Sage Journals: Discover world-class research

Abstract

To enhance their effectiveness, nonprofit fundraisers may wish to harness the power of identification. Informed by Social Identity Theory and Charitable Triad Theory, we meta-analyzed 40 years of research on social identification and charitable giving to quantify the overall relationship and conducted meta-regressions to investigate moderators. Across 109 effect sizes drawn from 89,570 participants, we found a medium-sized relationship (r = .29). Identification with other donors (r = .23), beneficiaries (r = .24), and fundraisers (r = .36) were all positively associated with giving. Strength of identification (r = .32) was more strongly associated with giving than was shared identity (i.e., in-group vs out-group target; r = .15). Effects were smaller for actual behavior (r = .20) than for self-reported giving (r = .33) and were only found when giving was mediated through charities (r = .34) but not when giving directly to individuals (r = .04). We include practical recommendations for ways that fundraisers can effectively leverage the power of identification in recruitment campaigns, copywriting, and selection of spokespeople.

Keywords

charitable giving fundraising social identification meta-analysis

Charitable donations—defined as voluntary contributions of money that benefit people outside the giver’s family (Bekkers & Wiepking, 2011)—now exceed $470 billion annually in the United States alone (Giving USA, 2021). To support their essential charity work, nonprofit marketing teams must fundraise effectively, using their resources as wisely as possible to maximize funding. To be effective, they need to know what motivates people to give and apply that understanding to their campaigns.

One concept that has gained significant traction in recent years is the notion that identities and identification processes are critical for understanding charitable behavior (Aaker & Akutsu, 2009; Levine & Manning, 2013; Zagefka & James, 2015). Indeed, a range of studies have examined empirically the power of identities and social identification within charitable giving contexts (e.g., Chapman et al., 2020; Kessler & Milkman, 2018; Puntoni et al., 2011; Shang et al., 2008; Winterich et al., 2013). Articles and blog posts about the importance of understanding donor identities are also common within nonprofit industry media (e.g., Ahern, 2020; Charity Link, 2022), suggesting the topic is also being taken seriously within professional fundraising communities.

Given this surging attention, and the fact that social identification has captured the zeitgeist both within and outside academia, it is particularly important to present a clear-eyed overview of the extent to which—and the conditions under which—social identification is associated with charitable giving. Meta-analyses, which provide bird’s-eye overviews of large amounts of literature, are one way to provide this clarity. Meta-analyses are particularly useful when existing literatures have traditionally been fragmented across disciplines, as they have been in the field of charitable giving (with studies published across, for example, nonprofit studies, marketing, psychology, economics, sociology, and management).

Another advantage of meta-analyses is their ability to delineate important boundary conditions which are not feasible to test within individual studies. For example, the current literature obscures a critical theoretical question: identification with whom? Charitable Triad Theory (Chapman et al., 2022) argues that three actors—donors, beneficiaries, and fundraisers—are critical to understanding charitable giving contexts, and identification can be measured in relation to any one of these three actors. However, existing studies are not well-equipped to referee whether the effects of identification are particularly strong (or weak) when applied to each of these actors. Knowing the answer to this question is important for theory development, as well as for practitioners who need to know how to best direct resources and marketing efforts.

The vast canvas provided by a meta-analysis can also help answer other important questions, such as whether identification is a particularly strong correlate of giving for some pockets of the population compared to others, or whether identification is differentially associated with self-reported donation amounts (which are prone to distortion and bias) versus objective donation amounts. Answering these questions helps provide practical guidance for nonprofit marketers about when, how, and for whom to leverage the power of social identification in their fundraising appeals and marketing communications.

Meta-analyses can also help nuance between theoretical distinctions that are sometimes glossed over in the literature. For example, as outlined in the next section, Social Identity Theory (e.g., Tajfel, 1978) distinguishes between two related but separate notions: strength of identification and shared identity. In the literature, these concepts are sometimes treated as interchangeable, and together, they are used to speak to the power of identification in promoting giving. However, there are theoretical reasons to believe that the effects of one may be stronger than the other (elaborated below), and subjecting this prediction to empirical rigor adds both theoretical nuance and practically useful information for nonprofit managers and fundraisers.

In this article, we meta-analyze data from 74 independent samples recruited in 15 countries to understand when social identification is (or is not) associated with giving. We consider the strength of the relationship between giving and identification and examine empirically—and for the first time—the strength of the relationship for identification with each actor in the Charitable Triad. We also offer practical guidance for nonprofit marketers about when and how to leverage the power of identification in their fundraising appeals and marketing communications. Before doing so, however, we outline relevant theories that are used to frame our hypotheses and research questions.

Social Identity Theory

According to Social Identity Theory (e.g., Hornsey, 2008; Tajfel, 1978), people have a range of both individual (personal) and group-based (social) identities that may be relevant or salient in different contexts. The social groups that a person belongs to, in turn, provide meaningful information that helps shape their attitudes and behaviors. In this paper, we follow the Social Identity Theory tradition and use the phrase “social identification” to refer to both shared identity (focused on the group membership of targets relative to participants, with a dichotomous in-group vs out-group distinction) and strength of identification (a distinction that occurs on a graded scale among in-group members). These constructs are theoretically related and are sometimes used interchangeably in social identity research on charitable giving. However, these are two qualitatively distinct concepts that may have different relationships with charitable giving.

The extent to which shared group membership shapes attitudes and behaviors depends on whether the identity is a valued and important part of a person’s self-definition (i.e., the strength of identification). Strength of identification is typically construed as reflecting several distinct but related dimensions, including the extent to which the group membership is thought about often and is subjectively important to the individual’s self-definition; the extent to which positive feelings arise because of their group membership; and perceptions of similarity with other group members (Cameron, 2004). Importantly for the current analysis, it is presumed that strongly identifying with a target will make you more positively predisposed to that target: one will have more congenial attitudes, beliefs, attributions, and behaviors toward people and organizations to which we feel strongly identified.

In addition to strength of identification, social identification can influence attitudes and behavior through mere shared identity. For example, one robust phenomenon is the tendency for people to show preference toward people who belong to the same social groups as them (known as in-group members) rather than people who belong to other social groups (i.e., out-group members; Hewstone et al., 2002). This in-group bias is proposed to emerge naturally from group members’ motivations to achieve and maintain positive distinctiveness (Jetten et al., 2004). Previous meta-analyses have shown that sharing an important identity with a target motivates target-favoring behavior in a range of contexts, including jury decision-making (Mitchell et al., 2005), cooperative tasks (Balliet et al., 2014), and job performance evaluations (where in-group bias was found among male raters only; Bowen et al., 2000). Individual studies have also shown in-group favoritism in general helping (e.g., Levine et al., 2005; Stürmer et al., 2006). However, it is not yet known how robust the overall effect is in charitable giving contexts.

In-group bias is not a reflexive or straight-forward response to belonging to the same social group. Rather, the extent to which group membership translates into in-group favoritism depends on sociostructural factors, including the relative status and power of the groups and the extent to which these hierarchies are legitimate and stable. Indeed, Nadler (2002) proposes that intergroup helping relations are inherently status relations, and that sometimes people help out-groups to assert their own group’s dominance and keep beneficiary out-groups in a relatively low status position. Furthermore, van Leeuwen (2007) proposes that people sometimes give to out-groups as a way to restore threatened in-group identities. This literature implies that the effects of shared identity on giving may be complex and resist simple prescriptions such as “shared identity = more giving.” A meta-analysis can help tease apart the conditions under which social identification—whether shared identity or strength of identification—does (or does not) promote giving.

Charitable Triad Theory

Another important nuance is that, no matter how it is assessed, social identification can only be understood with reference to a particular target. Charitable Triad Theory (Chapman et al., 2022) proposes that charitable decisions are triadic and relational. This theory of donor psychology rests on three key tenets. First, giving is triadic: characteristics of three key actors—donor, beneficiary, and fundraiser—influence giving decisions. In the language of Charitable Triad Theory, donors are entities considering making gifts of money, beneficiaries are the end users of any funds donated, and fundraisers are the individuals or, more commonly, the nonprofit organizations who request donations from potential donors on behalf of beneficiaries. Second, understanding the characteristics of each actor in the triad may be necessary but not sufficient to understand the complexities of giving behavior. In other words, donor behavior cannot be meaningfully understood with reference to only one of the three targets. Instead (the third tenet), giving is relational: interactive relationships between the triad determine charitable choices. Various relationships between the actors are possible, including relationships based on liking, values alignment, or social identification.

