Dynamic,Yet Stable: Separating Within- and Between-Person Components of Collective Action in Support of a Disadvantaged Outgroup and its Antecedents

Abstract

Despite an increasing interest in the drivers of intergroup solidarity, the within-person longitudinal relationships between advantaged group members’ engagement for disadvantaged groups and its postulated antecedents remain scarcely tested. In the context of the refugee crisis following Russia’s invasion of Ukraine, we conducted a three-wave longitudinal survey (N_T1 = 804, N_T2 = 702, and N_T3 = 624) assessing Poles’ (the advantaged group) willingness to act for Ukrainians (the disadvantaged group), together with three hypothesized predictors—moral convictions, intergroup contact, and politicized identity. Employing a random intercept cross-lagged panel model that separates between- from within-person variance, we found that within-person changes in moral convictions and friendship contact directly predicted subsequent action intentions. Contrary to past theorizing, politicized identity emerged as consequence rather than an antecedent of collective action. Superficial intergroup contact indirectly predicted engagement intentions by facilitating cross-group friendship. We discuss the implications of our findings for current models of collective action.

Keywords

collective action for a disadvantaged group politicized identity moral convictions intergroup contact longitudinal methodology

Recent years witnessed a substantial increase in collective action research (Agostini & van Zomeren, 2021). Next to the collective action of disadvantaged groups to improve their status, more attention has been paid to advantaged group members’ engagement for disadvantaged outgroups. Apart from high-quality studies (e.g., Hässler et al., 2020), the literature has been enriched by significant theoretical developments (Radke et al., 2020). Still, some issues remain rather underresearched. One of these are overtime associations between collective action for disadvantaged outgroups and its psychological drivers (for exceptions, see Reimer et al., 2017; Sengupta et al., 2022). Out of multiple antecedents of pro-outgroup engagement proposed in the literature, many (e.g., cross-group friendship) do not succumb to experimental manipulation (e.g., MacInnis & Page-Gould, 2015). Moreover, there are reasons to believe that the relationships between collective action for disadvantaged outgroups and its hypothesized determinants are, in fact, reciprocal and that the antecedents themselves are connected by a network of mutual relationships. In these circumstances, longitudinal data in connection with the appropriate statistical apparatus (see Rohrer & Murayama, 2021) seem to be the most optimal way to investigate the antecedents and consequences of pro-outgroup engagement.

In this contribution, we add to the existing literature by examining the longitudinal associations between collective action for a disadvantaged outgroup and its four postulated antecedents—politicized identity, moral convictions, personal intergroup contact, and cross-group friendship (see Radke et al., 2020). We do so in the context of the refugee crisis that arose following Russia’s invasion of Ukraine in February 2022. Using three-wave longitudinal data collected in a nationwide sample of Poles (the advantaged group), we estimate the within-person longitudinal associations between collective action for Ukrainians (the disadvantaged group) and its hypothesized determinants. To this aim, we employ the random intercept cross-lagged panel model (RI-CLPM; Hamaker et al., 2015)—an extension of the traditional cross-lagged panel model (CLPM; e.g., Wu et al., 2013) that distinguishes between-person and within-person variance.

Collective Action for Disadvantaged Outgroups

Many contributions to the collective action literature point to politicized identity (i.e., identification with a social movement; Simon & Klandermans, 2001) as the key antecedent of advantaged group members’ engagement for a disadvantaged outgroup (e.g., Thomas et al., 2012; van Zomeren et al., 2008). For instance, the extended social identity model of collective action (extended SIMCA; van Zomeren et al., 2018) posits that when individuals perceive a normative fit between their moral beliefs and the principles of a given social movement, moral convictions facilitate politicized identity development and, in consequence, motivate protest behavior (van Zomeren, Postmes, & Spears, 2012).

At the same time, protest behavior was shown to increase identification with social movements (e.g., Becker et al., 2011). Therefore, one may expect that politicized identity simultaneously motivates and originates from pro-outgroup engagement. However, longitudinal research testing this hypothesis is missing (but see Thomas et al., 2021). Our study aims to fill this gap by examining politicized identity both as an antecedent and as a consequence of collective action for a disadvantaged outgroup.

Moral beliefs (i.e., perceptions of what is right and what is wrong; Skitka, 2010) constitute another hypothesized determinant of engagement for disadvantaged outgroups. As moral principles have an absolute character (i.e., are believed to apply to everyone, regardless of group membership; Skitka, 2010), any behaviors or situations that offend these standards elicit anger and facilitate action. In the context of asymmetrical power relations, advantaged group members who endorse egalitarian ideals may be motivated to engage for a disadvantaged outgroup to uphold their moral standards (Radke et al., 2020). The explanatory power of moral beliefs was demonstrated in multiple cross-sectional studies, including those embedded in hierarchical group contexts (e.g., van Zomeren, Postmes, & Spears, 2012). For instance, moral convictions (i.e., attitudes on a particular issue that are rooted in one’s core values; Skitka, 2010) predicted Hungarians’ engagement in solidarity with refugees during the 2015 crisis (Kende et al., 2017). However, we are not aware of any study that has examined the longitudinal association between moral convictions and collective action for a disadvantaged outgroup.

Finally, advantaged group members’ engagement with a disadvantaged outgroup may originate from their positive relations with outgroup members. Intergroup contact was demonstrated to improve attitudes toward outgroups (Pettigrew & Tropp, 2006) as well as promote collective action on their behalf (e.g., Hässler et al., 2020; Reimer et al., 2017; but see Sengupta et al., 2022). The form of intergroup contact that may be especially effective in motivating engagement is cross-group friendship (MacInnis & Hodson, 2019). In the course of repeated interactions with their outgroup friends, advantaged group members are likely to empathize with the disadvantaged outgroup (e.g., Cakal et al., 2021) and recognize the inequality that exists between groups (e.g., Lytle et al., 2017). As these factors are associated with collective action for the disadvantaged outgroup (e.g., Mallett et al., 2008; Saab et al., 2015; Selvanathan et al., 2018), we propose that cross-group friendship would facilitate this type of engagement. Analogous effects are likely to be weaker, or even nonexistent, for superficial (i.e., lacking friendship-specific features such as repeated interactions or mutual self-disclosure; see Davies et al., 2011) forms of intergroup contact. On the contrary, intergroup contact is theorized to improve outgroup-directed attitudes primarily in the early stages of intergroup relationships (MacInnis & Page-Gould, 2015; Page-Gould, 2022). Therefore, when the coexistence of the advantaged and disadvantaged groups is relatively recent, superficial contact may be sufficient to motivate collective action.

