also a significant interaction between donation type (goods/money) and the communal frame (F(1, 561) = 4.05, p < .05, η2p= .01). In the control frame, a donation of goods versus money produced significantly higher perceptions of communal intent (MGoods = 4.72, SD = .88 vs. MMoney = 4.39, SD = 1.02; t(275) = 2.94, p < .01, d = .35). There was no difference in perceptions of communal intent in the communal frame condition (MGoods = 4.97, SD = 1.19 vs. MMoney = 4.99, SD = 1.13; t(286) = –.18, p = .86). Effort/Sacrifice The gift-giving manipulation had a significant main effect on perceived effort and sacrifice; the communal frame increased perceptions of effort and sacrifice compared to the control frame (F(1, 560) = 18.78, p < .001, η2p= .03). Donation type also showed a significant main effect; donations of goods were viewed as more effortful (F(1, 560) = 5.28, p = .022, η2p= .01). However, we did not observe a significant interaction between donation type and the communal frame (F(1, 560) = .13, p = .72). There was no difference in perceptions of effort and sacrifice by donation type for the control condition (MGoods = 4.67, SD = .99 vs. MMoney = 4.49, SD = 1.11; t(275) = 1.44, p = .15), and there was a marginally significant difference in effort perceptions for the communal-frame (gift) condition (MGoods = 5.10, SD = 1.16 vs. MMoney = 4.86, SD = 1.21; t(285) = 1.71, p = .09). Purchase Likelihood There was a significant main effect of the donor manipulation on purchase likelihood; overall participants were more likely to purchase from the company whose donation was framed communally (F(1, 555) = 4.60, p =.03, η2p= .01). Donation type also had a significant main effect; donations of goods led to higher purchase likelihood (F(1, 555) = 7.24, p < .01, η2p= .01). There was a marginally significant interaction between donation type and the gift-giving manipulation on purchase likelihood (F(1, 555) = 3.53, p = .06, η2p= .01). In the control frame condition, participants were more likely to purchase from Spades Hardware after a donation of goods (MGoods = 5.60, SD = .91) versus money (MMoney = 5.20, SD = 1.05; t(271) = 3.40, p < .001, d = .41). By contrast, in the communal frame condition, we found no difference in purchase intentions based on donation type (MGoods = 5.60, SD = 1.06 vs. MMoney = 5.56, SD = 1.11; t(284) = .30, p = .76). Measured Warmth We found a main effect of the communal frame on perceived warmth (F(1, 566) = 15.02, p < .001, η2p= .03), but a nonsignificant main effect of donation type on warmth (F(1, 566) = .167, p = .68). There was no significant interaction between communal framing and donation type on measured warmth (F(1, 566) = 1.96, p = .16). Moderated Mediation We conducted a moderated mediation analysis (Hayes 2013, model 8) to test the predicted relationship of donation type by companies who do or do not hand-deliver their gifts on charitable credit received, with communal intent and effort/sacrifice as mediators. We simultaneously tested the significance of both mediators by calculating standardized indirect effects for 5,000 bootstrapped samples and found that the model mediates the effect of donation type on charitable credit (direct effect = .08; 95% CI [–.04, .20]). We found that donation type produced an indirect effect of communal intent conditional on communal frame. As hypothesized, inferences about communal intent mediated the effect of donation type on charitable credit for the control condition (indirect effect = .09; 95% CI [.03, .16]), but not for the communal-frame (gift) condition (indirect effect = –.01; 95% CI [–.09, .06]). Effort and sacrifice did not mediate the effect for the control condition (indirect effect = .05; 95% CI [–.02, .12]), nor for the gift condition (indirect effect = .07; 95% CI [–.004, .14]). The index of moderated mediation was