Social identification is inherently relational, focused on a connection between two parties. Charitable Triad Theory, with its attention to relationships within the Charitable Triad, is therefore a useful frame to examine the way identification processes work in charitable giving contexts. Charitable Triad Theory proposes that characteristics of donors as well as donors’ relationships with both beneficiaries and fundraisers affect giving decisions. In this article, we suggest an enhancement to Charitable Triad Theory by proposing that the relationships donors have that influence their giving could also be with other donors. Specifically, we examine the extent to which relationships based on identification between a potential donor and (a) other donors, (b) beneficiaries, and (c) fundraisers influence charitable responses.

Based on Charitable Triad Theory and previous research, identification with all three members of the Charitable Triad should affect giving. Identification with other donors has been shown to promote giving. For example, customers who identify more strongly with a corporation are also more likely to donate to causes supported by that donor corporation (Lichtenstein et al., 2004). People are also more likely to model the generosity of a donor when they share an identity with that donor. For example, in one study, Italian students were willing to donate an average of $19.54 when they learned about the generous donation of a fellow Italian, but only $12.88 when the same donation was attributed to a German donor (Hysenbelli et al., 2013, Study 3). However, Croson et al. (2010) found that normative information about the giving of others was associated with the self-reported giving of women, but not men. Thus, identification with other donors should promote giving, but perhaps not for all people.

People who have stronger identification with beneficiaries appear to be more likely to give to charitable appeals and also to give more generously (e.g., Zagefka & Sun, 2021). For example, the more people identified with victims of Hurricane Katrina, the more likely they were to donate to disaster recovery efforts (Winterich et al., 2009). In a series of vignette studies, James and Zagefka (2017) also demonstrated preferences for donating to help disaster victims from participants’ national in-group compared to a fictional out-group country. In their third study, for example, participants reported that they would hypothetically make a £5.59 donation when they thought the victims were from their own country, compared to only £3.86 when victims were from another country. However, despite the expected tendency for people to help in-group members, sometimes donors offer help to people they do not identify with for strategic reasons such as to assert their group’s high status over others (Nadler, 2002). Thus, identification with beneficiaries may sometimes (but not always) promote giving.

Finally, identification with fundraisers can promote donations. For example, when people identify more with the fundraising nonprofit, they say they are more willing to donate to that nonprofit in the future (Boenigk & Helmig, 2013). Another study, however, shows that when supporters are seeking donations for a charity, their personal identification with the charity explains only 1% of variation in their fundraising success compared to other more pragmatic factors (Chapman et al., 2019). Thus, identification with fundraisers should promote giving, but the size of the relationship is disputable.

In sum, identification with other donors, beneficiaries, and fundraisers can sometimes, but perhaps not always, promote giving. Furthermore, the size of the relationships between identification with these targets and giving remains unknown. A meta-analysis can illuminate to what extent, and for which people, identification processes can be harnessed effectively for fundraising.

The Current Study

The purpose of the current study is to aggregate evidence to test the overall importance of social identification for charitable giving and identify caveats and boundary conditions to the relationship. Specifically, we meta-analyzed 109 effects from 74 independent samples recruited in 15 countries over a period of more than 40 years.

Our study is guided by several research questions and hypotheses (summarized in Figure 1). First, we anticipate an overall positive relationship between social identification and charitable giving. As elaborated earlier, this hypothesis is grounded in Social Identity Theory—which argues that identification is associated with various forms of prosocial behavior—and Charitable Triad Theory, which articulates that giving is inherently relational.

Figure 1

Conceptual Diagram of the Relationships and Moderators Being Tested in the Meta-Analysis.

Hypothesis 1 (H1): Social identification is positively associated with charitable giving.

An important nuance in the current analysis is that we distinguish between strength of identification (i.e., the extent to which people report being psychologically identified with targets) and shared identity (i.e., the extent to which targets share the same group membership as the giver). Both strength of identification and shared identity have been commonly used to assess identification effects in social identity research. From the perspective of Social Identity Theory, strength of identification provides a relatively clean index of the extent to which people are psychologically attached to a target, and so should have a relatively strong relationship with charitable giving. In contrast, as elaborated earlier, the effects of merely sharing an identity with a target are more complex and more ambiguous, potentially influenced by sociostructural relations between the groups and strategic considerations (Nadler, 2002; van Leeuwen, 2007). This reasoning leads us to propose:

Hypothesis 2 (H2): The positive relationship between strength of identification and charitable giving will be greater than the positive relationship between shared identity and charitable giving.

One important element of the current analysis is the overlay of a Charitable Triad lens to consider target of identification: Does it matter whether identification is measured in relation to other donors, beneficiaries, or fundraisers? Answering this question can be of considerable importance for practitioners, who will be interested in “where” to direct their marketing energies. There is not yet sufficient theorizing to provide a priori hypotheses around the target of identification that would be most important in promoting giving. Charitable Triad Theory makes no predictions about which actor would be most important, and with almost no empirical research considering all three actors simultaneously (Chapman et al., 2022), there is simply not enough comparative evidence to make clear predictions on this point. However, following the basic tenets of Charitable Triad Theory, we propose that identification with all three actors will be associated with giving:

Hypothesis 3 (H3): Social identification with other donors (H3a), beneficiaries (H3b), and fundraisers (H3c) will all be positively associated with charitable giving.

We also anticipate that the positive relationship between identification and charitable giving will be greater when giving is self-reported (e.g., as a behavioral intention or a report of past giving) than when it is measured objectively (e.g., observations of actual giving behavior). This hypothesis is grounded in two principles. First, intentions are an imperfect proxy for behavior, with larger relationships typically found for self-reported measures such as giving intentions than for observed giving behavior (e.g., Shang et al., 2019). Second, social desirability biases may lead people to over-report their giving to make a positive impression of themselves to others (Lee & Sargeant, 2011). Therefore:

Hypothesis 4 (H4): The association between social identification and charitable giving will be stronger when giving is self-reported than when it is assessed using objective measures.

In addition to these formal hypotheses, we also tested several other potential moderators that were not grounded in theory or previous research. Specifically, we examined whether the relationship between identification and charitable giving is affected by gender composition of the sample, mean age of the sample, national context, the charity’s cause, as well as how giving was operationalized. Testing these non-theoretical moderators provides useful information to fundraisers about who may be most responsive to identification-based appeals and which types of charities may achieve best results.

Method

We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA; Moher et al., 2015) to identify quantitative research on the relationship between social identification and charitable giving.

Search Strategy

Literature searches were run in November 2019 across four databases: the two largest multidisciplinary academic databases (Scopus and Web of Science) plus content-relevant databases in business (ABI/Inform) and psychology (PsycINFO). We used search terms nested around the concepts of charitable giving (donation*, donate, donor*, philanthrop*, charitable, “not for profit*”, “non profit*”, nonprofit*, NGO*, “non governmental”, “third sector”, “charit*”) and identification (identification, “social identity”, identifiers). Searches were based on title, abstract, and keywords and were limited to articles published in peer-reviewed journals since 1980. To ensure the corpus was as complete as possible, we supplemented the initial database searches with forward and backward citation searches on included articles (identifying all articles that the focal article had been cited by and had cited, respectively). These searches identified relevant articles published until the end of June 2022. Results were also supplemented with calls for unpublished data distributed through X (formerly Twitter) and the Association for Research on Nonprofit Organizations and Voluntary Action, American Marketing Association, and Society for Personality and Social Psychology mailing lists. All search efforts yielded 381 unique articles.

Eligibility and Screening

Screening was done in three rounds: first on title, second on abstract, third on full text. To be included, articles needed to be written in English and to have quantitatively measured both identification and charitable giving (specifically, donations of money to benefit non-kin others). For articles less than 10 years old that did not report bivariate effect sizes, we emailed authors to request this information. The PRISMA flowchart, summarizing the screening process, is presented in Figure 2.

Figure 2.

Flow Diagram of the Literature Search and Exclusion Process for Systematic Review and Meta-Analysis.

Coding

The type of bivariate effect size extracted depended on study design, but we typically extracted Pearson’s correlation coefficients (r) for correlational studies or Cohen’s ds for experimental studies. In situations where these values were not provided by the study, we used test statistics (e.g., means and standard deviations, F-values, frequency data) to calculate effect sizes. All effect sizes were transformed to Pearson r correlations for the meta-analysis.

