Where There Is a (Collective) Will,There Are (Effective) Ways

Abstract

The 21st century has borne witness to catastrophic natural and human-induced tragedies. These disasters necessitate humanitarian responses; however, the individual and collective bases of support are not well understood. Drawing on Duncan’s motivational model of collective action, we focus on how individual differences position a person to adopt group memberships and develop a “group consciousness” that provides the basis for humanitarian action. Longitudinal mediation analyses involving supporters of international humanitarian action (N = 384) sampled annually for 3 years provided support for the hypothesized model, with some twists. The results revealed that within time point, a set of individual differences (together, the “pro-social orientation”) promoted a humanitarian group consciousness that, in turn, facilitated collective action. However, longitudinally, there was evidence that a more general pro-social orientation undermined subsequent identification with, and engagement in, the humanitarian cause. Results are discussed in terms of understanding the interplay between individual and group in collective actions.

Keywords

collective action social identification values human rights personal political salience generosity humanitarian action

Global disparities in wealth are large and increasing (e.g., Milanovic, 2005). These entrenched global inequalities have been exacerbated in recent years by a series of humanitarian disasters, both natural and man-made. In 2015, the world witnessed the devastating Nepalese earthquake that killed an estimated 9,000 people and the continuing civil war in Syria is creating a humanitarian crisis that is spreading across the world (Reese, Rosenmann, & McGarty, 2015). These disasters have motivated humanitarian responses. However, little research directly considers the antecedents of support for humanitarian assistance, especially as it relates to entrenched global inequalities (Olson, 1997). Thus, while there are rich literatures considering the antecedents of collective action among disadvantaged or minority group members (women, sexual minorities, Black people; see van Zomeren, Postmes, & Spears, 2008, for a review and meta-analysis), there is a relative paucity of studies that explore the motivators of action among privileged or advantaged group members and international humanitarian action specifically (Pittinsky & Diamante, 2015). The current research addresses this gap in the specific context of global poverty reduction.

A number of different perspectives are needed to understand the complex antecedents of humanitarian action. Individuals form their own attitudes and act on their values to volunteer their time, provide donations and petition governments (e.g., see Bolitho, Carr, & Fletcher, 2007; Darnton & Kirk, 2011; Zagefka & James, 2015). However, individuals also work to address humanitarian issues in concert with others as members of non-governmental organizations (NGOs) or as part of a social movement (e.g., Oxfam, Make Poverty History; see Thomas, McGarty, & Mavor, 2010). It follows that to understand engagement in humanitarian collective action, we must consider not only the role of generalized, stable individual differences in ideologies, values, and motivations in shaping engagement with this agenda, but also how these individual differences are given expression through belonging to groups and co-acting with other people. Consistent with these points, Duncan (2012) recently argued that “integrating individual differences variables into the study of motivation for collective action allows a deeper, more complex understanding of this motivation and can explain why some group members develop group consciousness and become politically active, whereas others do not” (p. 781). Indeed, although the importance of considering both levels is often recognized in reviews on, for example, charitable behavior (Bekkers & Wiepking, 2011; Wiepking & Bekkers, 2012) or responses to humanitarian emergencies (Zagefka & James, 2015), empirical integrations of the two are rare. Accordingly, in the current research, we focus on the relationships between factors that drive people to support humanitarian collective action at the individual and group levels and how these interrelate over time to promote (or undermine) commitment to a humanitarian cause.

The current research draws on Duncan’s (1999, 2012) model of motivation for collective action to develop an integrative analysis of the individual and social-psychological motivators of humanitarian action. Duncan’s (1999, 2012) model is comprised of three key components: personality and life experiences, group consciousness, and collective action. We conceptualize collective action broadly, as acts taken by group members to improve the conditions of an entire group (as per Wright, Taylor, & Moghaddam, 1990) of disadvantaged others. In the case of humanitarian action, this not only includes acts of collective generosity (i.e., donations to humanitarian NGOs, volunteering time) but also strategic efforts to demonstrate support to political leaders (i.e., signing petitions, attending rallies, and the like). Personality and life experiences are conceptualized as distal antecedents for the development of group consciousness, which then forms the more proximal platform for engagement in coordinated, collective actions to address social injustice (Duncan, 1999, 2012). This model thus provides a resource for integrating the insights across the charity, donation, human rights, and collective action literatures, integrating individual differences and group processes approaches, as well as accounting for distal (relatively stable individual differences) and proximal explanations (relating to group formation around a specific cause or issue) for action.

Individual Differences and Humanitarian Action

There exist some excellent recent reviews of the role of individual differences in charity or philanthropy (e.g., Bekkers & Wiepking, 2011), engagement with a human rights agenda (e.g., McFarland, 2015), and the motivation to engage in collective action (Duncan, 2012). These reviews have identified a consistent set of individual differences predictors that form the basis for our hypotheses about the role of individual differences in promoting support for humanitarian collective action.

Ideology: Social Dominance Orientation (SDO)

SDO is defined as “the extent that one desires that one’s ingroup dominance and be superior to outgroups” (Pratto, Sidanius, Stallworth, & Malle, 1994, p. 742). It is expected that people who endorse socially dominant ideology would be less motivated to engage in support for people in developing countries because preserving the status quo allows them to maintain their group-based status and power. Previous research supports this notion: Reese, Proch, and Cohrs (2014) found that SDO (along with right wing authoritarianism and other personality indicators) was a negative predictor of intention to support global poverty reduction efforts (see also Reese, Berthold, & Steffens, 2012), and of the view that global inequality was unjust. Other research has found that SDO is a negative predictor of human rights behaviors (e.g., Cohrs, Maes, Moschner, & Kielmann, 2007; McFarland & Mathews, 2005).

Values and Human Rights Attitudes

Values are “abstract beliefs about desirable ways of behaving and desirable general goals that transcend specific objects and situations, that vary in their importance for individuals, and that serve as guiding principles in people’s lives” (Feather, Woodyatt, & McKee, 2012, p. 517). While values are understood to be more abstract ideals, attitudes are more specific or proximal indicators (e.g., Schwartz, 1992). That is, values often underpin specific attitudes. Universalism values are defined as values that reflect understanding, appreciation, tolerance, and protection for the welfare of all people and for nature (Schwartz, 1992). For these reasons, universalism values are strongly associated with support for humanitarian action on global poverty (McFarland, 2015; see also Darnton & Kirk, 2011, for a practitioner perspective).

Moreover, research has found that universalism values are moderate predictors of human rights attitudes and action (Cohrs et al., 2007) as well as for support for reparation to Indigenous Australians (Feather et al., 2012). We therefore consider universalism values (Schwartz, 1992) as well as specific attitudes about human rights (e.g., McFarland & Mathews, 2005) as individual-level predictors of support for humanitarian action.

Personal Political Salience (PPS)

Given that humanitarian action is an issue that is also on the inter-governmental agenda of governments of wealthy countries, we also consider the role of PPS. Duncan (1999) posited that PPS—or the degree to which one attaches personal meaning to political events—is a key antecedent to engagement in activism. Similarly, here, we expect that PPS will be a key individual difference predictor of humanitarian action. Thus, we expect that to the extent that individuals view contemporary social and political issues (e.g., climate change, the rights of indigenous persons and the LGBTI [lesbian, gay, bisexual, trans, and/or intersex] community; see Lavine, Sullivan, Borgida, & Thomsen, 1996) as meaningful to them personally (PPS), then this should be an antecedent to membership of politicized groups and promote humanitarian action in support of the global poverty agenda. Indeed, research on feminist activism has demonstrated that those people who endorse political events (wars, social movements) as personally meaningful and relevant to them are more likely to identify with politicized (activist) groups and engage in activism (e.g., Curtin, Stewart, & Duncan, 2010; Duncan & Stewart, 2007).