Importantly, collective action for a disadvantaged outgroup may feed back into its psychological antecedents (Becker & Tausch, 2015; van Zomeren, Leach, & Spears, 2012). For instance, equality-based moral beliefs may not only prompt advantaged group members to fight for social change but also develop as a consequence of such engagement. In a similar vein, participating in a demonstration aimed at improving the situation of the disadvantaged outgroup may be a good opportunity to befriend members of this group.

Furthermore, determinants of collective action for the disadvantaged outgroup are likely to be interrelated. For example, moral convictions may define politicized identity and, by doing so, stimulate engagement for social change (van Zomeren et al., 2018). A similar effect may be expected for intergroup contact—in consequence of positive, close relations with outgroup members, advantaged group members may recognize the injustice of intergroup status differences and develop identification with social movements that seek to flatten intergroup hierarchy (MacInnis & Hodson, 2019). To disentangle these potentially reciprocal relationships, an appropriate statistical framework is needed.

The Dual Nature of Collective Action

Collective action is not a static phenomenon. Instead of maintaining a constant level over time, people’s readiness to engage in protest behavior is subject to temporal fluctuations (e.g., Tausch & Becker, 2013). Several theoretical models have been proposed to capture the dynamic aspect of engagement (e.g., Becker & Tausch, 2015; Drury & Reicher, 2009; Louis et al., 2020; van Zomeren, Leach, & Spears, 2012). In these models, changes in collective action are hypothesized to originate from (and translate into) changes in a range of psychological constructs.

On the contrary, one may think of collective action as a relatively stable, trait-like characteristic. It has been acknowledged that protest participation operates in “packages”—rather than fighting for a single cause, people tend to engage in collective action related to multiple (albeit connected) issues (e.g., Andersen & Jennings, 2010; Louis et al., 2016). Moreover, a considerable body of research points to personality as a distal antecedent of collective action (e.g., Besta et al., 2021; Mondak et al., 2010).

Although the dual (i.e., dynamic and stable) nature of collective action has been recognized in the literature (e.g., Louis et al., 2020), relatively few studies so far have sought to separate the two aspects empirically. While Thomas et al. (2021) focused on determining individual differences in collective action trajectories, only Sengupta et al. (2022) simultaneously modeled between- and within-person relationships between intergroup contact and pro-outgroup engagement. The scarcity of research isolating between- and within-person components of the investigated variables is a serious neglect as the causal relations postulated in psychological models of collective action (e.g., Becker & Tausch, 2015; van Zomeren, Leach, & Spears, 2012) can be conceived of as operating at the within-person level of analysis.

Longitudinal data in different areas of psychology, including the collective action literature (e.g., Thomas et al., 2020), have by default been analyzed with the CLPM. In this framework, the constructs of interest assessed at time T are regressed on themselves (autoregressive effects) and other variables (cross-lagged effects) measured at T−1. Therefore, a cross-lagged effect in CLPM tests for the prospective effect of a given construct on change in individual differences in another (Orth et al., 2021).

Recently, however, the limitations of the CLPM have been highlighted. The main line of criticism is that the CLPM confounds between-person and within-person variance (e.g., Hamaker et al., 2015; Usami et al., 2019). It is therefore not an appropriate statistical tool to simultaneously account for stability and dynamism in collective action or to capture the within-person longitudinal associations between engagement and other variables. In addition, as shown by simulation analyses, the CLPM is likely to find spurious cross-lagged effects when there are none and underestimate cross-lagged effects when they exist (Lucas, 2022).

These limitations do not apply to the RI-CLPM. In this model, stable between-person variance in each construct is captured by a separate trait factor (i.e., random intercept factor). The within-person variance, on the other hand, serves as the basis for estimating longitudinal associations between the variables of interest. In contrast to the CLPM, the autoregressive and cross-lagged effects in the RI-CLPM are modeled on the residualized scores and not on observed scores. A positive autoregressive effect of Construct A thus reveals whether a within-person deviation from the trait level of Construct A predicts analogical deviation in the subsequent occasion. Thus, it indicates whether a positive or negative deviation from the trait level of a given construct carries over from one measurement to the next (Hamaker et al., 2015; Usami et al., 2019). Meanwhile, a significant cross-lagged effect of Construct A on Construct B means that a within-person deviation from the trait level of Construct A predicts a within-person deviation from the person’s mean for construct B in the following measurement. Therefore, RI-CLPM is an optimal choice when reciprocal relations between constructs are investigated (Usami et al., 2019; but see Lüdtke & Robitzsch, 2021).

The Current Research

The context for our research was the immediate aftermath of Russia’s invasion of Ukraine on February 24, 2022, which led to around 3.5 million Ukrainians seeking refuge from the war in Poland (Derewienko, 2022). The refugees received a warm welcome from Polish society, with thousands of private individuals helping them by raising funds and offering free transport and accommodation (Petriczko, 2022). Material help was also dispatched to Ukrainian cities (Brzozowski, 2022).

In this study, we sought to identify the psychological sources and consequences of Poles’ collective action for the people of Ukraine (both the refugees and those who remained in their country). To this end, we conducted an online longitudinal study with three measurements separated by 2-week intervals. Data collection lasted between March 14 and April 14, 2022. Data, code, and the full questionnaire are available on the project’s OSF website (https://osf.io/e54bk/?view_only=c23e5b247e694148b260ac65d958279b).¹

Method

Sample

Our study involved three (N_T1 = 804, N_T2 = 702, N_T3 = 624) 20-min measurements. Participants were recruited from an online panel held by an external company (Pollster) and were compensated with credit points that could be exchanged for material rewards. The intended sample size was no less than 500 individuals who participated in all three measurements. This was determined based on funding availability and, given the 80% power, sufficed to detect correlations of |r| > .12.² After excluding individuals who did not declare Polish nationality, data from 620 individuals (314 men and 306 women aged between 18 and 84, M = 48.87, SD = 16.11) who completed the questionnaire for all measurements were analyzed. The final sample was representative of the Polish adult population in terms of gender and age distribution.

Measures

The measures presented below were embedded in a longer questionnaire that assessed various social-psychological constructs. Unless otherwise noted, all measures used a 1 (strongly disagree) to 5 (strongly agree) response scale.

Collective Action Intentions3

To assess normative collective action intentions, we asked participants how likely they were to engage in seven behaviors (i.e., participating in a demonstration, donating money to an aid organization, providing material assistance, organizing transport for refugees, providing accommodation to refugees, hanging the Ukrainian flag in the window/on the balcony, adding the Ukrainian flag to their profile picture on social media) in the next 2 weeks to support Ukraine and Ukrainians. The response scale ranged from 1 (very unlikely) to 7 (very likely).