In addition to effect sizes, we extracted information to examine potential moderators of the association. Of primary consideration, we extracted the way identification was assessed (i.e., measured strength of identification vs manipulated shared identity), the target of identification (i.e., with other donors, beneficiaries, or fundraisers), and how charitable giving was assessed (i.e., self-reported vs objective). We examined three variables relevant to the sample: gender (i.e., the proportion of the sample who were female), age (i.e., the mean reported age of the sample), and whether the data were collected in Western, Educated, Industrialized, Rich, and Democratic (WEIRD; Henrich et al., 2010) or non-WEIRD countries. Moderators based on methodology included the type of giving measure (i.e. likelihood of giving at all vs value of donation) and charity type. Further details, including definitions, examples, and coding of moderators, are available on the Open Science Framework (OSF; https://osf.io/hc63e/). The coding results spreadsheet, meta-analysis file, and citation library are also available on the OSF.

Analyses

Pearson’s correlation (r) was used to represent the association between social identification and charitable giving. To approximate a normal sampling distribution, all effect sizes (Pearson r correlations) were transformed to Fisher’s z correlations for the analyses. For ease of interpretation, our results report values transformed back to Pearson r correlations.

Given that the majority of studies included in this meta-analysis contributed more than one effect size (i.e., we extracted all relevant combinations of identification and charitable giving in a given study), we employed the three-level meta-analytic approach. Specifically, the three sources of variance accounted for are: sampling variance (Level 1), within-study variance (Level 2), and between-study variance (Level 3; Assink & Wibbelink, 2016).

Our analyses followed the procedures outlined by Assink and Wibbelink (2016) and used the rma.mv function of the Metafor package (Viechtbauer, 2010) in the R environment (R Core Team, 2020). We used multilevel random effects models and the restricted maximum likelihood estimation method to calculate all model parameters. The t-distribution was used to test individual regression coefficients and for calculating confidence intervals (Knapp & Hartung, 2003).

Sensitivity Analyses

Sensitivity analyses were conducted using a leave-one-out analysis and screening for r-standardized residuals with a cutoff of ±3.29 standard deviations. These analyses revealed one outlier (ES #74). Overall analyses with this effect size included or excluded returned the same pattern of results, so we elected to include this effect size in subsequent analyses.

Publication Bias

We tested for publication bias in three ways: visual inspection of the funnel plot (Figure 3), using Egger’s method of regressing the standardized effect size on the precision of the effect size (where a significant intercept indicates publication bias may be present), and the trim-and-fill method (which estimates how many studies are potentially missing and from which side of the average effect). We ran these tests only on the published articles in the sample. The results of these tests returned mixed results. Egger’s test for funnel plot asymmetry was not significant, t(41) = −0.39, p = .699, suggesting no publication bias; however, the trim-and-fill procedure estimated 10 missing studies on the right side of the funnel plot (i.e., positive effects). When estimated effect sizes from the hypothetical missing studies were imputed in the analyses, the strength of the association between identification and giving increased. This suggests that, if anything, the findings of our meta-analysis may be an underestimation of the true effect size—that is, overestimation from publication bias is likely not an issue. Nevertheless, because 14% of the effect sizes were drawn from unpublished studies, we were also able to test publication status (i.e., published vs unpublished datasets) as a potential moderator. Whether or not a dataset was published did not affect the size of association between identification and giving, F(1,107) = 0.47, p = .497.

Figure 3.

Funnel Plot for the Meta-Analysis.

Results

We report results in two phases. First, we summarize literature in terms of study characteristics, geography, and methods. In this first section, we summarize all 64 eligible articles that were identified in the systematic review. Second, we discuss the meta-analysis, including the aggregate effect size and evaluation of moderators. In this second section, we analyze only the 50 eligible articles that we were able to extract effect sizes from.

Description of the Literature

Our systematic review identified 64 relevant articles published since 1980 (summarized in Table 1). Based on the number of publications per year (Figure 4), there has been a noticeable burst of research attention on the role of identification in giving over the past decade. These articles were published across diverse disciplines, but especially in psychology (30%), marketing (25%), and nonprofit study journals (11%). Data on the association between identification and giving were collected in 15 countries. However, 84% of samples were collected in WEIRD countries, especially the United States (k = 42) and the United Kingdom (k = 11). On average, females constituted 58.7% of the samples, and the average age was 33 years. Among the articles that noted type of sample, approximately half used student samples (k = 31), and half used community samples (k = 37), but none used nationally representative samples. Overall, 70% of studies reported correlational data—either using surveys or experimental designs with other focal independent variables and measuring strength of identification—and 30% experimentally manipulated group status of targets (in-group vs out-group) to assess the impacts of shared identity.

Table 1.

Summary of 64 Articles Including Quantitative Data Examining the Association Between Identification and Giving Published Between 1980 and 2022.