To summarize, our focus targets those factors repeatedly identified as key individual differences antecedents of charity, human rights and collective action, to paint a richer picture of the kinds of people who engage in action for social change (see also Curtin et al., 2010). Our approach spans SDO, universalism values, attitudes, and PPS. In the current research, we conceptualize these variables as an integrated cluster of variables, modeled as a latent “pro-social orientation” factor. This is consistent with previous research on the pro-social personality (e.g., Penner, 2002) and the generous disposition (Collett & Morrissey, 2007) and takes into account that multiple causal orderings of these individual differences variables are possible and, indeed, likely.

However, we go further to suggest that we need to look beyond individual differences in developing explanations of humanitarian collective action to support people in developing countries. Personality and individual differences explanations distinguish those who are active from those who are not but do not tell us why these individual differences are associated with collective action (Duncan, 2012). Consistent with research that identifies a group-based element to overcoming the disadvantage suffered by people in developing countries (e.g., Reese et al., 2012; Thomas & McGarty, 2009), we need to take account of the ways that people act as group members to achieve joint goals on specific social issues or problems. Consistent with this argument (and following Duncan, 1999, 2012), we conceive of these individual differences as antecedents to the formation of a specific, humanitarian group consciousness, where group consciousness is based on psychological group membership (Turner, Oakes, Haslam, & McGarty, 1994).

Group Consciousness and Humanitarian Action

Group consciousness is defined as “identification with a group in which an individual recognizes the group’s position in a power hierarchy, rejects rationalizations of relative positioning, and embraces a collective solution to group problems” (Duncan, 1999, p. 612). For Duncan (2012), group consciousness is analogous to politicized social identification (Simon & Klandermans, 2001), and her conceptualization of group consciousness draws heavily on recent collective action research. The social identity model of collective action (SIMCA; van Zomeren et al., 2008) identifies three key drivers of action: perceived injustice, group efficacy, and social identification.

Perceived Injustice and Group Efficacy

Contemporary research on collective action emphasizes the importance of perceptions of injustice and beliefs about the effectiveness of joint or group action (group efficacy, following Bandura, 1997; van Zomeren et al., 2008; van Zomeren, Spears, Fischer, & Leach, 2004). Put simply, people must perceive a situation as illegitimate and believe that group action can be successful in redressing disadvantage, to be motivated to take action. These antecedents are more powerful predictors of collective action when they are group-based appraisals rather than located in individualist explanations (H. J. Smith & Ortiz, 2002) and when the reactions are affective (i.e., emotions such as anger or outrage) rather than cognitive (van Zomeren et al., 2008). Thus, whereas political issues can be experienced as personally relevant and meaningful at the individual level (PPS, SDO, values), they are likely to be more proximal predictors of action at the group level (reflected by affective reactions to injustice and group-level efficacy).

Social Identification

There is also evidence consistent with the idea that social identification underpins humanitarian collective action. According to the insights of the social identity approach (and in particular, self-categorization theory, Turner, Hogg, Oakes, Reicher, & Wetherell, 1987), individuals can operate at three discrete levels of abstraction: as a unique individual (“me,” “I”), as a group member (“us,” “them”), and as a human. When individuals identify at the broadest level of abstraction they can identify with their fellow humans, who then become a psychological part of their ingroup. According to McFarland (2015), identification with humanity reflects a “deep concern and caring for all human beings,” (p. 19) a sense that all humans are a part of one’s family.

McFarland, Webb, and Brown (2012) suggested that identification with all of humanity is a key driver of human rights knowledge, concern, and behavior. Indeed, Reese et al. (2014) showed that identification with the whole world predicts support for global poverty reduction, over and above relevant individual differences (SDO, right wing authoritarianism, justice sensitivity). Reese and Kohlmann (2015) also demonstrated that global identification shaped the consumption of fair trade products because it is associated with the belief that global inequality is unjust. Furthermore, Reese, Proch, and Finn (2015) experimentally demonstrated that higher global identification resulted in global donation behavior (i.e., pledging to UNICEF), compared with lower global identification, suggesting a causal effect of all-human identification on humanitarian (collective) action.

Moreover, the social identity approach also recognizes that group memberships (i.e., perceiving one’s self as a member of groups) play an important role in prescribing attitudes, beliefs, and behaviors (Turner et al., 1994). When a particular social identity becomes salient, group members adopt the relevant norms, values, and beliefs that define that group. In this way, group membership gives collective meaning to what would otherwise be an individual attribute (Baray, Postmes, & Jetten, 2009). Building on this insight, Bliuc, McGarty, Reynolds, and Muntele (2007; McGarty, Bliuc, Thomas, & Bongiorno, 2009) suggested that group memberships based around shared opinions about how the world should be may be uniquely positioned to understand the psychological bases of collective action (also L. G. E. Smith, Thomas, & McGarty, 2015). To the extent that one identifies with an opinion-based group (e.g., anti-poverty; Thomas et al., 2010) that group membership prescribes norms for opinion and action. Such groups help us to develop an analysis of action across geographical and psychological boundaries because the identity is about relations between those groups rather than being based in those group memberships (see Bliuc, McGarty, Hartley, & Muntele Hendres, 2012). Indeed, Bliuc et al. (2015) recently showed how climate change believers and climate change skeptics have competing forms of group consciousness in that they could be said to be united only in their opposition to each other. Such group memberships also help us to understand allied collective action (i.e., collective action taken by members of advantaged groups; see Thomas, Mavor, & McGarty, 2012).

Thus, the social identity approach implicates two different ways that social identification may promote humanitarian engagement and action: global- or human-level identification and opinion-based identification. Thomas et al. (2015) recently compared the two types of identification as predictors of engagement in the viral online campaign Kony2012. It was found that opinion-based identification was a stronger predictor of engagement in humanitarian action on this campaign than global identification. The authors argued that opinion-based groups may more readily embody norms for action than broader categories such as global- or human-level identification (see also, Reese et al., 2012). In the current research, we therefore focus on an opinion-based group defined by support for humanitarian action to end global poverty.

In sum, our analysis posits group consciousness as the conceptual and psychological link between individual differences and humanitarian action, allowing us to consider the ways in which coordinated action to achieve justice for people in developing countries is likely to be premised in group membership. Although it is well recognized that social identities facilitate engagement in collective action (van Zomeren et al., 2008), there is little literature that considers the role of individual differences in giving rise to group memberships and only a handful of studies that test this process in relation to advantaged group collective action (e.g., Saab, Tausch, Spears, & Cheung, 2015; Thomas et al., 2012; van Zomeren, Postmes, Spears, & Bettache, 2011). As it has been demonstrated that multiple causal orderings of the identity, injustice, and efficacy variables are possible (Thomas et al., 2012; Thomas et al., 2015) and consistent with Duncan’s (1999, 2012) definition of group consciousness, we conceptualize these variables as an integrated cluster of variables modeled as a latent factor.

The Current Research

The current research adopts an integrative approach to consider the interplay between individual differences and group processes, distal and proximal factors, in understanding humanitarian action. We use a longitudinal survey of anti-poverty supporters sampled annually over 3 years to explore how a more general set of individual differences can position a person to adopt discrete group memberships; how one’s individual values, beliefs, and ideologies find expression and are given meaning, through the development of a specific group consciousness. Such an integration is timely and necessary because, as others have pointed out (see Duncan, 2012, in relation to collective action; McFarland, 2015, in relation to human rights), there is a need not only for an overarching theoretical framework to (a) bring together empirical findings about personality and individual differences, on one hand, and social-psychological group processes, on the other, but also (b) to acknowledge both the individual and group processes at play in generosity and collective action for social change.