significant for communal intent (index = –.10; 95% CI [–.21, –.01]), but not significant for effort/sacrifice (index = .02; 95% CI [–.09, .11]). Discussion In study 4, we found once again that a low-warmth donor (Spades Hardware) received more credit for a donation of goods than a donation of money. However, when the donations were framed as communal (i.e., a gift), both donations were perceived to be motivated by communal intentions and therefore creditworthy. By increasing the perceived communality of the monetary donation, the low-warmth donor was able to receive high levels of charitable credit for making a cash donation. GENERAL DISCUSSION This research finds that donors receive charitable credit based on who they are and what they give. High-warmth donors receive more credit for being generous than low-warmth donors because people assume high-warmth donors have better intentions. However, low-warmth donors can boost the charitable credit they receive by donating goods. This article documents a novel interaction between donor traits (high vs. low warmth) and donation types (goods vs. money) and is among the first to study goods donations. The conceptualization that trait warmth triggers strong assumptions about the nature of a donor’s intent generates numerous predictions about the level of credit donors will receive. For example, although we did not find a consistent effect of donation type (goods vs. money) on effort perceptions, effort likely has independent effects on charitable credit when varied through other means. If people assume that high-warmth donors have good intentions, then, similar to donation type, donation effort may minimally impact charitable credit for high-warmth donors; high-warmth donors are likely to receive high levels of charitable credit for any magnitude of donation effort. In contrast, low-warmth donors, whose intentions are more subject to skepticism, may receive minimal credit for a low-effort donation, but receive more credit for sending a stronger signal of good intentions via a high-effort donation. These predictions are analogous to other findings in the literature in which individuals with different traits are judged differently for the same act. For example, prior research finds that donors with good reputations (i.e., social workers) are viewed as equally altruistic whether they brag about their good deed or not, whereas donors with less altruistic reputations (i.e., investment bankers) are viewed as even less altruistic when they brag about a good deed (Berman et al. 2015, study 2). Future research should further examine these and related patterns regarding donor traits and assumptions about communal intent. Although little research exists that examines outcomes associated with goods donations, there is some that distinguishes donations of money versus time (Liu and Aaker 2008; Reed et al. 2007; Macdonnell and White 2015). For example, research finds that people are happier, and therefore likely to make larger charitable donations, when they think about time rather than money (Liu and Aaker 2008), and that when the value of the two resources is equivalent, individuals view donating time as more moral than donating money (Reed et al. 2007). Assuming that time, like goods, signals communal rather than market-exchange intent, we would expect donations of time from low-warmth donors to appear more charitable than equivalent donations of money. One interesting, and perplexing, implication of the current findings is that for image reasons, low-warmth donors, such as most corporations, may benefit from donating goods rather than money, even though charities can often do more good with monetary donations (Conan 2011; USAID 2017). In fact, in web appendix C, study 8, we find that participants appear aware that nonprofits generally prefer monetary donations, and