	Study	N	Country	% Female	M age	Method	Type of identification (and target)	Giving outcome
*	Mael & Ashforth (1992)	297	USA	NA	NA	Survey	Fundraiser (nonprofit)	Self-report (priorities)
*	Tom & Elmer (1994)	1235	USA	NA	NA	Survey	Fundraiser (university)	Self-report (willingness)
*	Bhattacharya et al. (1995)	306	USA	NA	NA	Survey	Fundraiser (nonprofit)	Self-report (past giving)
*	Platow et al. (1999)	390	Australia	NA	NA	Experiment	Fundraiser (in-group vs out-group)	Objective (observed)
*	Brady et al. (2002)	595	USA	NA	NA	Survey	Fundraiser (nonprofit)	Self-report (intentions)
*	Arnett et al. (2003)	953	USA	55	NA	Survey	Fundraiser (university)	Objective (database)
*	Levine et al. (2004)	50	UK	56	19	Experiment	Beneficiary (victims)	Self-reported (willingness)
*	Lichtenstein et al. (2004)	660	USA	NA	NA	Correlational	Donor (corporate donor)	Self-report (intentions); Objective (study payment)
*	Shang et al. (2008)	76	USA	NA	NA	Experiment	Donor (in-group vs out-group)	Objective (observed)
	Hung & Wyer (2009)	54	China (Hong Kong)	NA	NA	Correlational	Beneficiary (victim)	Self-report (hypothetical value)
*	Winterich et al. (2009)	634	USA	NA	30	Correlational	Beneficiary (inclusion of other in self)	Self-report (hypothetical value); objective (study payment)
*	Kim et al. (2010)	306	Korea	68	NA	Survey	Fundraiser (with university)	Self-report (intentions)
*	Porter et al. (2011)	110	USA	46	NA	Survey	Fundraiser (with university)	Self-report (intentions)
*	McFarland et al. (2012)	3534	USA	NA	NA	Survey, experiment	Beneficiary (with all humanity)	Self-report (willingness); Objective (study payment)
	Bennett et al. (2013)	90	USA	NA	NA	Experiment	Donor (with corporate donor)	Self-report (willingness)
*	Boenigk & Helmig (2013)	314	Germany	56	NA	Survey	Fundraiser (with nonprofit)	Self-report (willingness, past giving)
*	Hysenbelli et al. (2013)	345	Italy	66	22	Experiment	Donor (in-group vs out-group)	Self-report (willingness, hypothetical value)
*	Kim & Walker (2013)	305	USA	16	39	Survey	Fundraiser (with figurehead of nonprofit)	Self-report (intentions)
	Ko et al. (2013)	816	USA	23	55	Survey	Fundraiser (with nonprofit)	Self-report (intentions)
	Stephenson (2013) ^a	2763	USA	57	47	Survey	Fundraiser (with brand)	Objective (database)
*	Zagefka et al. (2013)	100	Mixed	57	25	Survey	Beneficiary (with victims)	Self-report (willingness)
	Beldad et al (2014)	304	Netherlands	59	NA	Correlational	Fundraiser (with cause)	Self-report (intentions)
	Coulter (2014)	489	USA	56	53	Survey	Fundraiser (with nonprofit)	Self-report (intentions, past giving)
*	Hou et al. (2014)	205	China	NA	NA	Survey	Fundraiser (with nonprofit)	Self-report (past giving)
	Paulin, Ferguson, Jost, et al. (2014) ^b	592	Canada	56	NA	Correlational	Fundraiser (with cause)	Self-report (intentions)
	Paulin, Ferguson, Schattke, et al. (2014) ^b	592	Canada	56	NA	Correlational	Fundraiser (with cause)	Self-report (intentions)
	Stephenson & Bell (2014) ^a	2763	USA	57	NA	Survey	Fundraiser (with brand)	Objective (database)
*	Stephenson & Yerger (2014) ^a	2763	USA	57	NA	Survey	Fundraiser (with brand)	Objective (database)
*	Beldad et al. (2015)	380	Netherlands; USA	52	NA	Survey	Fundraiser (with cause)	Self-report (intentions)
*	Charynsh et al. (2015)	448	India	34	29	Experiment	Beneficiary (with victims)	Self-report (value)
*	Gibson (2015)	756	USA	84	NA	Survey	Fundraiser (with University)	Objective (database)
*	Lorenz et al. (2015)	282	USA	61	32	Experiment	Beneficiary (in-group vs out-group)	Self-report (willingness)
*	Park & Lee (2015)	465	Korea	NA	NA	Experiment	Beneficiary (in-group vs out-group)	Self-report (willingness)
*	Parris et al. (2015)	480	USA	56	NA	Correlational	Other (identification with surfing)	Self-report (intentions)
	Reese et al. (2015)	80	Luxembourg	55	25	Experiment	Beneficiary (with all humanity)	Objective (study payment)
*	Telzer et al. (2015)	26	USA	54	19	Experiment	Beneficiary (in-group vs out-group)	Objective (study payment)
	Ashar et al. (2016)	258	USA	NA	NA	Correlational	Beneficiary (shared values; similar SES)	Objective (study payment)
*	Kim (2016)	511	USA	31	32	Survey	Beneficiary (with athlete/team)	Self-report (intentions)
*	Kwak & Kwon (2016)	201	USA	53	51	Survey	Donor (with donor sports team)	Self-report (intentions)
*	Thomas et al. (2016)	384	Australia	60	50	Survey	Donor (with supporters)	Self-report (past giving); objective (study payment)
*	Amiot & Bastian (2017)	162	Australia; USA; Canada	83	31	Survey	Beneficiary (with animals)	Self-report (hypothetical value)
*	James & Zagefka (2017)	274	UK	71	25	Experiment	Beneficiary (in-group vs out-group)	Self-report (willingness, hypothetical value)
*	Neary (2017)	707	USA	57	49	Survey	Fundraiser (with university)	Objective (database)
*	Alt et al. (2018)	92	Indonesia	74	23	Experiment	Beneficiary (in-group vs out-group)	Objective (study payment)
*	Kogut et al. (2018)	214	Israel	65	24	Correlational	Beneficiary (with victim)	Objective (study payment)
*	Peasley et al. (2018)	584	USA	68	45	Survey	Fundraiser (with nonprofit)	Self-report (intentions, past giving)
*	Schattke et al. (2018)^b	712	Canada	NA	NA	Correlational	Fundraiser (with cause)	Self-report (willingness)
*	Watson (2018)	530	USA	NA	NA	Survey	Fundraiser (with nonprofit)	Objective (database)
*	Chapman et al. (2019)	1646	Australia	66	NA	Survey	Fundraiser (with nonprofit)	Objective (database)
*	James (2019)	2566	USA	51	36	Correlational	Donor (with example donor)	Self-report (willingness)
	Jang et al. (2019)	145	USA	41	35	Correlational	Fundraiser (with figurehead of nonprofit)	Self-report (intentions)
*	Maki et al. (2019)	2023	Chile	NA	NA	Survey	Beneficiary (with country people)	Self-report (frequency; past giving)
*	Amiot et al. (2020)	215	USA	55	32	Survey	Beneficiary (with animals; with all humanity)	Self-report (hypothetical value)
	Fang et al. (2020)	2547	USA	NA	NA	Survey	Fundraiser (with nonprofit)	Objective (database)
*	Hwang et al. (2020)	490	USA	52	36	Survey	Beneficiary (with athletic department); fundraiser (with university)	Self-report (intentions)
*	Imada (2020)	234	USA; UK	NA	NA	Experiment	Beneficiary (in-group vs out-group)	Objective (study payment, lottery)
	Jin et al. (2020)	561	USA; China	51	29	Experiment	Beneficiary (in-group vs out-group)	Self-report (intentions); objective (lottery)
*	Madrigal (2020)	407	USA	42	NA	Survey	Fundraiser (with nonprofit)	Objective (database)
*	Linos et al. (2021)	79368	Greece	35	40	Experiment	Beneficiary (in-group vs out-group)	Objective (database)
*	Manokara et al. (2021)	1907	USA; Canada	45	35	Correlational	Beneficiary (with others)	Objective (study payment)
*	Peak & McGarty (2021)	159	Australia	79	21	Correlational	Donor (with supporters)	Self-report (intentions)
*	Zagefka & Sun (2021) ^c	500	UK	68	36	Survey	Beneficiary (with victims)	Self-report (willingness)
*	Chapman et al. (2022)	934	USA	59	39	Correlational	Beneficiary (with victims)	Self-report (willingness); objective (study payment)
*	Zagefka (2022)^c	500	UK	68	36	Survey	Beneficiary (with victims)	Self-report (willingness)

Note. NA = information was not disclosed; * in the left column indicates the article is included in the meta-analysis. Articles that share a superscript letter report the same data. Correlational refers to experiments that measured (but did not manipulate) identification. Ns reported indicate total number of participants involved in studies reported in the article; corresponding meta-analysis Ns are sometimes smaller if not all studies and/or experimental conditions were analyzed.

Figure 4.

Summary of Publications Including Measures of Both Identification and Charitable Giving Over Time.

Different targets of identification were assessed. Most examined identification with either beneficiaries (k = 46) or fundraisers (k = 34), while only 11 studies examined identification with other donors. Identification was typically assessed with self-report measures (e.g., “When I talk about this charitable organization, I usually say ‘we’ rather than ‘they’”; Hou et al., 2014) or by manipulating the group membership of the target (in-group vs out-group).

Studies that used self-reports of giving included those asking participants to report their willingness to donate (k = 31), future intentions to donate (k = 25), hypothetical amount they would donate (k = 10), and self-reported past giving (k = 10). Those studies that included objective measures of giving typically assessed the amount of a study payment that the participant donated (k = 26), analyzed real giving behavior from charity databases (k = 12), included a lottery where participants could donate a share if they win (k = 7), or directly observed giving in the field (k = 2).

Finally, studies examined giving to a range of charity types. Many studies considered giving to multiple or all charities (k = 19). However, when the focus was on a particular cause, the most popular causes studied were education (k = 14), emergency relief (k = 12), international charities (k = 12), and children’s charities (k = 9).

Meta-Analysis

For the meta-analysis, we extracted 109 effect sizes from 74 independent samples published in 50 articles. On average, each study yielded 1.43 effect size estimates (SD = 0.79, range = 1–6). Data came from 89,570 participants in 15 countries.

Overall Association

Overall, a reliable positive association was found between social identification and charitable giving, r = .29, t = 7.46, p < .001, 95% CI = [.21, .36].

Heterogeneity

There was significant variance in effect sizes observed at both the within-study level (p < .001) and between-study level (p < .001). Using Cheung’s (2014) formula, 1.85% of the total variance could be attributed to sampling variance, 33.74% to within-study variance, and 64.41% to between-study variance. Results of these heterogeneity tests suggest that the true effects vary significantly across studies and measures, more so than would be expected based on sampling error alone. Thus, moderator analyses were warranted to identify likely causes of this heterogeneity.

Moderator Analyses

Three-level meta-regressions examined potential moderators of the association between identification and giving. Before conducting these moderator analyses, all categorical variables were coded into dichotomous dummy variables, and continuous variables were centered around their mean. With a sample of 109 effect sizes, we did not have sufficient statistical power to run multivariate analyses. Therefore, following the work of Hox and colleagues (2010), we conducted univariate moderator analyses, whereby each moderator was tested independently. Further detail on how moderators were defined and coded is available on the OSF. Results are summarized in Table 2.

Table 2

Moderators of the Overall Association Between Identification and Charitable Giving.