Three waves of data allow for an examination of cross-sectional (contemporaneous) relations at each wave as well as the examination of changes between each wave (longitudinal). As Mackinnon (2008) pointed out, this is important because changes within a person can be different from changes between people. We test the primary hypothesis that it is the shift from personal to social that transforms individual into group behavior (Turner et al., 1994). We seek to map the between-person (contemporary) and within-person (longitudinal) processes of that shift (see also, Turner-Zwinkles, van Zomeren, & Postmes, 2015, for a complementary approach). Consistent with Duncan’s (2012) integrative model, we expect that humanitarian collective action will be predicted by a set of factors that capture individual differences in relation to a general tendency to endorse egalitarian, pro-social worldviews (reflecting low levels of SDO, endorsing social justice values and human rights attitudes, and expressing PPS; together a “pro-social orientation”) through group consciousness, a latent factor indicated by social identification with, and self-definition of, a humanitarian opinion-based group efficacy and perceived injustice. Thus, we expect that group consciousness will mediate the association between individual differences and humanitarian action both within time points (contemporaneous mediation) and over time (longitudinal mediation).

Humanitarian action was measured by a general measure of self-reported action, as well as two measures of observed humanitarian action. Specifically, participants were invited to allocate US$5 across three causes (global poverty reduction, domestic poverty reduction, and animal welfare); participants were not given the option of keep the money themselves. The allocation thus measures willingness to financially support global poverty specifically, relative to other charitable causes (see Darnton & Kirk, 2011). As a behavioral measure of socio-political action in line with collective action research, participants were invited to sign a letter addressed to the Australian Foreign Minister outlining their support for international development. Consistent with Duncan’s (2012; also Drury & Reicher, 2000; van Zomeren, Leach, & Spears, 2012) theorizing that there could also be reverse (feedback) effects from action to subsequent individual differences and group consciousness, we test whether these contribute to model fit: over and above the forward mediation process described earlier.

Method

Participants

Respondents were recruited as part of a larger, longitudinal project exploring the nexus of personality, social, and political factors of support for global poverty reduction. Participants were initially recruited through an electronic link distributed through the mailing lists of two global poverty NGOs and signed up to participate in the longitudinal research. The data reported here are from the second (referred to here as Time 1), third (referred to here as Time 2), and fourth (referred to here as Time 3) waves of this study, as these were the first waves to include comprehensive measures of individual differences.

Of the 571 people who participated at Time 1, there were 408 at Time 2, and 436 people participated at Time 3. Given the longitudinal approach, we retained only those participants for whom we had responses on at least two of the three occasions (leaving N = 384). We conducted a missing values analysis involving all of the key measures, as well as demographic variables (age, gender, and educational level), as these can help inform patterns of missing data (Enders, 2013). Modeling suggested the data (11%-13% for Time 1 variables, 19%-22% for Time 2 variables, 28% for Time 3 variables) were missing completely at random, χ²(1,001) = 1,056.21, p = .11, and we, therefore, addressed missing data using full information maximization likelihood (FIML) in MPlus Version 7.2. FIML does not impute values but estimates parameters using all the information that is already contained in the incomplete data set. It has been shown to produce unbiased estimates when the amount of missing data is not large (<50%) and is missing completely at random (Enders, 2013). This yielded a sufficient sample size to detect effects of a small-medium magnitude (.2-.3) with power = .80, α = .05, in models containing six latent variables and 29 observed variables (Soper, 2016).

At Time 1, participants were aged between 18 and 82 years (M = 49.79, SD = 14.50 years) and were primarily female (60.1%; 47 people did not indicate their gender). All but four people reported that they were Australian citizens or permanent residents. This sample was also highly educated, with most reporting a bachelor’s degree (29.43%) or higher (34.90%) while a smaller group had completed vocational training (6.77%) or an undergraduate diploma (8.07%); a minority had completed secondary school (6.00%) or had not completed secondary school (2.08%).

Measures and Procedure

Participants were sent an e-mail inviting them to participate in research on “Personality and Social Attitudes Towards Global Poverty.” The e-mail included a link to a secure web server. Participants were told that the questionnaire had three parts and completed items measured (primarily) on a Likert-type scale, which ranges from 1 to 7. Measures were identical at each of the three time points.

The pro-social orientation

We operationalized the key aspects of individual differences with measures of SDO, values, human rights attitudes, and PPS. Responses to each of these aggregated variables were parceled together and modeled as reflective indicators of the pro-social orientation. Parceling the items in this way deals with measurement error and allows for a more parsimonious model than if we used all the items separately and makes parameter estimates more reliable (Hall, Snell, & Foust, 1999).

SDO

Three items measured SDO (Pratto et al., 1994): “Some groups of people are simply inferior to other people,” “It’s OK if some groups have more of a chance in life than others,” and “It would be good if groups could be equal” (R). Reliability was poor (α = .53), probably because of the nature of the sample and floor effects on the items. As analyses involving the individual items did not affect results, we, nevertheless, created an aggregated scale to maintain consistency with previous research using this scale. To maintain consistency with the other items, SDO was reverse scored such that more positive scores reflect a preference for social equality.

Values

We adapted measures universalism values from Schwartz (1992). Participants were asked to rate three universal value items (equality, social justice, a world at peace; α = .75) based on how important they were as a guiding principle in their life (1 = not important, 7 = extremely important).

Human rights attitudes

Participants indicated how much they agreed with two statements adapted from the Human Rights Attitudes Scale (Diaz-Veizades, Widaman, Little, & Gibbs, 1995): “People are entitled to have the food, housing and medical care necessary to maintain their health and well-being” and “Everyone has the right to an adequate standard of living,” α = .87.

PPS

Following Lavine et al. (1996), participants indicated how important to them personally each of the following issues was: climate change, the rights of gay and lesbian people, global poverty, the rights of Australia’s indigenous people, the treatment of asylum seekers, equality between men and women (1 = not at all important, 7 = very important). Note that Duncan and Stewart’s (2007) research on PPS adopted a measure based on the salience of political or social events (wars, social movements) whereas the measure adopted here assesses meaning attached to specific causes (following Lavine et al., 1996). Responses were aggregated, and higher scores indicate a higher level of PPS, α = .82.

Group consciousness

Consistent with Duncan’s (2012) conceptualization of group consciousness as incorporating identification with groups that can mobilize action, perceptions of injustice, and the belief in the groups’ ability to act effectively, group consciousness is represented as a reflective latent factor comprised of the items.

Social identification and self-definition

Identification with the humanitarian opinion-based group was measured with three items, α = .85: “I see myself as a supporter of efforts to end global poverty,” “I am pleased to be a supporter of efforts to end global poverty,” and “I identify with other supporters of efforts to end global poverty” (following Cameron, 2004). These were supplemented with two items, adapted from Baray et al. (2009), assessing the degree to which the group is self-defining: “My support for efforts to end global poverty is important in shaping my personality” and “Being someone who supports efforts to end global poverty makes me who I am.” Given the importance of social identification and the social self-definition to the group consciousness factor, these were parceled as two indicators.

Affective injustice

As van Zomeren et al. (2008) demonstrated that affective reactions to injustice were stronger predictors than cognitive reactions, two items adapted from Thomas et al. (2012) measured affective reactions to injustice as outrage. The items were prefixed with the statement: “Thinking about the situation in developing countries, I feel: outrage [anger],” α = .86.