Personal Intergroup Contact⁴

The measure of personal intergroup contact consisted of 2 items: “Do you personally know any Ukrainians?” (1 = no, I don’t, 2 = yes, one or two, 3 = yes, a few, 4 = yes, a lot, 5 = yes, many) and “How often do you talk to Ukrainians?” (1 = never, 2 = once a year, 3 = once a month, 4 = once a week, 5 = everyday).

Cross-Group Friendship

To assess cross-group friendship, we employed two items: “Do you have any Ukrainians among your friends?” (1 = no, I don’t, 2 = yes, one or two, 3 = yes, a few, 4 = yes, a lot, 5 = yes, many) and “How often do you talk to Ukrainians who are your friends?” (1 = never, 2 = once a year, 3 = once a month, 4 = once a week, 5 = everyday).

Politicized Identity

To gauge politicized identity, two items were used: “I identify with people who help Ukrainians” and “I identify with the movement to support Ukrainians.”

Moral Convictions

Moral convictions were measured with two items: “My opinion about helping Ukrainians reflects an important part of who I am” and “My opinion about helping Ukrainians is an important part of my moral norms and values” (van Zomeren, Postmes, & Spears, 2012).

Results

Analytic Strategy

Our analyses involved estimating two (i.e., stationary and nonstationary, see below) RI-CLPMs (see Figure 1). Each of these models included five observed variables—collective action intentions, politicized identity, moral convictions, personal intergroup contact, and cross-group friendship—assessed across three measurement occasions. To account for between-person stability of the analyzed constructs, a latent random intercept (i.e., participants’ average level of the given construct over three measurements) for each construct was created. Factor loadings linking random intercepts with respective observed variables were fixed to 1. The five random intercepts were allowed to correlate. The within-person part of our model consisted of temporal deviations from participants’ expected scores (i.e., random intercepts). The 15 (5 variables × 3 measurements) deviation scores were reflected by latent variables linked to the observed variables by factor loadings fixed to 1. As the residual variances of the observed variables were fixed to 0, we decomposed variance in the observed variables into two parts—a between-person component (i.e., the random intercepts) and a within-person component (i.e., latent within-person deviation scores). The within-person autoregressive and cross-lagged effects were created by regressing each deviation score at T3 or T2 on all deviation scores obtained at T2 or T1, respectively.

Figure 1.

Conceptual Random Intercept Cross-Lagged Panel Model of the Associations Between Collective Action for a Disadvantaged Outgroup and its Antecedents

To test whether our data reflected a stationary process (i.e., where intraindividual change is stable across time), we compared the non-stationary (i.e., unconstrained) model with the stationary solution, which constrained the within-person autoregressive and cross-lagged effects registered between T1 and T2 to be equivalent to the corresponding effects observed between T2 and T3. A better fit of the nonstationary model would suggest that the within-person processes changed over time (Cole & Maxwell, 2003; Hamaker et al., 2015). The results of a better-fitting (or more parsimonious) model were then interpreted. In addition, we estimated 20 indirect effects of T1 variables on T3 variables (for a similar analysis, see Waddell et al., 2022).

All analyses were performed in Mplus 8.0 (Muthén & Muthén, 1998–2017) and employed the maximum likelihood (ML) estimator. The model fit was judged based on conventional criteria (Hu & Bentler, 1999). To account for the violation of multivariate normality assumption, the 95% confidence intervals for all estimates were obtained with bias-corrected bootstrapping (10,000 re-samples).⁵ There were no missing data.

Random Intercept Cross-Lagged Panel Model6

Stationarity Test

Table 1 presents descriptive statistics, reliabilities, and intercorrelations for the variables of interest. We began our analyses by testing for stationarity. As the non-stationary RI-CLPM did not fit data better than its more parsimonious, stationary counterpart (see Table 2), we interpreted the results of the latter.

Table 1.

Reliabilities, Descriptive Statistics (M, SD), and Intercorrelations for Key Variables.

Variable	Reliability	M (SD)	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. CA intentions T1	.84	3.43 (1.44)	—
2. Personal intergroup contact T1	.87	2.19 (1.10)	.23	—
3. Cross-group friendship T1	.84	1.50 (0.86)	.32	.60	—
4. Politicized identity T1	.90	3.98 (1.01)	.56	.08	.15	—
5. Moral convictions T1	.87	3.99 (0.89)	.49	.04	.12	.72	—
6. CA intentions T2	.85	3.21 (1.42)	.80	.19	.34	.51	.46	—
7. Personal intergroup contact T2	.86	2.15 (1.08)	.21	.82	.53	.09	.06	.18	—
8. Cross-group friendship T2	.85	1.48 (0.84)	.33	.54	.75	.15	.13	.35	.61	—
9. Politicized identity T2	.89	3.87 (1.02)	.57	.07	.18	.76	.62	.58	.09	.18	—
10. Moral convictions T2	.89	3.89 (0.96)	.49	.04	.16	.63	.66	.52	.08	.16	.73	—
11. CA intentions T3	.86	3.06 (1.41)	.79	.15	.28	.49	.42	.81	.14	.33	.53	.48	—
12. Personal intergroup contact T3	.87	2.13 (1.06)	.22	.77	.51	.11	.04	.20	.85	.57	.11	.08	.18	—
13. Cross-group friendship T3	.88	1.47 (0.82)	.33	.47	.70	.18	.14	.36	.53	.78	.19	.18	.33	.59	—
14. Politicized identity T3	.90	3.85 (1.02)	.56	.10	.18	.74	.59	.58	.09	.19	.79	.64	.57	.13	.20	—
15. Moral convictions T3	.89	3.83 (0.98)	.50	.06	.14	.63	.69	.51	.06	.17	.67	.69	.51	.09	.18	.77

Note. N = 620. Significant (p < .05) correlation coefficients are presented in bold. For two-item measures, Spearman–Brown coefficients are reported to assess reliability. CA = collective action.

Table 2.

Fit Statistics for the Nonstationary and Stationary RI-CLPM.

Model	χ²	df	p	AIC	BIC	RMSEA [90% CI]	CFI	SRMR	Δχ²	Δdf	Δp
Nonstationary RI-CLPM	20.19	10	.028	19324.93	19878.65	.041 [.013, .066]	.999	.016
Stationary RI-CLPM	33.63	35	.534	19288.37	19731.34	.000 [.000, .027]	1.000	.021	13.44	25	.971

Note. RI-CLPM = random intercept cross-lagged panel model; AIC = Akaike information criterion; BIC = Bayesian information criterion; RMSEA = root mean square error of approximation; CI = confidence interval; CFI = comparative fit index; SRMR = standardized root mean residual.