Moderator	k	F(df₁, df₂)	β0	β1	95% CI		t₀	t₁
Moderator	k	F(df₁, df₂)	β0	β1	Lower	Upper	t₀	t₁
Identification vs shared identity	109	F(1,107) = 4.77*
Strength of identification (survey)	74		.32	—	.25	.40	7.78***
Shared identity (experiment)	35		.15	—	.01	.29	2.05*	−2.18*
Target of identification	108	F(2,105) = 1.46
Other donor	13		.23	—	.03	.41	2.23*
Beneficiary	62		.24	—	.13	.34	4.37***	0.11
Fundraiser	33		.36	—	.25	.46	6.28***	1.17
Self-reported vs objective giving	109	F(1,107) = 5.36*
Self-reported	71		.33	—	.25	.41	7.89***
Objective	38		.20	—	.09	.30	3.59***	−2.32*
Sample characteristics
Proportion female	74	F(1,72) = 1.06	—	−.00	−.01	.00	—	—
Mean age	59	F(1,57) = 0.00	—	−.00	−.01	.01	—	—
WEIRD status	108	F(1,106) = 0.54
Non-WEIRD country	24		.34	—	.15	.51	3.41***
WEIRD country	84		.27	—	.19	.34	6.48***	0.74
Other
Charitable giving measured	103	F(1,101) = 5.31*
Likelihood	63		.34	—	.25	.41	7.70***
Value	40		.20	—	.10	.30	3.72***	−2.30*
Recipient of giving 1	93	F(1,91) = 6.15*
Individual	17		.04	—	−.19	.27	0.35
Charity	76		.34	—	.26	.41	7.94***	2.48*
Recipient of giving 2	74	F(1,72) = 0.03
Multiple charities	23		.34	—	.20	.48	4.44***
Single charity	51		.33	—	.23	.43	6.13***	−0.16
Charity type	69	F(5,63) = 0.87
Children	13		.27	—	−.04	.54	1.73
Education	17		.46	—	.30	.59	5.47***	1.16
Health	7		.27	—	.00	.50	2.01*	−0.01
International	10		.24	—	.02	.44	2.18*	−0.16
Emergency	9		.31	—	.08	.51	2.66**	0.21
Other	13		.26	—	.07	.43	2.76**	−0.07
Publication status	109	F(1,107) = 0.47
Published	94		.30	—	.22	.37	7.14***
Unpublished	15		.22	—	.02	.41	2.14*	0.68

Note. For each category of moderator variable, the reference variable is listed first. For continuous moderators, we report β₁ (i.e., estimated regression coefficient), which conveys the increase in mean effect size (r) for a unit increase in the moderator variable. For categorical moderators, we report β₀ (the mean r for each level of a given moderator variable). k = number of effect sizes contributing to analysis; F(df1, df2) = omnibus test; t₀ = difference in mean r with zero, t₁ = difference in mean r compared with reference variable.

p < .05; **p < .01; ***p < .001

Strength of Identification versus Shared Identity

The way identification was assessed significantly moderated the association between identification and giving, F(1,107) = 4.77, p = .031. Specifically, the relation between identification and giving was significantly stronger in survey studies that measured both strength of identification and giving, r = .32, p < .001, than in studies that experimentally manipulated shared identity (i.e., in-group vs out-group status of targets), r = .15, p = .043.¹ Two forest plots summarize all effect sizes included in the meta-analysis: Figure 5 shows effect sizes for studies that measured strength of identification, and Figure 6 shows effect sizes for studies that manipulated shared identity.

Figure 5.

Forest Plot Summarizing 74 Effect Sizes (Pearson’s Correlation Coefficients; r) From Studies That Measured Strength of Identification.

Figure 6

Forest Plot Summarizing 35 Effect Sizes (Pearson’s Correlation Coefficients; r) From Studies That Manipulated Shared Identity (i.e., In-group vs Out-group Status of Targets).

Target of Identification

Target of identification did not significantly moderate the association between identification and giving, F(2,105) = 1.46, p = .237. Instead, identification with all three of the Charitable Triad actors was significantly associated with giving: identification with other donors (r = .23, p = .028), beneficiaries (r = .24, p < .001), and fundraisers (r = .36, p < .001).

Self-Reported Versus Objective Giving

How giving was measured affected the strength of association.² Studies that captured self-reported giving (e.g., intentions to give, self-reported past giving value) yielded a stronger relation, r = .33, p < .001, than did studies that captured objective giving (e.g., actual giving behavior), r = .20, p < .001, F(1,107) = 5.36, p = .023.

Other Moderators

The proportion of females in the sample, F(1,72) = 1.06, p = .307, the average age of the sample, F(1,57) = 0.00, p = .967, or whether the data were collected in WEIRD or non-WEIRD countries, F(1,106) = 0.54, p = .463, did not change the size of the relationship between identification and charitable giving. Identification was more strongly associated with the likelihood of giving at all, r = .34, p < .001, than with the value of donations, r = .20, p < .001, F(1,101) = 5.31, p = .023. Finally, the recipient of giving also moderated the effects, F(1,91) = 6.15, p = .015. Identification was associated with giving to charitable organizations, r = .34, p < .001, but not directly to individuals, r = .04, p = .725. It did not matter whether the donation would benefit a single charity or multiple charities, F(1,72) = 0.03, p = .875; nor did the type of charitable cause in question moderate the relationship, F(5,63) = 0.87, p = .503.

Discussion

We meta-analyzed over 40 years of available quantitative data on the relationship between social identification and giving. Overall, we found a moderate positive association that varied depending on the type of identification (strength of identification vs shared identity) but not across different types of people or causes. Identification with all three actors of the Charitable Triad—other donors, beneficiaries, and fundraisers—was positively associated with charitable giving. However, identification was only associated with donations made through charities and not for those offered directly to individuals. In combination, these findings suggest that identification processes may be relatively universal in mediated giving contexts and could therefore be incorporated into fundraising strategies in diverse ways, as discussed below.

Consistent with H1, and in line with Social Identity Theory, our meta-analysis found a positive association between social identification and charitable giving (r = .29). Although the size of this effect is moderate, it should be noted that charitable giving is a complex behavior, determined by multiple psychological principles and curtailed by numerous practical constraints. The complex nature of the phenomenon means that it is rare for any one dimension to have large effects on giving. To benchmark the size of the current effect, it should be noted that it exceeds previous meta-analytic estimates on the effects of trust (r = .22; Chapman et al., 2021), having one’s giving observed (r = .15; Bradley et al., 2018), and exposure to prosocial media (r = .09; Coyne et al., 2018).

Consistent with H2, identification was found to be more strongly associated with giving when strength of identification was measured compared to when a shared identity based on common group membership was experimentally manipulated. As discussed earlier, measuring strength of identification captures a subjective sense of connection between participants and the relevant targets, providing a relatively clean index of the psychological relationship between participants and a target. In contrast, the mere effects of shared group membership are a relatively crude proxy for identification, albeit one that is commonly used. Although people’s preference for helping in-group members over out-group members has been well established (Levine et al., 2002, 2005; Levine & Thompson, 2004; Stürmer et al., 2005, 2006), the relatively weak association found here suggests that charitable giving may be a unique form of intergroup behavior that evokes distinct psychological processes compared to more generalized forms of helping. Perhaps donating money to out-groups is sometimes seen as a way to benefit the in-group, by restoring group-based esteem or to maintain a sense of higher status (see also the study by van Leeuwen, 2007). The notion that people sometimes help out-groups for strategic reasons has been established empirically (e.g., Halabi et al., 2008; Nadler & Chernyak-Hai, 2014; Nadler et al., 2009) and may explain why effects of the in-group (vs out-group) status of the relevant target were not very strong.

Relatedly, identities bring with them perceptions of what group members typically do or approve of; such perceptions are called social norms (Jetten et al., 1996; Lay et al., 2020; Smith & Louis, 2009; Terry & Hogg, 1996). It may be that certain identities contain norms for giving (see also the study by Chapman et al., 2023), which means that when the identity is made salient, then giving is more likely regardless of target. For example, if there is a perceived norm that British people are typically generous, then making the British identity salient may promote charitable giving to both in-group (i.e., British) and out-group (i.e., non-British) targets. This would also help explain the weaker effects observed for studies that manipulated shared identity status.

Finally, it may simply be that identification can more easily be harnessed than coaxed. In other words, if a subjective sense of identification exists (as when strength of identification is measured), then it promotes giving. However, simply manipulating group membership to highlight shared identities may not be as effective if the social identity in question is not central, positively valued, or does not generate a sense of connection with other group members (Cameron, 2004).

Our analyses were framed around the basic tenets of Charitable Triad Theory (Chapman et al., 2022): that charitable giving is triadic and relational. Results show how relationships of identification between a potential donor and (1) other donors, (2) beneficiaries, and (3) fundraisers are all positively associated with giving outcomes (supporting H3a–c). This evidences Charitable Triad Theory’s central argument, that to understand donor psychology one must consider all three actors in the triad (and the relationships between them) because each can influence charitable outcomes. Although no significant differences were found in the importance of identification with each of the actors, inspection of the trends suggests that identification with fundraisers (e.g., the nonprofit raising money for the cause) may be an especially powerful lever for promoting charitable giving. This is particularly important to note given the comparative neglect that fundraisers have experienced in the charitable giving literature (Breeze, 2017; Chapman et al., 2022).