Group efficacy

Two items adapted from Thomas et al. (2012) measured the belief in the group’s ability to act effectively to redress global poverty: “Together supporters of efforts to end global poverty can improve the outcomes for people in developing countries” and “Together supporters of efforts to end global poverty can make a positive difference for people in developing countries,” α = .93.

Self-reported humanitarian action and observed behavior

Self-reported humanitarian action

Participants indicated the frequency with which they engaged in 12 actions to support people in developing countries (1 = never, 7 = daily). Consistent with our broad conceptualization, the items included acts of personal generosity (e.g., “Make a donation to a global poverty charity or organization”) as well as political actions (e.g., “Sign a petition relating to global poverty reduction,” “Write a letter to a member of Parliament outlining support for poverty reduction,” “Share a post on Facebook or Twitter about global poverty”), α = .82.

Observed behavior: Donation and letter

The self-reported anti-poverty action was supplemented with two measures of observed humanitarian action. First, as a behavioral measure of collective generosity on this issue, participants were each given US$5 as reimbursement for participation but were asked to allocate this sum across three NGOs: an NGO that seeks to address global poverty, an animal welfare NGO, and an NGO that seeks to combat domestic (i.e., within Australia) poverty. Participants could allocate proportions freely across the three groups, but their allocations had to sum to US$5. This measure is thus a measure of generosity on the issue of global humanitarian action relative to other worthy causes. Second, as a behavioral measure of socio-political action, participants were presented with two letters, addressed to the Australian Foreign Affairs Minister, advocating their support for, or opposition to, Australia’s expressed commitment to the Millennium Development Goals. These were coded −1 (opposed to Australia’s commitment), 0 (signed neither letter), and 1 (support Australia’s commitment). The letter signing variable was not available at Time 3.

Results

To test the integrative model of humanitarian action, we conducted structural equation modeling (SEM) using MPlus Version 7 (Muthén & Muthén, 1998-2012). Given that the chi-square test is unreliable with large samples, we considered three other indices to assess model fit: The root mean square error of approximation (RMSEA) is one of the most widely reported and informative criteria in covariance structure modeling, where values of .01, .05, and .08 indicate excellent, good, and acceptable fit, respectively; the standardized root mean square residual (SRMR), where a value of less than .08 is generally considered evidence of good fit; and the comparative fit index (CFI), where values approaching .95 indicate good model fit (Hu & Bentler, 1999).

Construct Validation and Measurement Invariance

We first sought to test whether (a) the pro-social orientation and group consciousness are indeed valid, discrete factors and, if so, (b) whether the factor structure is stable over time. We first subjected all our measures relating to individual and group factors to an exploratory factor analysis. Results supported a two-factor solution over a one-factor solution for all three time points, Δχ²(7) = 101.67, p < .001 for Time 1, Δχ²(7) = 133.13, p < .001 for Time 2, and Δχ²(7) = 139.25, p < .001 for Time 3. Consistent with our conceptualization of these as discrete components, social identification, self-definition as a group member, and group efficacy loaded strongly and significantly onto the first “group” factor (geomin rotated loadings = .53-.81 for Time 1, .55-.92 for Time 2, and .55-.87 for Time 3), while human rights, PPS, values, and SDO loaded strongly onto a second “person” factor (geomin rotated loadings = .50-.84 for Time 1, .57-.81 for Time 2, .54-.89 for Time 3). Unexpectedly, affective injustice loaded significantly onto the person factor (.43) at Time 1 and cross-loaded onto both factors at Times 2 and 3. Given that the injustice component of the group consciousness played an ambiguous role in distinguishing between the person and group factors here, we omitted it from further tests.

We next conducted a confirmatory factor analysis to validate the two-factor solution overall and to verify whether it fits in the same way across the three time points. Measurement invariance is a key precondition for longitudinal methods: It is established by increasingly constraining the parameters for models across time points to be the same and then observing whether the chi-square value changes significantly (deterioration is a sign of undesirable variability in measurement; see Vandenberg & Lance, 2000). The group consciousness factor was indicated by social identification, self-definition as a group member, and group efficacy at Times 1 to 3. The pro-social orientation was indicated by human rights, PPS, values, and SDO at Times 1 to 3. The latent factors were allowed to correlate with each other, and the same indicator was allowed to correlate with itself across time. Table 1 displays the fit statistics for the models that were tested as well as an explanation of the forms of invariance testing.

Table 1.

Fit Statistics for Tests of Measurement Invariance.

		χ²(df), p	RMSEA [90% CI]	CFI	SRMR
Configural invariance	Same pattern of significant indicator to factor loadings over time	χ²(153) = 259.58, p < .001	.04 [.03, .05]	.97	.05
Metric invariance	Factor loadings are the same over time	χ²(163) = 266.25, p < .001	.04 [.03, .05]	.97	.05
Scalar invariance	The intercepts are the same over time	χ²(173) = 281.51, p < .001	.04 [.03, .05]	.97	.05
Residual invariance	The residual variances are the same over time	χ²(187) = 303.74, p < .001	.04 [.03, .05]	.97	.07

Note. RMSEA = root mean squared error of approximation; CI = confidence interval; CFI = comparative fit index; SRMR = standardized root mean square residual.

Model fit for the initial model was good establishing configural invariance (Table 1). We then constrained the factor loadings to be the same. Model fit was good (Table 1), and there was no significant deterioration of model fit, confirming metric invariance, Δχ²(10) = 6.67, p = .76. We next constrained the intercepts to be the same across time. Model fit was good (Table 1), and scalar invariance was confirmed, Δχ²(10) = 15.27, p = .12. Finally, we tested residual invariance by constraining the residual variances to be the same. Model fit was again satisfactory (Table 1), and residual invariance was confirmed, Δχ²(14) = 22.27, p = .07. Thus, we established full measurement invariance for the two factors across the three time points. This measurement model was then used as the basis for our tests of structural relations between variables.

Table 2 displays the descriptive information (means, standard deviations) and correlations for the latent factors over time, as well as the outcome variables (self-reported action, donation allocation, and letter signing). It can be seen not only that there were strong, significant correlations between the factors over time but also that both factors were significant small-moderate predictors of the key outcomes (except for the Time 1 donation allocation that was marginal).

Table 2.

Means (SDs) and Correlations Between Person and Group Latent Variables and Outcome Measures.

	M (SD)	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Pro-social orientation T1	4.02 (0.46)	—	.91***	.74***	.66***	.59***	.51***	.31***	.28***	.23***	.07*	.15***	.15***	.19***	.17***
2. Pro-social orientation T2	3.93 (0.44)		—	.82***	.60***	.67***	.58***	.28***	.33***	.27***	.07^†	.15***	.16***	.18***	.22***
3. Pro-social orientation T3	4.14 (0.50)			—	.49***	.55***	.63***	.23***	.27***	.09	.06^†	.12***	.21***	.14***	.18***
4. Group consciousness T1	3.78 (0.59)				—	.90***	.81***	.47***	.42***	.27***	.11*	.23***	.24***	.29***	.26***
5. Group consciousness T2	3.79 (0.67)					—	.90***	.39***	.46***	.30***	.10*	.23***	.26***	.26***	.32***
6. Group consciousness T3	3.93 (0.70)						—	.35**	.41***	.27***	.09*	.20***	.32***	.24***	.29***
7. Self-report action T1	2.26 (0.71)							—	.80***	.47***	.05^†	.11**	.11**	.14**	.10**
8. Self-report action T2	2.31 (0.74)								—	.58***	.05^†	.11***	.42***	.12***	.15***
9. Self-report action T3	2.09 (1.01)									—	.03	.11**	.08**	.08***	.09***
10. Donation allocation T1	2.96 (1.95)										—	.26***	.12***	.03	.03
11. Donation allocation T2	3.13 (2.06)											—	.42***	.07**	.07**
12. Donation allocation T3	3.34 (1.89)												—	.07**	.08**
13. Letter signing T1	41.9%													—	.32***
14. Letter signing T2	38.5%														—

Note. The correlations are based on the values estimated using FIML (as this is what was used in the structural equation models); the means and standard deviations are based on the data without the estimated missing values. Letter signing is a categorical variable so the value denotes the percentage of people who signed the letter supporting humanitarian commitment.