yet still show more favorable evaluations of donors who make goods donations. For low-warmth donors, such as most for-profit companies, who wish to maximize both their impact and credit received, we propose two potential solutions. First, as in study 4, low-warmth donors can describe a monetary donation as a gift, and thus frame it more communally. Alternatively, low-warmth donors can coordinate closely with charities to identify their most pressing tangible needs, and fund those needs quickly and directly via goods donations. Indeed, working closely with nonprofits to identify and meet their greatest needs may serve as an additional signal to consumers of a donor’s communal intentions. In summary, despite the fact that monetary donations are the most frequent donation type (USA.gov 2012), and the type that charities often prefer, we find that low-warmth donors who donate goods receive more credit for their generosity than those who make equivalent monetary donations. We conclude that low-warmth donors, including many corporations (Aaker et al. 2010), aiming to maximize both charitable credit and actual impact may benefit from spending their philanthropic funds on donations of goods that are coordinated with a charity’s needs, or alternatively, describing a monetary donation communally. Future research should continue to explore our differing judgments of donors based on both donation type and donor traits and the implications for charitable credit and sustained giving. DATA COLLECTION INFORMATION The first and second authors conducted the study in the article’s introduction on Amazon Mechanical Turk in March 2017; study 1 on Amazon Mechanical Turk in August 2017; study 2A on Amazon Mechanical Turk in May 2017; study 2B on Amazon Mechanical Turk in February 2017; study 3A on Amazon Mechanical Turk in October 2017; study 3B on Amazon Mechanical Turk in December 2016; and study 4 on Amazon Mechanical Turk in June 2017. The first author analyzed these data with help from the second author. Web appendix C studies: the first and second authors conducted appendix study 1 on Amazon Mechanical Turk in September 2016; appendix study 2 on Amazon Mechanical Turk in September 2017; appendix study 3 on Amazon Mechanical Turk in February 2015; appendix study 4 on Amazon Mechanical Turk in October 2016; appendix study 5 on Amazon Mechanical Turk in January 2015; appendix study 6 on Amazon Mechanical Turk in November 2015; and appendix studies 7 and 8 on Amazon Mechanical Turk in January 2017. The first author analyzed these data with help from the second author. The authors wish to thank Lucy Gao and Nicole Cooper for research assistance. They would also like to thank Alix Barasch and the Washington University Journal Club for helpful feedback on an earlier draft of the article. REFERENCES Aaker Jennifer L., Vohs Kathleen D., Mogilner Cassie ( 2010), “Non-Profits Are Seen as Warm and For-Profits as Competent: Firm Stereotypes Matter,” Journal of Consumer Research , 37 ( 2), 224– 37. Google Scholar CrossRef Search ADS   Aggarwal Pankaj ( 2004), “The Effects of Brand Relationship Norms on Consumer Attitudes and Behavior,” Journal of Consumer Research , 31 ( 1), 87– 101. Google Scholar CrossRef Search ADS   Aggarwal Pankaj, Law Sharmistha ( 2005), “Role of Relationship Norms in Processing Brand Information,” Journal of Consumer Research , 32 ( 3), 453– 64. Google Scholar CrossRef Search ADS   Andrzejewski Susan A., Mooney Emily C. ( 2016), “Service with a Smile: Does the Type of Smile Matter?” Journal of Retailing and Consumer Services , 29 ( March), 135– 41. Google Scholar CrossRef Search ADS   Ariely Dan, Bracha Anat, Meier Stephan ( 2007), “Doing Good or Doing Well? Image Motivation and Monetary Incentives in Behaving Prosocially,” American Economic Review , 99 ( 1), 544– 55. Google Scholar CrossRef Search