Between-Person Level

Among the between-person relationships (see Table 3; Figure 2), only the association between personal acquaintance and moral convictions random intercepts did not reach significance. The remaining associations between the trait factors were positive and significant.

Table 3.

Between-Person Associations for Collective Action Intentions, Personal Intergroup Contact, Cross-Group Friendship, Politicized Identity, and Moral Convictions (Stationary Model).

Variable 1	Variable 2	B	SE	p	95% CI
Collective action intentions	Personal intergroup contact	0.27	0.07	<.001	[0.15, 0.40]
Collective action intentions	Cross-group friendship	0.34	0.06	<.001	[0.24, 0.46]
Collective action intentions	Politicized identity	0.73	0.07	<.001	[0.60, 0.85]
Collective action intentions	Moral convictions	0.59	0.06	<.001	[0.48, 0.70]
Personal intergroup contact	Cross–group friendship	0.40	0.08	<.001	[0.25, 0.52]
Personal intergroup contact	Politicized identity	0.11	0.05	.022	[0.02, 0.20]
Personal intergroup contact	Moral convictions	0.03	0.04	.552	[–0.06, 0.11]
Cross–group friendship	Politicized identity	0.13	0.04	<.001	[0.06, 0.20]
Cross–group friendship	Moral convictions	0.08	0.03	.012	[0.02, 0.15]
Politicized identity	Moral convictions	0.56	0.05	<.001	[0.46, 0.65]

Note. 95% CIs were obtained with bootstrapping (10,000 re-samples). CI = confidence interval; SE = standard error.

Figure 2.

Between-Person Associations for Collective Action Intentions, Personal Intergroup Contact, Cross-Group Friendship, Politicized Identity, and Moral Convictions (Stationary Random Intercept Cross-Lagged Panel Model).

Within-Person Level

Table 4 and Figure 3 present the within-person components of our model. As evidenced by mostly nonsignificant autoregressive paths, only personal intergroup contact showed a carry-over effect (i.e., the between-measurement persistence of a deviation from the individual mean; see Osborne et al., 2021). For the cross-lagged effects, we found that positive deviations from the trait level of moral convictions were associated with deviations from the mean level of collective action in the following measurement. In other words, when participants’ moralization of engagement for the disadvantaged outgroup was greater than usual, they experienced a subsequent increase in collective action intentions. In a similar vein, positive deviations from participants’ average level of cross-group friendship predicted positive deviations from the trait level of collective action in the next wave. Thus, advantaged group members who reported greater cross-group friendship than usual were more willing than expected to take collective action for the disadvantaged outgroup in the next measurement. By contrast, deviations from the trait level of collective action did not depend on the preceding deviations from participants’ mean levels of politicized identity and personal intergroup contact. At the same time, positive deviations from participants’ mean level of collective action were associated with subsequent positive deviations in politicized identity. This suggests that engagement for the disadvantaged outgroup strengthens identification with pro-outgroup social movements.

Table 4.

Within-Person Autoregressive and Cross-Lagged Effects (Stationary Model).

Predictor	Outcome	B	SE	p	95% CI
Collective action intentions_t	Collective action intentions_t+1	0.09	0.12	.484	[−0.14, 0.33]
Personal intergroup contact_t	Collective action intentions_t+1	−0.16	0.10	.115	[−0.34, 0.04]
Cross-group friendship_t	Collective action intentions_t+1	0.32	0.10	.001	[0.12, 0.52]
Politicized identity_t	Collective action intentions_t+1	0.04	0.09	.652	[−0.13, 0.21]
Moral convictions_t	Collective action intentions_t+1	0.15	0.08	.048	[0.01, 0.31]
Collective action intentions_t	Personal intergroup contact_t+1	−0.01	0.06	.927	[−0.12, 0.11]
Personal intergroup contact_t	Personal intergroup contact_t+1	0.36	0.11	.001	[0.16, 0.58]
Cross-group friendship_t	Personal intergroup contact_t+1	0.13	0.09	.154	[−0.06, 0.29]
Politicized identity_t	Personal intergroup contact_t+1	−0.06	0.07	.355	[−0.19, 0.07]
Moral convictions_t	Personal intergroup contact_t+1	0.13	0.06	.020	[0.03, 0.25]
Collective action intentions_t	Cross-group friendship_t+1	0.07	0.05	.110	[−0.02, 0.17]
Personal intergroup contact_t	Cross-group friendship_t+1	0.16	0.08	.041	[0.01, 0.30]
Cross-group friendship_t	Cross-group friendship_t+1	0.18	0.12	.136	[−0.05, 0.42]
Politicized identity_t	Cross-group friendship_t+1	−0.04	0.06	.516	[−0.15, 0.07]
Moral convictions_t	Cross-group friendship_t+1	0.08	0.04	.059	[0.002, 0.16]
Collective action intentions_t	Politicized identity_t+1	0.12	0.06	.041	[0.004, 0.24]
Personal intergroup contact_t	Politicized identity_t+1	−0.13	0.08	.086	[−0.29, 0.02]
Cross-group friendship_t	Politicized identity_t+1	0.13	0.09	.124	[−0.04, 0.31]
Politicized identity_t	Politicized identity_t+1	0.19	0.10	.073	[−0.003, 0.41]
Moral convictions_t	Politicized identity_t+1	0.07	0.07	.331	[−0.06, 0.20]
Collective action intentions_t	Moral convictions_t+1	0.08	0.06	.173	[−0.04, 0.19]
Personal intergroup contact_t	Moral convictions_t+1	−0.03	0.08	.674	[−0.18, 0.12]
Cross-group friendship_t	Moral convictions_t+1	0.19	0.09	.030	[0.03, 0.36]
Politicized identity_t	Moral convictions_t+1	0.19	0.11	.068	[−0.01, 0.40]
Moral convictions_t	Moral convictions_t+1	−0.03	0.10	.778	[−0.23, 0.16]

Note. Unstandardized effects reported. 95% CIs were obtained with bootstrapping (10,000 re-samples). CA = collective action; CI = confidence interval; SE = standard error.

Figure 3.

Within-Person Longitudinal Associations for Collective Action Intentions, Personal Intergroup Contact, Cross-Group Friendship, Politicized Identity, and Moral Convictions (Stationary Random Intercept Cross-Lagged Panel Model).