†

p < .06. *p < .05. **p < .01. ***p < .001.

Testing the Integrative Model of Humanitarian Collective Action

We first tested a model in which contemporaneous and longitudinal mediation were specified, over and above autoregressive effects (Mackinnon, 2008). The pro-social orientation at Time 1 predicted the pro-social orientation at Time 2, as well as contemporaneous (Time 1) group consciousness and subsequent (Time 2) group consciousness; similarly, the pro-social orientation at Time 2 predicted the pro-social orientation at Time 3, as well as contemporaneous (Time 2) and subsequent (Time 3) group consciousness. Group consciousness at Time 1 predicted group consciousness at Time 2 as well as contemporaneous (Time 1) subsequent (Time 2) actions (self-reported action, donation allocation, and letter); similarly, group consciousness at Time 2 predicted group consciousness at Time 3 as well as contemporaneous (Time 1) subsequent (Time 2) actions (self-reported action, donation allocation, and letter). The items were allowed to correlate with themselves over time (i.e., SDO at Time 1 correlated with that at Times 2 and 3), and each of the actions was predicted by its earlier measurement (autoregressive effects).

The model fit well (see Table 3, Model 1, for fit statistics). Having established good fit for the “pure” mediational model, we tested three additional, theoretically explicable models to see if the inclusion of additional pathways significantly contributed to model fit. First, on the basis that group consciousness could also feedback to inform the pro-social person (group in person), we tested a reverse causal model whereby the pro-social orientation factor at Times 2 and 3, respectively, was shaped by prior group consciousness at Times 1 and 2, respectively. Including these paths did not significantly improve model fit, Δχ²(2) = 0.48, p > .05 (see Table 3, Model 2).

Table 3.

Fit Statistics for Structural Tests of the Integrative Model of Humanitarian Collective Action.

		χ²(df), p	RMSEA [90% CI]	CFI	SRMR
Model 1	Pure mediation model. Contemporary and longitudinal mediation specified, autoregressive paths.	χ²(375) = 613.55, p < .001	.040 [.035, .046]	.945	.064
Model 2	First alternative model testing whether group consciousness impacts subsequent pro-social orientation	χ²(373) = 613.07, p < .001	.040 [.035, .047]	.945	.064
Model 3	Second alternative model testing whether action affects subsequent orientation and group factors	χ²(363) = 589.25, p < .001	.040 [.034, .046]	.948	.063
Model 4	Third alternative model testing whether the orientation factor predicts contemporaneous and longitudinal action (over and above the group factor)	χ²(362) = 574.67, p < .001	.039 [.033, .045]	.951	.061
Model 5	Final model. Contains the paths specified in Model 1, as well as three additional significant paths identified in Models 3-4	χ²(372) = 580.28, p < .001	.038 [.032, .044]	.952	.064

Note. RMSEA = root mean squared error of approximation; CI = confidence interval; CFI = comparative fit index; SRMR = standardized root mean square residual.

Second, on the basis that engagement in prior action might affect subsequent pro-social orientation and group consciousness (as per the related arguments of Drury & Reicher, 2000), we allowed for paths between each of the three actions and the subsequent pro-social orientation and group consciousness factors (see Table 3, Model 3, for fit statistics). Adding these paths produced a better fitting model, Δχ²(12) = 24.30, p = .02; however, only the path from self-reported action (Time 1) to group consciousness (Time 2) was significant.

Third, we tested whether the pro-social orientation factor predicts contemporaneous and longitudinal variation in the action outcomes, over and above the group consciousness factor. We therefore allowed the pro-social orientation factor to predict each of the three actions not only within each time point but also over time (i.e., we specified the same relationships for the pro-social orientation factors as we had for the group consciousness factors.) These inclusions also produced a significantly better fitting model, Δχ²(13) = 38.88, p < .001 (see Table 3, Model 4). Of the paths that contributed significantly to model fit, it was the prediction of self-reported action at Time 3 that was predicted by the pro-social orientation at Time 2 (β = .69, p = .004) and Time 3 (β = −.71, p = .002). All the other paths were non-significant.

Our final model therefore included the “pure” mediational model that we specified originally, as well as the significant paths identified in our tests of the alternative models: specifically, the path from self-reported action (Time 1) to group consciousness (Time 2) derived from our test of Model 3 and those from the pro-social orientation at Times 2 and 3, respectively, to self-reported action at Time 3 derived from our test of Model 4. The model fit was good (see Table 3, Model 5). Figures 1 and 2 display the final model with standardized regression coefficients; for ease of interpretation, we present the measurement (Figure 1) and structural models (Figure 2) separately. Table 4 displays the tests of the indirect effects of the pro-social orientation on the action outcomes by group consciousness, for the key paths relating to contemporaneous and longitudinal mediation.

Figure 1.

Measurement model for the pro-social person and group consciousness factors at each of the three time points.

Figure 2.

Structural model for the longitudinal test of the integrative model of humanitarian collective action.

Table 4.

Tests of the Indirect Effects in the Final Model of the Integrative Model of Humanitarian Collective Action.

Effect	Indirect effect (SE), p value
Contemporaneous mediation
Pro-social orientation 1 → Group consciousness 1 → Letter signing 1	.31 (.04), p < .001
Pro-social orientation 1 → Group consciousness 1 → Donation allocation 1	.07 (.04), p = .05
Pro-social orientation 1 → Group consciousness 1 → Self-reported action 1	.19 (.04), p < .001
Pro-social orientation 2 → Group consciousness 2 → Letter signing 2	.36 (.14), p = .01
Pro-social orientation 2 → Group consciousness 2 → Donation allocation 2	.08 (.13), p = .56
Pro-social orientation 2 → Group consciousness 2 → Self-reported action 2	.45 (.12), p < .001
Pro-social orientation 3 → Group consciousness 3 → Donation allocation 3	.41 (.17) p = .02
Pro-social orientation 3 → Group consciousness 3 → Self-reported action 3	.32 (.19), p = .09
Longitudinal mediation
Pro-social orientation 1 → Group consciousness 2 → Letter signing 2	−.33 (.14) p = .01
Pro-social orientation 1 → Group consciousness 2 → Donation allocation 2	−.07 (.12) p = .57
Pro-social orientation 1 → Group consciousness 2 → Self-reported action 2	−.41 (.12) p = .001
Pro-social orientation 1 → Group consciousness 2 → Donation allocation 3	.18 (.14) p = .19
Pro-social orientation 1 → Group consciousness 2 → Self-reported action 3	.45 (.25) p = .07

Figure 1 shows that PPS was the strongest indicator of the individual differences factor across the three measurement points, while social identification was the strongest indicator of the group consciousness factor. Figure 2 shows that consistent with Duncan’s (2012) theorizing, there was good evidence of contemporaneous mediation of the effect of the pro-social orientation on action within each time point. Consistent with Figures 1 and 2, Table 4 shows that the contemporaneous indirect effects were all significant except for the effect of the pro-social orientation on the donation allocation through group consciousness at Time 2, and the effect of the pro-social orientation on self-reported action through group consciousness at Time 3 was marginal.