ADS   Barasch Alixandra, Levine Emma E., Berman Jonathan Z., Small Deborah A. ( 2014), “Selfish or Selfless? On the Signal Value of Emotion in Altruistic Behavior,” Journal of Personality and Social Psychology , 107 ( 3), 393– 413. Google Scholar CrossRef Search ADS PubMed  BCLC, “Hurricane Sandy—Corporate Aid Tracker,” U.S. Chamber of Commerce Foundation, https://www.uschamberfoundation.org/corporate-citizenship-center/hurricane-sandy-corporate-aid-tracker. Bechwati Nada N., Xia Lan ( 2003), “Do Computers Sweat? The Impact of Perceived Effort of Online Decision Aids on Consumer’s Satisfaction with the Decision Process,” Journal of Consumer Psychology , 13 ( 1), 139– 48. Google Scholar CrossRef Search ADS   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   Berman Jonathan Z., Levine Emma E., Barasch Alixandra, Small Deborah A. ( 2015), “The Braggart’s Dilemma: On the Social Rewards and Penalties of Advertising Prosocial Behavior,” Journal of Marketing Research , 52 ( 1), 90– 104. Google Scholar CrossRef Search ADS   Casciaro Tiziana, Lobo Miguel Sousa ( 2008), “When Competence Is Irrelevant: The Role of Interpersonal Affect in Task-Related Ties,” Administrative Science Quarterly , 53 ( 4), 655– 84. Google Scholar CrossRef Search ADS   Charity Navigator ( 2017), “Guide to Donating Noncash Items,” https://www.charitynavigator.org/index.cfm?bay=content.view&cpid=335. Cheal David ( 1987), “Showing Them You Love Them: Gift Giving and the Dialectic of Intimacy,” Sociological Review , 35 ( 1), 150– 69. Google Scholar CrossRef Search ADS   Clark Margaret S., Mills Judson R. ( 1979), “Interpersonal Attraction in Exchange and Communal Relationships,” Journal of Personality and Social Psychology , 37 ( 1), 12– 24. Google Scholar CrossRef Search ADS   Clark Margaret S., Mills Judson R. ( 2011), “A Theory of Communal (and Exchange) Relationships,” in Handbook of Theories of Social Psychology , Vol. 2, ed. Van Lange Paul A. M., Kruglanski Arie W., Higgins E. Tory, Thousand Oaks, CA: Sage, 232– 50. Conan Neal ( 2011), “A Case for Cash Donations, Instead of Cans,” NPR, http://www.npr.org/2011/11/22/142661882/a-case-for-cash-donations-instead-of-cans. Cuddy Amy J. C., Fiske Susan T., Glick Peter ( 2004), “When Professionals Become Mothers, Warmth Doesn’t Cut the Ice,” Journal of Social Issues , 60 ( 4), 701– 18. Google Scholar CrossRef Search ADS   Cuddy Amy J. C., Fiske Susan T., Glick Peter. ( 2008), “Warmth and Competence as Universal Dimensions of Social Perception: The Stereotype Content Model and the BIAS Map,” Advances in Experimental Social Psychology , 40, 61– 149. Google Scholar CrossRef Search ADS   Cuddy Amy J. C., Glick Peter, Beninger Anna ( 2011), “The Dynamics of Warmth and Competenc Judgments, and their Outcomes in Organizations,” Research in Organizational Behavior , 31, 73– 98. Google Scholar CrossRef Search ADS   Douglas Mary, Isherwood Baron ( 1979), “The World of Goods,”  New York: Basic. Google Scholar CrossRef Search ADS   Ellingsen Tore, Johannesson Magnus ( 2011), “Conspicuous Generosity,” Journal of Public Economics , 95 ( 9–10), 1131– 43. Google Scholar CrossRef Search ADS   Fiske Susan T., Cuddy Amy J. C., Glick Peter ( 2007), “Universal Dimensions of Social Cognition: Warmth and Competence,” Trends in Cognitive Sciences , 11 ( 2), 77– 83. Google Scholar CrossRef Search ADS PubMed  Fiske Susan T., Cuddy Amy J. C., Glick Peter, Xu Jun ( 2002), “A Model of (Often Mixed) Stereotype Content: Competence and Warmth Respectively Follow from Perceived Status and Competition,” Journal of Personality and Social Psychology , 82 ( 6), 878– 902. Google Scholar CrossRef Search ADS PubMed  Flynn Francis J. ( 2003), “How Much Should I Give and How Often? The Effects of Generosity and Frequency of Favor Exchange on Social Status and Productivity,” Academy of Management Journal , 46 ( 5), 539– 