Regarding the associations between collective action antecedents, we found a mutual positive relationship between moral convictions and cross-group friendship. Moreover, temporal increases in moral convictions were associated with the subsequent positive deviations from the trait level of personal intergroup contact.

In terms of within-person indirect effects (Table 5), we found a positive indirect effect of moral convictions at T1 on collective action intentions at T3 via cross-group friendship at T2. Similarly, there was a positive indirect effect of cross-group friendship at T1 on collective action at T3 via moral convictions at T2. By exerting a positive effect on collective action at T2, moral convictions at T1 had a positive indirect effect on politicized identity at T3. In a similar vein, cross-group friendship at T1 had a positive indirect effect on politicized identity at T3 by increased collective action at T2. Finally, there was a positive indirect effect of personal intergroup contact at T1 on collective action at T3 by cross-group friendship at T2.

Table 5.

Within-Person Indirect Effects Involving Collective Action Intentions (Stationary Model).

Focal predictor	Mediator	Outcome	IE	SE	p	95% CI
Personal intergroup contact_t1	Cross-group friendship_t2	Collective action intentions_t3	0.05	0.03	.072	[0.01, 0.12]
Personal intergroup contact_t1	Politicized identity_t2	Collective action intentions_t3	−0.01	0.01	.683	[−0.05, 0.01]
Personal intergroup contact_t1	Moral convictions_t2	Collective action intentions_t3	−0.01	0.01	.677	[−0.04, 0.01]
Cross-group friendship_t1	Personal intergroup contact_t2	Collective action intentions_t3	−0.02	0.02	.291	[−0.08, 0.004]
Cross-group friendship_t1	Politicized identity_t2	Collective action intentions_t3	0.01	0.01	.688	[−0.01, 0.05]
Cross-group friendship_t1	Moral convictions_t2	Collective action intentions_t3	0.03	0.02	.129	[0.003, 0.09]
Politicized identity_t1	Personal intergroup contact_t2	Collective action intentions_t3	0.01	0.01	.433	[−0.004, 0.06]
Politicized identity_t1	Cross-group friendship_t2	Collective action intentions_t3	−0.01	0.02	.515	[−0.06, 0.02]
Politicized identity_t1	Moral convictions_t2	Collective action intentions_t3	0.03	0.02	.149	[0.001, 0.09]
Moral convictions_t1	Personal intergroup contact_t2	Collective action intentions_t3	−0.02	0.02	.203	[−0.07, 0.001]
Moral convictions_t1	Cross-group friendship_t2	Collective action intentions_t3	0.02	0.02	.111	[0.003, 0.07]
Moral convictions_t1	Politicized identity_t2	Collective action intentions_t3	0.003	0.01	.730	[−0.01, 0.03]
Personal intergroup contact_t1	Collective action intentions_t2	Personal intergroup contact_t3	0.001	0.01	.928	[−0.01, 0.03]
Personal intergroup contact_t1	Collective action intentions_t2	Cross-group friendship_t3	−0.01	0.01	.240	[−0.04, 0.001]
Personal intergroup contact_t1	Collective action intentions_t2	Politicized identity_t3	−0.02	0.01	.172	[−0.06, 0.00]
Personal intergroup contact_t1	Collective action intentions_t2	Moral convictions_t3	−0.01	0.01	.257	[−0.05, 0.001]
Cross-group friendship_t1	Collective action intentions_t2	Personal intergroup contact_t3	−0.002	0.02	.927	[−0.04, 0.03]
Cross-group friendship_t11	Collective action intentions_t2	Cross-group friendship_t3	0.02	0.02	.242	[−0.001, 0.08]
Cross-group friendship_t1	Collective action intentions_t2	Politicized identity_t3	0.04	0.02	.074	[0.004, 0.09]
Cross-group friendship_t1	Collective action intentions_t2	Moral convictions_t3	0.03	0.02	.185	[−0.01, 0.08]
Politicized identity_t1	Collective action intentions_t2	Personal intergroup contact_t3	0.00	0.01	.970	[−0.01, 0.01]
Politicized identity_t1	Collective action intentions_t2	Cross-group friendship_t3	0.003	0.01	.706	[−0.01, 0.03]
Politicized identity_t1	Collective action intentions_t2	Politicized identity_t3	0.01	0.01	.689	[−0.01, 0.04]
Politicized identity_t1	Collective action intentions_t2	Moral convictions_t3	0.003	0.01	.700	[−0.01, 0.03]
Moral convictions_t1	Collective action intentions_t2	Personal intergroup contact_t3	−0.001	0.01	.923	[−0.02, 0.01]
Moral convictions_t1	Collective action intentions_t2	Cross-group friendship_t3	0.01	0.01	.172	[0.00, 0.04]
Moral convictions_t1	Collective action intentions_t2	Politicized identity_t3	0.02	0.01	.145	[0.01, 0.06]
Moral convictions_t1	Collective action intentions_t2	Moral convictions_t3	0.01	0.01	.297	[−0.001, 0.05]
Collective action intentions_t1	Personal intergroup contact_t2	Cross-group friendship_t3	−0.001	0.01	.938	[−0.02, 0.02]
Collective action intentions_t1	Personal intergroup contact_t2	Politicized identity_t3	0.001	0.01	.932	[−0.01, 0.02]
Collective action intentions_t1	Personal intergroup contact_t2	Moral convictions_t3	0.00	0.01	.973	[−0.01, 0.01]
Collective action intentions_t1	Cross-group friendship_t2	Personal intergroup contact_t3	0.01	0.01	.348	[−0.003, 0.04]
Collective action intentions_t1	Cross-group friendship_t2	Politicized identity_t3	0.01	0.01	.316	[−0.002, 0.04]
Collective action intentions_t1	Cross-group friendship_t2	Moral convictions_t3	0.01	0.01	.235	[0.00, 0.05]
Collective action intentions_t1	Politicized identity_t2	Personal intergroup contact_t3	−0.01	0.01	.420	[−0.04, 0.004]
Collective action intentions_t1	Politicized identity_t2	Cross-group friendship_t3	−0.004	0.01	.535	[−0.03, 0.01]
Collective action intentions_t1	Politicized identity_t2	Moral convictions_t3	0.02	0.02	.167	[0.001, 0.08]
Collective action intentions_t1	Moral convictions_t2	Personal intergroup contact_t3	0.01	0.01	.259	[−0.002, 0.04]
Collective action intentions_t1	Moral convictions_t2	Cross-group friendship_t3	0.01	0.01	.262	[−0.001, 0.03]
Collective action intentions_t1	Moral convictions_t2	Politicized identity_t3	0.01	0.01	.439	[−0.002, 0.03]
Collective action intentions_t1	Personal intergroup contact_t2	Collective action intentions_t3	0.001	0.01	.928	[−0.01, 0.03]
Collective action intentions_t1	Cross-group friendship_t2	Collective action intentions_t3	0.02	0.02	.242	[−0.001, 0.08]
Collective action intentions_t1	Politicized identity_t2	Collective action intentions_t3	0.01	0.01	.689	[−0.01, 0.04]
Collective action intentions_t1	Moral convictions_t2	Collective action intentions_t3	0.01	0.01	.297	[−0.001, 0.05]

Note. Unstandardized effects are reported. 95% CIs were obtained with bootstrapping (10,000 re-samples). For the full list of indirect effects, see the Online Supplement. CA = collective action; IE = indirect effect; CI = confidence interval.