There was also evidence consistent with longitudinal mediation of effects but here, unexpectedly, the relationship between the pro-social orientation and subsequent group consciousness was negative across both sets of time points, as was the relationship between group consciousness and self-reported action in the following wave (Figure 2; also Table 4). Given that the zero-order correlations between these variables are positive (see Table 1), this pattern of results is consistent with a suppressor effect. This unanticipated pattern of results is discussed further below.

Discussion

The current research considers the role of individual- and group-level factors in explaining humanitarian collective action and generosity. Drawing on Duncan’s (2012) integrative model of collective action, we hypothesized that a pro-social orientation would be an antecedent of the development of a humanitarian group consciousness that in turn mediates the effects of individual differences on engagement in humanitarian action. We found not only good support for the integrative model but also some unexpected twists.

Figure 2 shows that consistent with Duncan’s (2012) theorizing, there was good evidence of contemporaneous mediation of the effect of the pro-social orientation on action within each time point: The pro-social orientation promoted group consciousness that, in turn, predicted all three action variables, although donation allocation was not significantly predicted by group consciousness at Time 2. Unsurprisingly, there were strong autoregressive effects such that the actions were all significantly predicted by past actions; similarly, previous group consciousness strongly predicted subsequent group consciousness, and the pro-social orientation was strongly predicted by the pro-social orientation in previous waves. This pattern of results suggests that within each time point, the pattern of causal associations identified by the integrative model best explains current or contemporaneous action at an aggregate (between persons) level (Mackinnon, 2008).

However, the tests of longitudinal mediation revealed some unexpected findings that raise new questions about the nature of intra-individual changes in personal and group-based consciousness. Specifically, the longitudinal relationship between the pro-social orientation (Times 1 and 2) and subsequent group consciousness (Times 2 and 3) was negative. Similarly, group consciousness (Time 1) and subsequent (Time 2) self-reported action, and marginally (p = .06) the relationship between group consciousness (Time 2) and subsequent (Time 3) self-reported action was also negative. These findings suggest that the more a person cares in the abstract about an equal world at initial measurement, the less they report a humanitarian group consciousness a year later, with flow on effects undermining self-reported action 2 years hence. We note that Table 1 shows that the zero-order correlations between these variables are positive; thus, it is only when we account for autoregressive effects and variability associated with past and current action that the “true” negative relationships are revealed (a suppressor effect). That these effects are present across two time points increases confidence that these are reliable, if unexpected, findings (Figure 2). In the following sections, we consider the implications of these findings for theorizing of individual differences and group processes in humanitarian action, as well as the practical implications of these findings.

An Empirical Integration of Individual and Group in Action

The current research presents a unifying framework for considering both the distal antecedents (in terms of broad individual differences) and the more proximal causes (in terms of psychological group formation) of collective action (following Duncan, 1999, 2012; Duncan & Stewart, 2007). In doing so, the current integration fills gaps in both individual and group theorizing of action. The literature on individual differences in collective action is largely empirical with “no coherent unifying theory” (Duncan, 2012, p. 789). Embedding the individual differences in the theoretical framework provided by social identity research on collective action (e.g., van Zomeren et al., 2008) implies that there is a set of individual variables that (taken together as indicators of an underlying latent factor) are likely to increase group consciousness and therefore engagement. At three time points (across persons) we found that the humanitarian group consciousness (“who we are”) gave expression to more distal individual differences in egalitarian worldviews (“who I am”). This highlights the collective bases of humanitarian action: To be motivated to act, people need to believe, feel, and identify collectively (Thomas, McGarty, & Mavor, 2009). This point bears close attention because the group or collective dimension is missing from much of the analysis on responses to humanitarian issues and global poverty in particular (e.g., Small & Loewenstein, 2003) as well as charitable responses more generally (Bekkers & Wiepking, 2011).

However, the group processes literature, especially when following the social identity approach typically pays less attention to the role of individual differences as providing a platform or key antecedent to group formation and therefore collective action (with some exceptions, for example, Curtin et al., 2010; Duncan & Stewart, 2007; Reese et al., 2014). Although there have been calls for an increasingly social interactionist perspective on social action, one that recognizes how individual differences shape—and are shaped by—group processes (Duncan, 2012), the two traditions often follow distinct conceptual and empirical trajectories. This disconnection is important because, as Turner-Zwinkles et al. (2015) noted, personal change and social change are inherently intertwined, and understanding the relation between them is central to understanding this symbiosis (see Postmes, Haslam, & Swaab, 2005, for a discussion of how personal values can promote social identity formation).

However, contrary to other research that suggests that personal and political interests merge over time (Blackwood & Louis, 2012; Turner-Zwinkles et al., 2015), we found an unexpected longitudinal pattern whereby greater levels of individual and group engagement, respectively, subsequently undermined group formation and action, respectively. Similarly, contrary to theorizing, which suggests that engagement in collective action should empower (Drury & Reicher, 2000) and positively reinforce group consciousness (Duncan, 2012; also van Zomeren et al., 2012), we found that self-reported action weakly negatively predicted subsequent group consciousness.

There are two plausible explanations for these unexpected, negative, effects. First, this pattern of results could be seen as evidence consistent with a principle-implementations gap (Schuman, Steeh, Bobo, & Krysan, 1997) whereby the suppression effects reveal a false commitment that is not supported by specific actions. People who are strongly committed and people who pretend to be strongly committed both score highly on pro-social orientation. This leads to high scores on the pro-social orientation at subsequent waves but fails to materialize as group consciousness in subsequent waves once the shared variance associated with genuine pro-social orientation is controlled for. A similar pattern occurs with group consciousness and action, whereby a genuine commitment to the cause predicts contemporaneous action and is stable over time, but when we control for stable commitment, it reveals an earlier (overstated) commitment to the cause that is associated with lower levels of action. In this way, there may be an inconsistency between a person’s orientation toward achieving social justice, generally, and for people in developing nations, specifically (their “principles”), and their engagement in specific actions (the “implementation”; see also, Durrheim & Dixon, 2004).

A second plausible explanation is that at the intra-individual level, caring about others and an equal world may lead a person to care about many social issues and events (and see these as of political relevance); this may effectively dilute subsequent commitment to, and engagement, in a specific cause. This may also explain the weaker associations between group consciousness and the donation allocation because people who have a group consciousness around global poverty reduction are also likely to have a similar orientation to redressing domestic poverty (Duncan, 2012). This second explanation suggests a paradox: A higher pro-social orientation generally negates the likelihood of engagement in one specific cause, perhaps because these people are more susceptible to compassion fatigue and burn-out (see Moeller, 1999). Alternatively, it may be that these findings represent a failure to integrate the humanitarian group into the broader self-concept (Louis, Amiot, Thomas, & Blackwood, 2016; Turner-Zwinkles et al., 2015) and/or that the group-level costs of acting did not effectively translate to personal benefits (Blackwood & Louis, 2012).

Policy and Practical Implications

The year 2015 signals the beginning of the sustainable development goals (SDGs), an inter-governmental agreement of specific targets for global prosperity over the next 15 years (United Nations, 2015). Understanding the individual and group antecedents of support for the global humanitarian agenda is thus important from the perspective of generating not only donor charity but also international political will (McFarland, 2015).