The relationship between identification and charitable giving was significant regardless of whether giving was self-reported or measured objectively. However, consistent with H4, the size of the effect was somewhat stronger when giving was measured through self-reports. As discussed earlier, self-report measures can be notoriously imperfect proxies for reality, both because practical constraints curtail our ability to translate intentions into actions, and because people tend to exaggerate or over-report their prosocial behaviors (Lee & Sargeant, 2011). Having said that, the relationship between identification and objective giving was non-trivial in size (r = .20; 4% of variance in giving explained). Relatedly, the association was stronger when considering whether someone would give at all (giving likelihood) than how much they would give (giving value). This also likely stems from the fact that many pragmatic concerns (e.g., income) can affect the way an individual’s desire to give manifests.

The relationship between identification and giving was found to be stable across men and women, people of different ages, and cultural contexts. A ramification of this finding is that fundraising strategies based on identification with campaign-relevant targets should be effective for diverse segments and in a broad range of fundraising markets. However, we found identification effects when people were giving in ways that were mediated by charities, but not when people were giving directly to individual strangers. It is possible that design confounds this question, as most of the studies assessing giving to individuals were economic game experiments, which may not have ecological validity for assessing giving behaviors. Alternatively, it may be that social identities (sense of “we”) are more salient in formal giving contexts, while personal identities (sense of “I”) may be more meaningful in informal, one-on-one helping contexts. It is also possible that empathy may override identification processes when dealing with an individual recipient of charity; indeed, past research indicates that empathy works best on solitary targets and that people are more generous with individual victims (e.g., Bloom, 2006; Slovic, 2007). Future research may wish to examine whether it is indeed the individual that breaks the power of identification (rather than an artifact of study design) and why that happens. Experimental approaches would lend themselves well to these research questions.

Strengths, Limitations, and Future Directions

By aggregating data from 74 independent samples that used diverse methods, we can be confident about the generalizability of findings. Nevertheless, meta-analyses are limited by what data are available, and we found comparably few studies that examined identification with other donors. Those that did consider other donors often studied the impact of identification with corporate donors or celebrity supporters. Furthermore, none of the studies employed nationally representative samples. Future research may wish to examine how identification with everyday donors (i.e., people like you or I that support the target cause) may influence giving responses and should ideally collect data from nationally representative samples.

Our meta-regressions consider each moderator independently. There simply have not yet been enough studies examining identification and charitable giving to have sufficient power (i.e., enough effect sizes) to conduct multivariate meta-regressions and consider all moderators simultaneously. Doing so would add nuance to the current findings. We therefore recommend multivariate moderation analyses be conducted in the future when the corpus of available research is larger.

A further limitation of the available data is that the evidence base considered identification with other donors, beneficiaries, and fundraisers separately. We were therefore not able to look at their additive or interactive effects. This reflects a deficit in the literature: as highlighted by Chapman and colleagues (2022), the vast majority of studies on charitable giving focus on only one dimension of the triad, and only 2 of 1,337 articles in their review examined the full triad of actors simultaneously. We encourage future scholarship to engage with the dyadic and triadic relationships between the three Charitable Triad actors and to examine how identification with each actor interacts with identification with others.

Finally, only one of the studies examining donors’ identification with fundraisers considered individual fundraisers, with the others looking at identification with fundraising organizations or causes. This relative emphasis is reflective of the broader charitable giving literature, and not confined to social identification research (Chapman et al., 2022). We believe identification dynamics between donors and individual fundraisers, especially in the context of major and peer-to-peer giving, would be a fruitful direction for future research.

Managerial Implications

Having aggregated and examined 40 years’ worth of research, we conclude that social identification plays an important role in giving. People who identify more with other donors, with beneficiaries, or with fundraisers are more likely to give and are more generous when they do give. Fundraisers and nonprofit marketers could therefore benefit from incorporating identification cues into the framing of their appeals and in their donor solicitations.

If a subjective sense of identification exists (as when strength of identification is measured) then it promotes giving. However, simply highlighting shared identities may not be as effective if the social identity in question is not central, positively valued, or does not generate a sense of connection with other group members. Although shared identities can promote giving, simply highlighting the in-group status of a beneficiary (e.g., campaigns prompting potential donors to help people “in their own backyard”) or a fundraiser (e.g., highlighting that the fundraising organization has the same religious affiliation as potential donors) may not be as effective. Fundraisers must instead work to nourish a subjective sense of identification with the relevant Charitable Triad targets.

Nonprofit marketers have three key opportunities to leverage the power of social identification. They can leverage identification with other donors; for example, by recruiting existing major donors as spokespeople to solicit and attract new high-level givers. They can leverage identification with beneficiaries; for example, by drawing attention to points of similarity between donors and beneficiaries in campaign materials. Finally, they can leverage identification with fundraisers, which our analyses suggest may have a particularly strong relationship with charitable giving. One approach would be to work to increase the diversity of fundraisers themselves. In the UK, for example, 77% of fundraisers are women, and they are generally comparatively young and highly educated (Breeze, 2017). By increasing diversity in their ranks, fundraising teams may find more opportunities for shared identities with different kinds of donors. Beyond diversity, some useful roadmaps for how to cultivate meaningful relationships between individual fundraisers and high-value supporters have been laid out, such as Burnett’s (2002) relationship fundraising approach and Shaker and Nelson’s (2022) five-tier model of relationship building in fundraising. It may also be worth considering what not to do, such as Harrison’s (2023) recent research on common mistakes in donor stewardship and why donor-fundraiser relationships sometimes fail. Drawing attention to the relationship the donor has with either the fundraising organization or their spokespeople may also be particularly powerful in the contexts of alumni giving, celebrity endorsement, and peer-to-peer campaigns.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Cassandra Chapman is the recipient of an Australian Research Council Discovery Early Career Researcher Award (project number DE220100903) funded by the Australian Government.

ORCID iD

Cassandra M. Chapman

Notes

Author Biographies

Cassandra M. Chapman is an Associate Professor of Marketing at the University of Queensland researching the psychology of charitable giving, effective and ethical fundraising, and charity scandals.

Jessica L. Spence is a postdoctoral researcher at the University of Queensland researching how intergroup biases shape social decision-making.

Matthew J. Hornsey is a Professor of Management at the University of Queensland researching social influence.

Lucas Dixon completed a PhD at the University of Queensland on the psychology of magical thinking in business and consumer decision-making.

References

Aaker

J. L.

Akutsu

(2009). Why do people give? The role of identity in giving. Journal of Consumer Psychology, 19(3), 267–270. https://doi.org/10.1016/j.jcps.2009.05.010

Ahern

(2020). Identity-based fundraising: What 6 years of research revealed [Webinar]. https://www.youtube.com/watch?v=tZCKzqFqd8s

Ashar

Y. K.

Andrews-Hanna

J. R.

Yarkoni

Sills

Halifax

Dimidjian

Wager

T. D.

(2016). Effects of compassion meditation on a psychological model of charitable donation. Emotion, 16(5), 691–705. https://doi.org/10.1037/emo0000119

Assink

Wibbelink

C. J.

(2016). Fitting three-level meta-analytic models in R: A step-by-step tutorial. Quantitative Methods for Psychology, 12(3), 154–174. https://doi.org/10.20982/tqmp.12.3.p154

Balliet

De Dreu

C. K. W.

(2014). Ingroup favoritism in cooperation: A meta-analysis. Psychological Bulletin, 140, 1556–1581. https://doi.org/10.1037/a0037737

Bekkers

Wiepking

(2011). A literature review of empirical studies of philanthropy: Eight mechanisms that drive charitable giving. Nonprofit and Voluntary Sector Quarterly, 40(5), 924–973. https://doi.org/10.1177/0899764010380927

Beldad

Gosselt

Hegner

Leushuis

(2015). Generous but not morally obliged? Determinants of Dutch and American donors’ repeat donation intention (REPDON). Voluntas, 26(2), 442–465. http://www.jstor.org/stable/43654690

Beldad

Snip

van Hoof

(2014). Generosity the second time around: Determinants of individuals’ repeat donation intention. Nonprofit and Voluntary Sector Quarterly, 43(1), 144–163. https://doi.org/10.1177/0899764012457466

Bennett

C. M.

Kim

Loken

(2013). Corporate sponsorships may hurt nonprofits: Understanding their effects on charitable giving. Journal of Consumer Psychology, 23(3), 288–300. https://doi.org/10.1016/j.jcps.2012.10.010

10.

Bloom

(2006). Against empathy. The Bodley Head.

11.

Boenigk

Helmig

(2013). Why do donors donate? Examining the effects of organizational identification and identity salience on the relationships among satisfaction, loyalty, and donation behavior. Journal of Service Research, 16(4), 533–548. https://doi.org/10.1177/1094670513486169

12.