Discussion

This research sought to investigate the within-person longitudinal relationships between collective action for a disadvantaged outgroup and its three widely postulated antecedents—politicized identity, moral convictions, and intergroup contact. Adopting RI-CLPM as a statistical framework, we analyzed data collected at the time of the refugee crisis spurred by Russia’s invasion of Ukraine. We found that advantaged group members (i.e., Poles) who reported stronger cross-group friendship and moral convictions than usual exhibited greater positive deviations from their typical level of willingness to engage for a disadvantaged outgroup (i.e., Ukrainians) 2 weeks later. By contrast, temporal changes in personal intergroup contact and politicized identity did not translate directly to subsequent deviations in participants’ collective action intentions. At the same time, however, positive deviations in collective action were associated with positive divergences from the average level of politicized identity (but not with changes in the remaining variables) in the following measurement. Moreover, we found a reciprocal positive association between cross-group friendship and moral convictions, and a positive prospective effect of moral convictions on personal intergroup contact. These significant cross-lagged effects served as building blocks for several indirect effects.

Our study extends previous research in several ways. First, it provides a longitudinal test for three hypothesized predictors of collective action for the disadvantaged outgroup—moral convictions, intergroup contact, and politicized identity. In accordance with past theorizing (Radke et al., 2020; van Zomeren et al., 2018), moral convictions had a positive prospective effect on engagement intentions. In other words, a temporal increase in viewing engagement for the disadvantaged outgroup as a matter of morality entailed the subsequent positive deviation from participants’ mean level of collective action intentions.

Likewise, a within-person increase in cross-group friendship predicted a consecutive increase in willingness to engage for the disadvantaged outgroup (see Hässler et al., 2020; Reimer et al., 2017). This, however, was not the case for personal intergroup contact. These findings corroborate the hypothesis that intergroup contact is particularly likely to motivate engagement for the disadvantaged outgroup when it reaches a critical quality threshold (see MacInnis & Hodson, 2019). The within-person indirect effect of cross-group friendship at T1 on collective action at T3 via moral convictions at T2 points to one of the mechanisms behind the positive effect of cross-friendship on engagement for the disadvantaged group. Specifically, having friends from the disadvantaged outgroup seems to prompt advantaged group members to perceive the fight for equality as a moral issue, which translates into higher engagement intentions. At the same time, the within-person indirect effect of personal intergroup contact at T1 on collective action at T3 via cross-group friendship at T2 was significant. This suggests that superficial intergroup contact motivated collective action for the disadvantaged outgroup by facilitating cross-group friendship.

Importantly, the current results diverge from the findings reported by Sengupta et al. (2022). As demonstrated by these authors, European New Zealanders’ (the advantaged group) contact with Māori (the disadvantaged group) had no within-person longitudinal effect on solidarity with the latter. One reason behind this inconsistency may be the different types of disadvantages experienced by Ukrainian refugees and Māori. The inferior status of Ukrainian refugees is incidental—it originates from an external cause, which, once removed, would leave this group no longer disadvantaged. By contrast, Māori face structural disadvantage resulting from years of systemic discrimination. As shown by van Zomeren et al.’s (2008) meta-analysis (but see Agostini & van Zomeren, 2021), structural disadvantage works as a contextual constraint upon some antecedents of collective action—the explanatory power of psychological variables is significantly smaller for structural than incidental inequalities. Thus, the inconsistent results of the two studies may reflect a general difference between collective action taken to tackle structural vs. incidental disadvantage. Second, the 2-week interval employed in the current research was much shorter than the 1-year lag used by Sengupta et al. (2022). Perhaps, long intervals between subsequent measurements are not optimal to detect longitudinal associations between intergroup contact and pro-outgroup engagement. At the same time, 2-week intervals may be insufficient to observe theory-based processes that the current study failed to find (i.e., the over-time effect of politicized identity on collective action). Finally, it is possible that our findings depart from those of Sengupta et al. (2022) because they capture a different phase in the lifespan of intergroup relationships. In their asymptotic model of intergroup contact, MacInnis and Page-Gould (2015; see also Page-Gould et al., 2022) proposed that the improvement in outgroup-directed attitudes is the strongest in early intergroup interactions and diminishes over time. If a similar dynamic applies to pro-outgroup engagement, intergroup contact should fuel collective action to a greater extent at the initial rather than late stages of intergroup relationships. As the percentage of Poles knowing at least one Ukrainian increased only recently,⁷ we believe that intergroup relationships registered in the present research were on average newer than those examined by Sengupta et al. (2022).

The current results also add to the literature on social identity and collective action. Unexpectedly, temporal within-person increases in politicized identity did not translate into subsequent positive deviations from the trait level of collective action. As the positive effect of social (e.g., politicized) identity on engagement is a common denominator of several influential models of collective action, the present results were not entirely consistent with past theorizing. Most importantly, the nonsignificant prospective effect of politicized identity is at odds with the predictions based on the social identity model of collective action (SIMCA; van Zomeren et al., 2008) and the encapsulated model of social identification and collective action (EMSICA; Thomas et al., 2009, 2012). Moreover, contrary to the extended SIMCA (van Zomeren et al., 2018), deviations from the trait level of politicized identity did not mediate the association between the temporal deviations in moral convictions and engagement intentions. Finally, inconsistent with the dual chamber model of collective action (Agostini & van Zomeren, 2021), social identity and morality did not work in parallel to motivate engagement for the disadvantaged outgroup. At the same time, temporal increases in politicized identity were predicted by preceding deviations in collective action—when participants’ engagement for the disadvantaged outgroup was greater than usual, their identification with the social movement helping the outgroup exceeded its trait level in the following measurement (see Becker et al., 2011; Thomas et al., 2020). These results invite two important conclusions. First, in some contexts, politicized identity may serve as a consequence rather than the antecedent of collective action. Perhaps this is the case when engagement for the disadvantaged outgroup does not originate from group-level motives (Radke et al., 2020). Second, given the positive association between random intercepts for politicized identity and collective action in the RI-CLPM, the established models of collective action may be a more accurate description of between- than within-person processes.