Results highlight the need to consider collective dimensions of support for the humanitarian agenda but also offer a tale of caution. Although a recent practitioner report (Darnton & Kirk, 2011) considered the important role of universalism values in advocating for the humanitarian agenda, the longitudinal findings here suggest that targeting universalism values may be counterproductive over the longer term. Rather, the current findings attest to the crucial importance of allowing idiosyncratic individual differences and life experiences to find expression in group membership. How might this be achieved? Some research points to the powerful role of small group interactions (Thomas & McGarty, 2009; Thomas, McGarty, & Mavor, 2016). It is through (structured) small group discussion that people generate new collective identities that fuse individual orientations with collective motivation (Postmes et al., 2005; also Zwinkles-Turner et al., 2015). Practitioners would do well to consider the mechanisms through which their outreach and engagement can foster the social dimension of support.

Limitations and Future Directions

The current research used longitudinal SEM to understand the between-person (contemporaneous) and within-person (longitudinal) processes behind support for humanitarian action. Adopting a latent measurement approach to individual differences and group variables allows us to synthesize relevant variables, account for measurement error, and test hypotheses that it would otherwise be difficult to test with measured variables (due to the complexity of the model and the power required). However, as discussed above, the two aspects of this modeling are contradictory. Curran and Bauer (2011) discussed many examples of when the aggregate (between-person) results are contrary to within-person change. To untangle these competing findings, future research must fully disaggregate the influence of between-person and within-person effects on outcomes using multi-level growth models (Curran & Bauer, 2011).

The current research included those individual differences variables most widely implicated in reviews of philanthropy, human rights behavior, and collective action; nevertheless, this selection was not intended to be exhaustive and there are likely to be other relevant individual differences variables that we have not considered here. Similarly, we did not consider the role of human-level or global identification in group consciousness; these may involve specific dimensions that provide similarly applicable behavioral norms (e.g., global self-investment; Reese, Proch, & Finn, 2015). Future research might consider these possibilities. Nevertheless, we suggest that the motivational model provides a framework that can be readily applied to many social issues, where researchers can test which individual and group factors are most relevant to that context.

In this vein, one notable omission from the current investigation is the catalyzing role of life experiences. Duncan (2012) not only placed life experiences alongside individual differences as distal antecedents to an emerging group consciousness but also proposed that specific life experiences may moderate the effects of individual differences and group consciousness on action. Future research might incorporate life experiences (e.g., relating to education, political efficacy, well-being; see Duncan, 2012) to consider the ways in which specific events or life experiences interact with individual differences and/or group consciousness to promote action.

Concluding Comments

Although humanitarian issues are often viewed by members of the public as issues of charity (e.g., Darnton & Kirk, 2011), the retributive reaction of many governments to people fleeing persecution and the recent negotiation of the SDGs reminds us that humanitarian issues are also political ones (see Reese, Rosenmann, & McGarty, 2015). The current research sought to develop a more complex understanding of the kinds of people who develop the will to act collectively to seek humanitarian justice for people in developing countries. These are not just questions of concern to psychologists (McFarland, 2015). Harmonizing and integrating these literatures is of critical importance to understand the ways that the engagement of citizens can foster the political will necessary to advance human rights, especially beyond the borders of one’s own nation.

Footnotes

Author’s Note

Ana-Maria Bliuc is currently affiliated with Western Sydney University.

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: This research was supported by the Australian Research Council’s Discovery Early Career Researcher Award (DE120101029).

Supplemental Material

The online supplemental material is available at .

References

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: W.H. Freeman.

Baray

Postmes

Jetten

(2009). When I equals we: Exploring the relation between social and personal identity of extreme right-wing political party members. British Journal of Social Psychology, 48, 625-647.

Bekkers

Wiepking

(2011). Who gives? A literature review of predictors of charitable giving part one: Religion, education, age and socialisation. Voluntary Sector Review, 2, 337-365.

Blackwood

L. M.

Louis

W. R.

(2012). If it matters for the group then it matters to me: Collective action outcomes for seasoned activists. British Journal of Social Psychology, 51, 72-92.

Bliuc

A.-M.

McGarty

Hartley

L. K.

Hendres

D. M.

(2012). Manipulating national identity: The strategic use of rhetoric by supporters and opponents of the “Cronulla riots” in Australia. Ethnic and Racial Studies, 35, 2174-2194.

Bliuc

A.-M.

McGarty

Reynolds

Hendres

D. M.

(2007). Opinion-based group membership as a predictor of commitment to political action. European Journal of Social Psychology, 37, 19-32.

Bliuc

A.-M.

McGarty

Thomas

Lala

Berndsen

Misajon

(2015). Public division about climate change rooted in conflicting socio-political identities. Nature Climate Change, 5, 226-229.

Bolitho

F. H.

Carr

S. C.

Fletcher

R. B.

(2007). Public thinking about poverty: Why it matters and how to measure it. International Journal of Nonprofit and Voluntary Sector Marketing, 12, 13-22.

Cameron

J. E.

(2004). A three-factor model of social identity. Self and Identity, 3, 239-262.

10.

Cohrs

J. C.

Maes

Moschner

Kielmann

(2007). Determinants of human rights attitudes and behavior: A comparison and integration of psychological perspectives. Political Psychology, 28, 441-469.

11.

Collett

J. L.

Morrissey

C. A.

(2007). The social psychology of generosity: The state of current interdisciplinary research ( Electronic Report for the John Templeton Foundation Generosity Planning Project.–45 Vol. 15, pp. 2008-09).

12.

Curran

P. J.

Bauer

D. J.

(2011). The disaggregation of within-person and between-person effects in longitudinal models of change. Annual Review of Psychology, 62, 583-619.

13.

Curtin

Stewart

A. J.

Duncan

L. E.

(2010). What makes the political personal? Openness, personal political salience, and activism. Journal of Personality, 78, 943-968.

14.

Darnton

Kirk

(2011). Finding frames: New ways to engage the UK public in global poverty. London, England: Bond.

15.

Diaz-Veizades

Widaman

K. F.

Little

T. D.

Gibbs

K. W.

(1995). The measurement and structure of human rights attitudes. The Journal of Social Psychology, 135, 313-328.

16.

Drury

Reicher

(2000). Collective action and psychological change: The emergence of new social identities. British Journal of Social Psychology, 39, 579-604.

17.

Duncan

L. E.

(1999). Motivation for collective action: Group consciousness as mediator of personality, life experiences, and women’s rights activism. Political Psychology, 20, 611-635.

18.

Duncan

L. E.

(2012). The psychology of collective action. In Deaux

Snyder

(Eds.), The Oxford handbook of personality and social psychology (pp. 781-803). New York, NY: Oxford University Press.

19.

Duncan

L. E.

Stewart

A. J.

(2007). Personal political salience: The role of personality in collective identity and action. Political Psychology, 28, 143-164.

20.

Durrheim

Dixon

(2004). Attitudes in the fiber of everyday life: The discourse of racial evaluation and the lived experience of desegregation. American Psychologist, 59, 626-636.

21.

Enders

C. K.

(2013). Dealing with missing data in developmental research. Child Development Perspectives, 7, 27-31.

22.

Feather

N. T.

Woodyatt

McKee

I. R.

(2012). Predicting support for social action: How values, justice-related variables, discrete emotions, and outcome expectations influence support for the Stolen Generations. Motivation and Emotion, 36, 516-528.

23.

Hall

R. J.

Snell

A. F.

Foust

M. S.

(1999). Item parceling strategies in SEM: Investigating the subtle effects of unmodeled secondary constructs. Organizational Research Methods, 2, 233-256.

24.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

25.

Lavine

Sullivan

J. L.

Borgida

Thomsen

C. J.

(1996). The relationship of national and personal issue salience to attitude accessibility on foreign and domestic policy issues. Political Psychology, 293-316.

26.

Louis

W. R.

Amiot

C. E.

Thomas

E. F.

Blackwood

(2016). The “activist identity” and activism across domains: A multiple identities analysis. Journal of Social Issues, 72, 242-263.