Bowen

C.-C.

Swim

J. K.

Jacobs

R. R.

(2000). Evaluating gender biases on actual job performance of real people: A meta-analysis. Journal of Applied Social Psychology, 30(10), 2194–2215. https://doi.org/10.1111/j.1559-1816.2000.tb02432.x

13.

Bradley

Lawrence

Ferguson

(2018). Does observability affect prosociality? Proceedings Biological Sciences/The Royal Society, 285, 1875. https://doi.org/10.1098/rspb.2018.0116

14.

Breeze

(2017). The new fundraisers: Who organises charitable giving in contemporary society? Policy Press.

15.

Burnett

(2002). Relationship fundraising: A donor-based approach to the business of raising money (2nd ed.). Jossey-Bass.

16.

Cameron

J. E.

(2004). A three-factor model of social identity. Self and Identity, 3(3), 239–262. https://doi.org/10.1080/13576500444000047

17.

Chapman

C. M.

Dixon

Wallin

Young

Masser

B. M.

Louis

W. R.

(2023). We usually give like this: Social norms describe typical charitable causes supported by group members. Nonprofit and Voluntary Sector Quarterly, 53(1), 29–53. https://doi.org/10.1177/08997640231160467

18.

Chapman

C. M.

Hornsey

M. J.

Gillespie

(2021). To what extent is trust a prerequisite for charitable giving? A systematic review and meta-analysis. Nonprofit and Voluntary Sector Quarterly, 50(6), 1274–1303. https://doi.org/10.1177/08997640211003250

19.

Chapman

C. M.

Louis

W. R.

Masser

B. M.

Thomas

E. F.

(2022). Charitable Triad Theory: How donors, beneficiaries, and fundraisers influence charitable giving. Psychology & Marketing, 39(9), 1826–1848. https://doi.org/10.1002/mar.21701

20.

Chapman

C. M.

Masser

B. M.

Louis

W. R.

(2019). The Champion Effect in peer-to-peer giving: Successful campaigns highlight fundraisers more than causes. Nonprofit and Voluntary Sector Quarterly, 48(3), 572–592. https://doi.org/10.1177/0899764018805196

21.

Chapman

C. M.

Masser

B. M.

Louis

W. R.

(2020). Identity motives in charitable giving: Explanations for charity preferences from a global donor survey. Psychology & Marketing, 37(9), 1277–1291. https://doi.org/10.1002/mar.2136

22.

Charity Link. (2022, 17 August). Why donor identity matters to charities and fundraisers. https://www.charitylink.net/blog/reasons-people-give-to-charity

23.

Cheung

M. W.

(2014). Modeling dependent effect sizes with three-level meta-analyses: A structural equation modeling approach. Psychological Methods, 19(2), 211–229. https://doi.org/10.1037/a0032968

24.

Coulter

T. L.

(2014). Alumni organizational identification and motivation [Masters Thesis, Florida State University]. http://purl.flvc.org/fsu/fd/FSU_migr_etd-9162

25.

Coyne

S. M.

Padilla-Walker

L. M.

Holmgren

H. G.

Davis

E. J.

Collier

K. M.

Memmott-Elison

M. K.

Hawkins

A. J.

(2018). A meta-analysis of prosocial media on prosocial behavior, aggression, and empathic concern: A multidimensional approach. Developmental Psychology, 54(2), 331–347. https://doi.org/10.1037/dev0000412

26.

Croson

Handy

Shang

(2010). Gendered giving: The influence of social norms on the donation behavior of men and women. International Journal of Nonprofit and Voluntary Sector Marketing, 15(2), 199–213. https://doi.org/10.1002/nvsm.385

27.

Fang

Fombelle

P. W.

Bolton

R. N.

(2020). Member retention and donations in nonprofit service organizations: The balance between peer and organizational identification. Journal of Service Research, 24(2), 187–205. https://doi.org/10.1177/1094670520933676

28.

Giving USA. (2021). Giving USA 2021: In a year of unprecedented events and challenges, charitable giving reached a record $471.44 billion in 2020. https://www.givinginstitute.org/resource/resmgr/gusa/2021_resources/gusa_2021_press_release_fina.pdf

29.

Halabi

Dovidio

J. F.

Nadler

(2008). When and how do high status group members offer help: Effects of Social Dominance Orientation and status threat. Political Psychology, 29(6), 841–858. https://doi.org/10.1111/j.1467-9221.2008.00669.x

30.

Harrison

V. S.

(2023). Why relationships fail: Donor insights on defining and understanding successes and failures in nonprofit stewardship. Journal of Public Relations Research, 35(4), 236–258. https://doi.org/10.1080/1062726X.2023.2198237

31.

Henrich

Heine

S. J.

Norenzayan

(2010). The weirdest people in the world? Behavioral and Brain Sciences, 33(2–3), 61–83. https://doi.org/10.1017/S0140525X0999152X

32.

Hewstone

Rubin

Willis

(2002). Intergroup bias. Annual Review of Psychology, 53(1), 575–604. https://doi.org/10.1146/annurev.psych.53.100901.135109

33.

Hornsey

M. J.

(2008). Social Identity Theory and Self-Categorization Theory: A historical review. Social and Personality Psychology Compass, 2(1), 204–222. https://doi.org/10.1111/j.1751-9004.2007.00066.x

34.

Hou

Eason

C. C.

Zhang

(2014). The mediating role of identification with a nonprofit organization in the relationship between competition and charitable behaviors. Social Behavior and Personality: An International Journal, 42(6), 1015–1028. https://doi.org/10.2224/sbp.2014.42.6.1015

35.

Hox

J. J.

Moerbeek

Van de Schoot

(2010). Multi-level analysis: Techniques and applications (2nd ed.). Routledge.

36.

Hung

I. W.

Wyer

R. S.

(2009). Differences in perspective and the influence of charitable appeals: When imagining oneself as the victim is not beneficial. Journal of Marketing Research, 46(3), 421–434. https://doi.org/10.1509/jmkr.46.3.421

37.

Hysenbelli

Rubaltelli

Rumiati

(2013). Others’ opinions count, but not all of them: Anchoring to ingroup versus outgroup members’ behavior in charitable giving. Judgment and Decision Making, 8(6), 678–690.

38.

James

T. K.

Zagefka

(2017). The effects of group memberships of victims and perpetrators in humanly caused disasters on charitable donations to victims. Journal of Applied Social Psychology, 47(8), 446–458. https://doi.org/10.1111/jasp.12452

39.

Jang

Kim

Lee

J. S.

Wann

D. L.

(2019). The impact of hedonic and meaningful messages on sport consumers’ responses to athlete foundations: A focus on fan identification level. Communication and Sport, 8(3), 346–363. https://doi.org/10.1177/2167479519830063

40.

Jetten

Spears

Manstead

A. S. R.

(1996). Intergroup norms and intergroup discrimination: Distinctive self-categorization and social identity effects. Journal of Personality and Social Psychology, 71, 1222–1233. https://doi.org/10.1037/0022-3514.71.6.1222

41.

Jetten

Spears

Postmes

(2004). Intergroup distinctiveness and differentiation: A meta-analytic integration. Journal of Personality and Social Psychology, 86, 862–879. https://doi.org/10.1037/0022-3514.86.6.862

42.

Jin

Zhu

(2020). How recipient group membership affects the effect of power states on prosocial behaviors. Journal of Business Research, 108, 307–315. https://doi.org/10.1016/j.jbusres.2019.10.052

43.

Kessler

J. B.

Milkman

K. L.

(2018). Identity in charitable giving. Management Science, 64(2), 845–859. https://doi.org/10.1287/mnsc.2016.2582

44.

Knapp

Hartung

(2003). Improved tests for a random effects meta-regression with a single covariate. Statistics in Medicine, 22(17), 2693–2710. https://doi.org/10.1002/sim.1482

45.

Y. J.

Rhee

Y. C.

Walker

Lee

J.-H.

(2013). What motivates donors to athletic programs: A new model of donor behavior. Nonprofit and Voluntary Sector Quarterly, 43(3), 523–546. https://doi.org/10.1177/0899764012472065

46.

Kogut

Ritov

Rubaltelli

Liberman

(2018). How far is the suffering? The role of psychological distance and victims’ identifiability in donation decisions. Judgment and Decision Making, 13(5), 458–466.

47.

Lay

Zagefka

González

Álvarez

Valdenegro

(2020). Don’t forget the group! The importance of social norms and empathy for shaping donation behaviour. International Journal of Psychology, 55(4), 518–531. https://doi.org/10.1002/ijop.12626

48.