53. Google Scholar CrossRef Search ADS   Flynn Francis J., Reagans Ray E., Amanatullah Emily T., Ames Daniel R. ( 2006), “Helping One’s Way to the Top: Self-Monitors Achieve Status by Helping Others and Knowing Who Helps Whom,” Journal of Personality and Social Psychology , 91 ( 6), 1123– 37. Google Scholar CrossRef Search ADS PubMed  Fournier Susan, Alvarez Claudio ( 2012), “Brands as Relationship Partners: Warmth, Competence, and In-Between,” Journal of Consumer Psychology , 22 ( 2), 177– 85. Google Scholar CrossRef Search ADS   Frey Bruno S., Goette Lorenz ( 1999), Does Pay Motivate Volunteers ? Vol. 7, Zurich, Switzerland: Institute for Empirical Research in Economics, University of Zurich. Gasiorowska Agata, Chaplin Lan N., Zaleskiewicz Tomasz, Wygrab Sandra, Vohs Kathleen D. ( 2016), “Money Cues Increase Agency and Decrease Prosociality among Children: Early Signs of Market-Mode Behaviors,” Psychological Science , 27 ( 3), 331– 44. Google Scholar CrossRef Search ADS PubMed  Goette Lorenz, Stutzer Alois, Frey Beat M. ( 2010), “Prosocial Motivation and Blood Donations: A Survey of the Empirical Literature,” Transfusion Medicine and Hemotherapy , 37 ( 3), 149– 54. Google Scholar CrossRef Search ADS PubMed  Goodman Joseph K., Cryder Cynthia E., Cheema Amar ( 2013), “Data Collection in a Flat World: The Strengths and Weaknesses of Mechanical Turk Samples,” Journal of Behavioral Decision Making , 26 ( 3), 213– 24. Google Scholar CrossRef Search ADS   Graham Melanie ( 2012), “Williams Elementary 5th Graders Help Hurricane Sandy Relief Efforts,” Newton Patch , December 9, https://patch.com/massachusetts/newton/wlliams-elementary-5th-graders-help-hurricane-sandy-rbfba38878d. Hayes Andrew F. ( 2013), Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-based Approach , New York: Guilford. Heyman James, Ariely Dan ( 2004), “Effort for Payment: A Tale of Two Markets,” Psychological Science , 15 ( 11), 787– 93. Google Scholar CrossRef Search ADS PubMed  Imhoff Roland, Woelki Jonas, Hanke Sebastian, Dotsch Ron ( 2013), “Warmth and Competence in Your Face! Visual Encoding of Stereotype Content,” Frontiers in Psychology , 4, 386. Google Scholar CrossRef Search ADS PubMed  Interbrand ( 2017), “Best Global Brands 2016 Rankings,” http://interbrand.com/best-brands/best-global-brands/2016/ranking/. Jiang Yuwei, Chen Zhansheng, Wyer Robert S.Jr. ( 2014), “Impact of Money on Emotional Expression,” Journal of Experimental Psychology , 55 ( November), 228– 33. Kenworthy Jared B., Tausch Nicole ( 2008), “Expectations about the Accuracy and Stability of Warmth versus Competence Traits: An Intergroup Analysis,” European Journal of Social Psychology , 38 ( 7), 1121– 29. Google Scholar CrossRef Search ADS   Kervyn N., Fiske Susan T., Malone Chris ( 2012), “Brands as Intentional Agents Framework: How Perceived Intentions and Ability Can Map Brand Perception,” Journal of Consumer Psychology , 22 ( 2), 166– 76. Google Scholar CrossRef Search ADS   Koschate-Fischer Nicole, Stefan Isabel V., Hoyer Wayne D. ( 2012), “Willingness to Pay for Cause-Related Marketing: The Impact of Donation Amount and Moderating Effects,” Journal of Marketing Research , 49 ( 6), 910– 27. Google Scholar CrossRef Search ADS   Lacetera Nicola, Macis Mario, Slonim Robert ( 2013), “Economic Rewards to Motivate Blood Donations,” Science , 340 ( 6135), 927– 8. Google Scholar CrossRef Search ADS PubMed  Lawrence Steven, Ronna Brown, Michael Hamill Remaley, Nina Stack ( 2014), Philanthropy and Hurricane Sandy: A Report on the Foundation and Corporate Response , New York: Foundation Center, Center for Disaster Philanthropy, Philanthropy New York, and Council of New Jersey Grantmakers. Lin-Healy Fern, Small Deborah A. ( 2012), “Cheapened Altruism: Discounting Personally Affected Prosocial Actors,” Organizational Behavior and Human Decision Processes , 117 ( 2), 269– 74. Google Scholar CrossRef Search ADS   Lin-Healy Fern, Small Deborah A. ( 2013), “Nice Guys Finish Last and Guys in Last Are Nice: The Clash between Doing Well and Doing Good,” Social Psychological and Personality Science , 4 ( 6), 692– 98. Google Scholar CrossRef Search ADS   Liu Wendy, Aaker Jennifer ( 2008), “The Happiness of Giving: The Time-Ask Effect,” Journal of Consumer Research , 35 ( 3), 543– 57. Google Scholar CrossRef Search ADS   Macdonnell Rhiannon, White Katherine ( 2015), “How Construals of Money versus Time Impact Consumer Charitable Giving,” Journal of Consumer Research , 42 ( 4), 551– 63. McGraw Peter A., Tetlock Philip E. ( 2005), “Taboo Trade-Offs, Relational Framing, and the Acceptability of Exchanges,” Journal of Consumer Psychology , 15 ( 1), 2– 15. Google Scholar CrossRef Search ADS   Newman George E., Cain Daylian M. ( 2014), “Tainted Altruism When Doing Some Good Is Evaluated as Worse than Doing No Good at All,” Psychological Science , 25 ( 3), 648– 55. Google Scholar CrossRef Search ADS PubMed  Oppenheimer Daniel M., Meyvis Tom, Davidenko Nicolas ( 2009), “Instructional Manipulation Checks: Detecting Satisficing to Increase Statistical Power,” Journal of Experimental Social Psychology , 45 ( 4), 867– 72. Google Scholar CrossRef Search ADS   Rampell Catherine ( 2012), “Volunteers Flock to Disaster Areas, Overwhelming City Relief Centers,” New York Times , November 4, http://www.nytimes.com/2012/11/05/nyregion/volunteers-and-donations-flock-to-areas-affected-by-hurricane-sandy.html. Reed AmericusII, Aquino Karl, Levy Eric ( 2007), “Moral Identity and Judgments of Charitable Behaviors,” Journal of Marketing , 71 ( January), 178– 93. Google Scholar CrossRef Search ADS   Simmons Joseph P., Nelson Leif D., Simonsohn Uri ( 2012), “A 21-Word Solution,” Dialogue , 26, 4– 12. USAID ( 2017), “Why Cash Is Best,” USAID Center for International Disaster Information CIDI, https://www.cidi.org/how-disaster-relief-works/monetary-contributions-workbestwhy-cash-is-best/#.Wh2rOGhSxPY. USA.gov ( 2012), “Donating to Charity,” USA.gov, www.usa.gov/donate-to-charity. Vlachos Pavlos A., Tsamakos Argiris, Vrechopoulos Adam P., Avramidis Panagiotis K. ( 2009), “Corporate Social Responsibility: Attributions, Loyalty, and the Mediating Role of Trust,” Journal of the Academy of Marketing Science , 37 ( 2), 170– 80. Google Scholar CrossRef Search ADS   Wagner Tillmann, Lutz Richard J., Weitz Barton A. ( 2009), “Corporate Hypocrisy: Overcoming the Threat of Inconsistent Corporate Social Responsibility Perceptions,” Journal of Marketing , 73 ( 6), 77– 91. Google Scholar CrossRef Search ADS   Webley Paul, Lea S. E., Portalska R. ( 1983), “The Unacceptability of Money as a Gift,” Journal of Economic Psychology , 4 ( 3), 223– 38. Google Scholar CrossRef Search ADS   White Martha C. ( 2015), “Is It Better to Donate Food or Money around the Holidays?” Time , http://time.com/money/4123363/donations-food-money/. Wojciszke Bogdan, Abele Andrea E. ( 2008), “The Primacy of Communion over Agency and Its Reversals in Evaluations,” European Journal of Social Psychology , 38 ( 7), 1139– 47. Google Scholar CrossRef Search ADS   Wojciszke Bogdan, Bazinska Rosa, Jaworski Marcin ( 1998), “On the Dominance of Moral Categories in Impression Formation,” Personality and Social Psychology Bulletin , 24 ( 12), 1251– 63. Google Scholar CrossRef Search ADS   Yoon Yeosun, Gürhan-Canli Zeynep, Schwarz Norbert ( 2006), “The Effect of Corporate Social Responsibility (CSR) Activities on Companies with Bad Reputations,” Journal of Consumer Psychology , 16 ( 4), 377– 90. Google Scholar CrossRef Search ADS   © The Author 2017. 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

Journal

Journal of Consumer ResearchOxford University Press

Published: Dec 28, 2017

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