27.

Mackinnon

D. P.

(2008). Introduction to statistical mediation analysis. New York, NY: Lawrence Erlbaum.

28.

McFarland

(2015). Culture, individual differences, and support for human rights: A general review. Peace and Conflict: Journal of Peace Psychology, 21, 10.

29.

McFarland

Mathews

(2005). Who cares about human rights? Political Psychology, 26, 365-385.

30.

McFarland

Webb

Brown

(2012). All humanity is my ingroup: A measure and studies of identification with all humanity. Journal of Personality and Social Psychology, 103, 830-853.

31.

McGarty

Bliuc

A. M.

Thomas

E. F.

Bongiorno

(2009). Collective action as the material expression of opinion-based group membership. Journal of Social Issues, 65, 839-857.

32.

Milanovic

(2005). Worlds apart: Measuring international and global inequality. Princeton, NJ: Princeton University Press.

33.

Muthén

L. K.

Muthén

B. O.

(1998-2012). Mplus: The comprehensive modeling program for applied researchers [Computer program]. Los Angeles, CA: Author.

34.

Olson

J. T.

(1997). Perceptions of global inequality: A call for research. Social Justice Research, 10, 39-62.

35.

Penner

L. A.

(2002). Dispositional and organizational influences on sustained volunteerism: An interactionist perspective. Journal of Social Issues, 58, 447-467.

36.

Pittinsky

T. L.

Diamante

(2015). Global bystander non-intervention. Peace and Conflict: Journal of Peace Psychology, 21, 226-247.

37.

Postmes

Haslam

S. A.

Swaab

R. I.

(2005). Social influence in small groups: An interactive model of social identity formation. European Review of Social Psychology, 16, 1-42.

38.

Pratto

Sidanius

Stallworth

L. M.

Malle

B. F.

(1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741.

39.

Reese

Berthold

Steffens

M. C.

(2012). We are the world—And they are not: Prototypicality for the world community, legitimacy, and responses to global inequality. Political Psychology, 33, 683-700.

40.

Reese

Kohlmann

(2015). Feeling global, acting ethical: Global identification and fairtrade consumption. Journal of Social Psychology, 155, 98-106.

41.

Reese

Proch

Cohrs

J. C.

(2014). Individual differences in responses to global inequality. Analysis of Social Issues and Public Policy, 14, 217-238.

42.

Reese

Proch

Finn

(2015). Identification with all humanity: The role of self-investment and self-definition. European Journal of Social Psychology, 45, 426-440.

43.

Reese

Rosenmann

McGarty

(2015). Globalisation and global concern: Developing a social psychology of human responses to global challenges. European Journal of Social Psychology, 45, 799-805.

44.

Saab

Tausch

Spears

Cheung

W. Y.

(2015). Acting in solidarity: Testing an extended dual pathway model of collective action by bystander group members. British Journal of Social Psychology, 54, 539-560.

45.

Schuman

Steeh

Bobo

Krysan

(1997). Racial attitudes in America: Trends and interpretations. Cambridge, MA: Harvard University Press.

46.

Schwartz

S. H.

(1992). Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. Advances in Experimental Social Psychology, 25, 1-65.

47.

Simon

Klandermans

(2001). Politicized collective identity: A social psychological analysis. American Psychologist, 56, 319-331.

48.

Small

D. A.

Loewenstein

(2003). Helping a victim or helping the victim: Altruism and identifiability. Journal of Risk and Uncertainty, 26, 5-16.

49.

Smith

H. J.

Ortiz

D. J.

(2002). Is it just me? The different consequences of personal and group relative deprivation. In Walker

Smith

H. J.

(Eds.), Relative deprivation: Specification, development, and integration (pp. 91-115). New York, NY: Cambridge University Press.

50.

Smith

L. G. E.

Thomas

E. F.

McGarty

(2015). “We must be the change we want to see in the world:” Integrating norms and identities through social interaction. Political Psychology, 36, 543-557.

51.

Soper

(2016). Free statistics calculators version 4.0. Retrieved from http://www.danielsoper.com/statcalc/default.aspx

52.

Thomas

E. F.

Mavor

K. I.

McGarty

(2012). Social identities facilitate and encapsulate action-relevant constructs: A test of the social identity model of collective action. Group Processes & Intergroup Relations, 15, 75-88.

53.

Thomas

E. F.

McGarty

(2009). The role of efficacy and moral outrage norms in creating the potential for international development activism through group-based interaction. British Journal of Social Psychology, 48, 115-134.

54.

Thomas

E. F.

McGarty

Lala

Stuart

Hall

L. J.

Goddard

(2015). Whatever happened to Kony2012? Understanding a global Internet phenomenon as an emergent social identity. European Journal of Social Psychology, 45, 356-367.

55.

Thomas

E. F.

McGarty

Mavor

K. I.

(2009). Aligning identities, emotions, and beliefs to create commitment to sustainable social and political action. Personality and Social Psychology Review, 13, 194-218.

56.

Thomas

E. F.

McGarty

Mavor

K. I.

(2010). Social psychology of making poverty history: Motivating anti-poverty action in Australia. Australian Psychologist, 45, 4-15.

57.

Thomas

E. F.

McGarty

Mavor

K. I.

(2016). Group interaction as the crucible of social identity formation: A glimpse at the foundations of social identities for collective action. Group Processes & Intergroup Relations, 19, 137-151.

58.

Turner

J. C.

Hogg

M. A.

Oakes

P. J.

Reicher

S. D.

Wetherell

M. S.

(1987). Rediscovering the social group: A self-categorization theory. Cambridge, MA: Basil Blackwell.

59.

Turner

J. C.

Oakes

P. J.

Haslam

S. A.

McGarty

(1994). Self and collective: Cognition and social context. Personality and Social Psychology Bulletin, 20, 454-463.

60.

Turner-Zwinkels

van Zomeren

Postmes

(2015). Politicization during the 2012 US presidential elections bridging the personal and the political through an identity content approach. Personality and Social Psychology Bulletin, 433-445.

61.

United Nations. (2015). Sustainable development goals. Retrieved http://www.un.org/sustainabledevelopment/sustainable-development-goals/

62.

Vandenberg

R. J.

Lance

C. E.

(2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3, 4-70.

63.

van Zomeren

Leach

C. W.

Spears

(2012). Protesters as “passionate economists” a dynamic dual pathway model of approach coping with collective disadvantage. Personality and Social Psychology Review, 16, 180-199.

64.

van Zomeren

Postmes

Spears

(2008). Toward an integrative social identity model of collective action: A quantitative research synthesis of three socio-psychological perspectives. Psychological Bulletin, 134, 504-535.

65.

van Zomeren

Postmes

Spears

Bettache

(2011). Can moral convictions motivate the advantaged to challenge social inequality? Extending the social identity model of collective action. Group Processes & Intergroup Relations, 14, 735-753.

66.

van Zomeren

Spears

Fischer

A. H.

Leach

C. W.

(2004). Put your money where your mouth is! Explaining collective action tendencies through group-based anger and group efficacy. Journal of Personality and Social Psychology, 87, 649-664.

67.

Wiepking

Bekkers

(2012). Who gives? A literature review of predictors of charitable giving. Part Two: Gender, family composition and income. Voluntary Sector Review, 3, 217-245.

68.

Wright

S. C.

Taylor

D. M.

Moghaddam

F. M.

(1990). Responding to membership in a disadvantaged group: From acceptance to collective protest. Journal of Personality and Social Psychology, 58, 994-1003.

69.

Zagefka

James

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