Furthermore, our findings shed light on the associations between the antecedents of collective action for a disadvantaged outgroup. For instance, we found a within-person reciprocal association between moral convictions and cross-group friendship—cross-group friendship had a positive prospective effect on moral convictions and was simultaneously predicted by the preceding deviations in moral convictions. Importantly, moral convictions did not show a reciprocal relationship with personal intergroup contact—while the within-person prospective effect of moral convictions on personal intergroup contact was positive, the opposite effect from personal intergroup contact on moral convictions did not reach significance. As such, in contrast to cross-group friendship, superficial intergroup contact did not entail the moralization of collective action for the disadvantaged outgroup (see MacInnis & Hodson, 2019).

Next, our study responds to calls for the greater use of longitudinal data in collective action research (e.g., Hässler et al., 2022). By employing a general community sample and short, 2-week intervals, it complements previous studies that used data collected among the committed supporters of a given cause (e.g., Thomas et al., 2021) or utilized longer (e.g., Sengupta et al., 2022) time lags between the subsequent measurements.

Finally, the current research adds to collective action literature by using the RI-CLPM (Hamaker et al., 2015) as an analytical framework. While collective action may be conceptualized both as a dynamic phenomenon and as a stable individual difference, this duality has rarely been recognized in empirical research so far (for the exceptions, see Sengupta et al., 2022; Thomas et al., 2021). Instead, past studies investigated temporal changes in engagement using the traditional CLPM (e.g., Górska et al.,2020, 2022) which does not separate within- and between-person variance (Hamaker et al., 2015), risking incorrect conclusions about intraindividual processes. By disentangling the within-person processes from stable between-person differences, the RI-CLPM allowed us to capture the dual (i.e., both static and dynamic) character of collective action.

In terms of practical implications, current findings suggest that, besides providing refugees with material assistance, decision-makers should not forget about creating opportunities for positive intergroup contact. As shown by our results, cross-group friendship motivates host society members to help refugees, either directly or by changing their moral convictions. Moreover, positive intergroup contact could mitigate the potential increase in antirefugee attitudes that may occur amid fears of economic recession (see Glick, 2005).

The present research has some limitations. Although RI-CLPM is the model of choice when one’s goal is to estimate reciprocal relations between the variables of interest (Usami et al., 2019), separating between- and within-person variance permits causal inference only after certain assumptions are made (Rohrer & Murayama, 2021). Specifically, one needs to assume that (a) all time-varying confounders are included in the model, (b) people do not vary in their change over time (i.e., do not have heterogeneous slopes, and (c) there are no contemporaneous causal effects (e.g., the effect of T1 cross-group friendship on T1 politicized identity).

Next, our findings are qualified by specific time lags between the subsequent measurements. As some processes (e.g., developing politicized identity based on one’s moral convictions) may require more than 2 weeks to fully unfold, the null results reported above could originate from the short timeframe of the present research. Relatedly, while a stationary model was a good fit to our data, it is entirely possible that using other (i.e., longer or interaction-based) intervals would reveal nonstationarity.

Supplemental Material

sj-docx-1-spp-10.1177_19485506221133882 – Supplemental material for Dynamic, Yet Stable: Separating Within- and Between-Person Components of Collective Action in Support of a Disadvantaged Outgroup and its Antecedents

Supplemental material, sj-docx-1-spp-10.1177_19485506221133882 for Dynamic, Yet Stable: Separating Within- and Between-Person Components of Collective Action in Support of a Disadvantaged Outgroup and its Antecedents by Paulina Górska and Nicole Tausch in Social Psychological and Personality Science

Footnotes

Handling Editor: Mallett, Robyn

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: Preparation of this manuscript was supported by the National Science Centre Poland (Opus Grant [2017/25/B/HS6/01116] and Harmonia Grant [2017/26/M/HS6/00689]).

ORCID iD

Paulina Górska

Supplemental Material

The supplemental material is available in the online version of the article.

Notes

Author Biographies

Paulina Górska is an adjunct professor at the Faculty of Psychology, University of Warsaw, Poland. Her research interests involve collective action, prejudice, minority stress, and political polarization.

Nicole Tausch is reader in social psychology at the University of St Andrews, Scotland. She is broadly interested in intergroup relations and conflict, prejudice and discrimination, and social perception. Her research examines the antecedents of different forms of collective action and the effects of intergroup contact in unequal intergroup relations.

References

Agostini

van Zomeren

(2021). Toward a comprehensive and potentially cross-cultural model of why people engage in collective action: A quantitative research synthesis of four motivations and structural constraints. Psychological Bulletin, 147(7), 667–700. https://doi.org/10.1037/bul0000256

Andersen

E. A.

Jennings

M. K.

(2010). Exploring multi-issue activism. PS: Political Science & Politics, 43(1), 63–67. https://doi.org/10.1017/S1049096510990604

Becker

J. C.

Tausch

(2015). A dynamic model of engagement in normative and non-normative collective action: Psychological antecedents, consequences, and barriers. European Review of Social Psychology, 26(1), 43–92. https://doi.org/10.1080/10463283.2015.1094265

Becker

J. C.

Tausch

Spears

Christ

(2011). Committed dis (s) idents: Participation in radical collective action fosters disidentification with the broader in-group but enhances political identification. Personality and Social Psychology Bulletin, 37(8), 1104–1116. https://doi.org/10.1177/0146167211407076

Besta

Pastwa-Wojciechowska

Jaśkiewicz

Piotrowski

Szulc

(2021). Radicalisation and individual differences: Disinhibition, boldness and meanness as predictors of support for radical collective action. Personality and Individual Differences, 168, Article 110354. https://doi.org/10.1016/j.paid.2020.110354

Bilewicz

Bukowski

Cichocka

Winiewski

Wójcik

(2009). Uprzedzenia etniczne w Polsce. Raport z ogólnopolskiego sonda u Polish Prejudice Survey 2009 [Ethnic prejudice in Poland: A report on the national Polish Prejudice Survey 2009] (Project report). University of Warsaw.