Lee

Sargeant

(2011). Dealing with social desirability bias: An application to charitable giving. European Journal of Marketing, 45(5), 703–719. https://doi.org/10.1108/03090561111119994

49.

Levine

Cassidy

Brazier

Reicher

(2002). Self-categorization and bystander non-intervention: Two experimental studies. Journal of Applied Social Psychology, 32(7), 1452–1463. https://doi.org/10.1111/j.1559-1816.2002.tb01446.x

50.

Levine

Manning

(2013). Social identity, group processes, and helping in emergencies. European Review of Social Psychology, 24(1), 225–251. https://doi.org/10.1080/10463283.2014.892318

51.

Levine

Prosser

Evans

Reicher

(2005). Identity and emergency intervention: How social group membership and inclusiveness of group boundaries shape helping behavior. Personality and Social Psychology Bulletin, 31(4), 443–453. https://doi.org/10.1177/0146167204271651

52.

Levine

Thompson

(2004). Identity, place, and bystander intervention: Social categories and helping after natural disasters. The Journal of Social Psychology, 144(3), 229–245. https://doi.org/10.3200/SOCP.144.3.229-245

53.

Lichtenstein

D. R.

Drumwright

M. E.

Braig

B. M.

(2004). The effect of Corporate Social Responsibility on customer donations to corporate-supported nonprofits. Journal of Marketing, 68(4), 16–32. https://doi.org/10.1509/jmkg.68.4.16.42726

54.

Mitchell

T. L.

Haw

R. M.

Pfeifer

J. E.

Meissner

C. A.

(2005). Racial bias in mock juror decision-making: A meta-analytic review of defendant treatment. Law and Human Behavior, 29(6), 621–637. https://doi.org/10.1007/s10979-005-8122-9

55.

Moher

Shamseer

Clarke

Ghersi

Liberati

Petticrew

Shekelle

Stewart

L. A.

Altman

D. G.

Booth

Chan

A. W.

Chang

Clifford

Dickersin

Egger

Gøtzsche

P. C.

Grimshaw

J. M.

Groves

Helfand

. . .Whitlock

(2015). Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Reviews, 4(1), 1–9. https://doi.org/10.1186/2046-4053-4-1

56.

Nadler

(2002). Inter–group helping relations as power relations: Maintaining or challenging social dominance between groups through helping. Journal of Social Issues, 58(3), 487–502. https://doi.org/10.1111/1540-4560.00272

57.

Nadler

Chernyak-Hai

(2014). Helping them stay where they are: Status effects on dependency/autonomy-oriented helping. Journal of Personality and Social Psychology, 106(1), 58–72. https://doi.org/10.1037/a0034152

58.

Nadler

Harpaz-Gorodeisky

Ben-David

(2009). Defensive helping: Threat to group identity, ingroup identification, status stability, and common group identity as determinants of intergroup help-giving. Journal of Personality and Social Psychology, 97(5), 823–834. https://doi.org/10.1037/a0015968

59.

Paulin

M. J.

Ferguson

Jost

Fallu

J.-M.

(2014). Motivating millennials to engage in charitable causes through social media. Journal of Service Management, 25(3), 334–348. https://doi.org/10.1108/JOSM-05-2013-0122

60.

Paulin

M. J.

Ferguson

Schattke

Jost

(2014). Millennials, social media, prosocial emotions, and charitable causes: The paradox of gender differences. Journal of Nonprofit & Public Sector Marketing, 26(4), 335–353. https://doi.org/10.1080/10495142.2014.965069

61.

Puntoni

Sweldens

Tavassoli

N. T.

(2011). Gender identity salience and perceived vulnerability to breast cancer. Journal of Marketing Research, 48(3), 413–424. https://doi.org/10.1509/jmkr.48.3.413

62.

R Core Team. (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/

63.

Reese

Proch

Finn

(2015). Identification with all humanity: The role of self-definition and self-investment. European Journal of Social Psychology, 45(4), 426–440. https://doi.org/10.1002/ejsp.2102

64.

Shaker

G. G.

Nelson

(2022). A grounded theory study of major gift fundraising relationships in U.S. Higher Education. Nonprofit and Voluntary Sector Quarterly, 51, 1054–1073. https://doi.org/10.1177/08997640211057437

65.

Shang

Reed

Croson

(2008). Identity congruency effects on donations. Journal of Marketing Research, 45(3), 351–361. https://doi.org/10.1509/jmkr.45.3.351

66.

Shang

Sargeant

Carpenter

(2019). Giving intention versus giving behavior: How differently do satisfaction, trust, and commitment relate to them? Nonprofit and Voluntary Sector Quarterly, 48(5), 1023–1044. https://doi.org/10.1177/0899764019843340

67.

Slovic

(2007). “If I look at the mass I will never act”: Psychic numbing and genocide. Judgment and Decision Making, 2(2), 79–95.

68.

Smith

J. R.

Louis

W. R.

(2009). Group norms and the attitude–behaviour relationship. Social and Personality Psychology Compass, 3(1), 19–35. https://doi.org/10.1111/j.1751-9004.2008.00161.x

69.

Stephenson

A. L.

(2013). The effect of brand identification on alumni supportive behaviors [Ph.D. Thesis, Indiana University of Pennsylvania].

70.

Stephenson

A. L.

Bell

(2014). Motivation for alumni donations: A social identity perspective on the role of branding in higher education. International Journal of Nonprofit and Voluntary Sector Marketing, 19(3), 176–186. https://doi.org/https://doi.org/10.1002/nvsm.1495

71.

Stephenson

A. L.

Yerger

D. B.

(2014). Optimizing engagement: Brand identification and alumni donation behaviors. International Journal of Educational Management, 28(6), 765–778. https://doi.org/10.1108/IJEM-04-2013-0057

72.

Stürmer

Snyder

Kropp

Siem

(2006). Empathy-motivated helping: The moderating role of group membership. Personality and Social Psychology Bulletin, 32(7), 943–956. https://doi.org/10.1177/0146167206287363

73.

Stürmer

Snyder

Omoto

A. M.

(2005). Prosocial emotions and helping: The moderating role of group membership. Journal of Personality and Social Psychology, 88(3), 532–546. https://doi.org/10.1037/0022-3514.88.3.532

74.

Tajfel

(1978). Differentiation between social groups: Studies in the social psychology of intergroup relations. Academic Press.

75.

Terry

D. J.

Hogg

M. A.

(1996). Group norms and the attitude-behavior relationship: A role for group identification. Personality and Social Psychology Bulletin, 22(8), 776–793. https://doi.org/10.1177/0146167296228002

76.

van Leeuwen

. (2007). Restoring identity through outgroup helping: Beliefs about international aid in response to the December 2004 tsunami. European Journal of Social Psychology, 37(4), 661–671. https://doi.org/10.1002/ejsp.389

77.

Viechtbauer

(2010). Conducting meta-analyses in R with the metafor package. Journal of Statistical Software, 36(3), 1–48.

78.

Winterich

K. P.

Mittal

Aquino

(2013). When does recognition increase charitable behavior? Toward a moral identity-based model. Journal of Marketing, 77(3), 121–134. https://doi.org/10.1509/jm.11.0477

79.

Winterich

K. P.

Mittal

Ross

W. T.

Jr. (2009). Donation behavior toward in-groups and out-groups: The role of gender and moral identity. Journal of Consumer Research, 36(2), 199–214. https://doi.org/10.1086/596720

80.

Zagefka

James

(2015). The psychology of charitable donations to disaster victims and beyond. Social Issues and Policy Review, 9(1), 155–192. https://doi.org/10.1111/sipr.12013

81.

Zagefka

Sun

(2021). Ingroup identification moderates blame attributions for the COVID-19 crisis, and willingness to help ingroup and outgroup members. Analyses of Social Issues and Public Policy, 21(1), 1202–1221. https://doi.org/10.1111/asap.12281

Social Identification and Charitable Giving: A Systematic Review and Meta-Analysis

Abstract

Keywords

Social Identity Theory

Charitable Triad Theory

The Current Study

Method

Search Strategy

Eligibility and Screening

Coding

Analyses

Sensitivity Analyses

Publication Bias

Results

Description of the Literature

Meta-Analysis

Overall Association

Heterogeneity

Moderator Analyses

Strength of Identification versus Shared Identity

Target of Identification

Self-Reported Versus Objective Giving

Other Moderators

Discussion

Strengths, Limitations, and Future Directions

Managerial Implications

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

Notes

Author Biographies

References