Brzozowski

(2022). Polish city creates “green corridor” for aid to Ukraine. Euractive. https://www.euractiv.com/section/europe-s-east/news/polish-city-creates-green-corridor-for-aid-to-ukraine/

Cakal

Halabi

Cazan

A. M.

Eller

(2021). Intergroup contact and endorsement of social change motivations: The mediating role of intergroup trust, perspective-taking, and intergroup anxiety among three advantaged groups in Northern Cyprus, Romania, and Israel. Group Processes & Intergroup Relations, 24(1), 48–67. https://doi.org/10.1177/1368430219885163

Cole

D. A.

Maxwell

S. E.

(2003). Testing mediational models with longitudinal data: Questions and tips in the use of structural equation modeling. Journal of Abnormal Psychology, 112(4), 558–577. https://doi.org/10.1037/0021-843X.112.4.558

10.

Davies

Tropp

L. R.

Aron

Pettigrew

T. F.

Wright

S. C.

(2011). Cross-group friendships and intergroup attitudes: A meta-analytic review. Personality and Social Psychology Review, 15(4), 332–351. https://doi.org/10.1177/1088868311411103

11.

Derewienko

(2022). Ilu uchodźców z Ukrainy jest w Polsce? [How many Ukrainian refugees are there in Poland?]. https://300gospodarka.pl/news/uchodzcy-z-ukrainy-w-polsce-liczba

12.

Drury

Reicher

(2009). Collective psychological empowerment as a model of social change: Researching crowds and power. Journal of Social Issues, 65(4), 707–725. https://doi.org/10.1111/j.1540-4560.2009.01622.x

13.

Glick

(2005). Choice of scapegoats In Dovidio

J. F.

Glick

Rudman

L. A.

et al. (Eds.), On the nature of prejudice: Fifty years after Allport (pp. 244–261). Wiley. https://doi.org/10.1002/9780470773963.ch15

14.

Górska

Stefaniak

Malinowska

Lipowska

Marchlewska

Budziszewska

Maciantowicz

(2020). Too great to act in solidarity: The negative relationship between collective narcissism and solidarity-based collective action. European Journal of Social Psychology, 50(3), 561–578. https://doi.org/10.1002/ejsp.2638

15.

Górska

Stefaniak

Marchlewska

Matera

Kocyba

Łukianow

Malinowska

Lipowska

(2022). Refugees unwelcome: Narcissistic and secure national commitment differentially predict collective action against immigrants and refugees. International Journal of Intercultural Relations, 86, 258–271. https://doi.org/10.1016/j.ijintrel.2021.11.009

16.

Hamaker

E. L.

Kuiper

R. M.

Grasman

R. P.

(2015). A critique of the cross-lagged panel model. Psychological Methods, 20(1), 102–116. https://doi.org/10.1037/a0038889

17.

Hässler

Ullrich

Bernardino

Shnabel

Laar

C. V.

Valdenegro

Ugarte

L. M.

(2020). A large-scale test of the link between intergroup contact and support for social change. Nature Human Behaviour, 4(4), 380–386. https://doi.org/10.1038/s41562-019-0815-z

18.

Hässler

Ullrich

Sebben

Shnabel

Bernardino

Valdenegro

Pistella

(2022). Need satisfaction in intergroup contact: A multinational study of pathways toward social change. Journal of Personality and Social Psychology, 22(4), 634–658. https://doi.org/10.1037/pspi0000365

19.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

20.

Kende

Lantos

N. A.

Belinszky

Csaba

Lukács

Z. A.

(2017). The politicized motivations of volunteers in the refugee crisis: Intergroup helping as the means to achieve social change. Journal of Social and Political Psychology, 5(1), 260–281. https://doi.org/10.5964/jspp.v5i1.642

21.

Louis

Thomas

McGarty

Lizzio-Wilson

Amiot

Moghaddam

(2020). The volatility of collective action: Theoretical analysis and empirical data. Political Psychology, 41, 35–74. https://doi.org/10.1111/pops.12671

22.

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(2), 242–263. http://doi.org/10.1111/josi.12165

23.

Lucas

R. E.

(2022). Time to abandon the cross lagged panel model. https://doi.org/10.31234/osf.io/pkec7

24.

Lüdtke

Robitzsch

(2021). A critique of the random intercept cross-lagged panel model [Preprint]. https://doi.org/10.31234/osf.io/6f85c

25.

Lytle

Dyar

Levy

S. R.

London

(2017). Essentialist beliefs: Understanding contact with and attitudes towards lesbian and gay individuals. British Journal of Social Psychology, 56(1), 64–88. https://doi.org/10.1111/bjso.12154

26.

MacInnis

C. C.

Hodson

(2019). Extending the benefits of intergroup contact beyond attitudes: When does intergroup contact predict greater collective action support? Journal of Theoretical Social Psychology, 3(1), 11–22. https://doi.org/10.1002/jts5.23

27.

MacInnis

C. C.

Page-Gould

(2015). How can intergroup interaction be bad if intergroup contact is good? Exploring and reconciling an apparent paradox in the science of intergroup relations. Perspectives on Psychological Science, 10(3), 307–327. https://doi.org/10.1177/1745691614568482

28.

Mallett

R. K.

Huntsinger

J. R.

Sinclair

Swim

J. K.

(2008). Seeing through their eyes: When majority group members take collective action on behalf of an outgroup. Group Processes & Intergroup Relations, 11(4), 451–470. https://doi.org/10.1177/1368430208095400

29.

Mondak

J. J.

Hibbing

M. V.

Canache

Seligson

M. A.

Anderson

M. R.

(2010). Personality and civic engagement: An integrative framework for the study of trait effects on political behavior. American Political Science Review, 104(1), 85–110. https://doi.org/10.1017/S0003055409990359

30.

Muthén

L. K.

Muthén

B. O.

(1998–2017). Mplus user’s guide (8th ed.).

31.

Orth

Clark

D. A.

Donnellan

M. B.

Robins

R. W.

(2021). Testing prospective effects in longitudinal research: Comparing seven competing cross-lagged models. Journal of Personality and Social Psychology, 120(4), 1013–1034. http://doi.org/10.1037/pspp0000358

32.

Osborne

Satherley

Little

T. D.

Sibley

C. G.

(2021). Authoritarianism and social dominance predict annual increases in generalized prejudice. Social Psychological and Personality Science, 12(7), 1136–1145. https://doi.org/10.1177/1948550620969608

33.

Page-Gould

Harris

MacInnis

C. C.

Danyluck

C. M.

Miller

I. D.

(2022). The intergroup perspective on cross-group friendship. Advances in Experimental Social Psychology, 65, 1–36. https://doi.org/10.1016/bs.aesp.2021.10.001

34.

Petriczko

(2022). For this border crisis, Poles extend a warm welcome, unlike last time New York Times. https://www.nytimes.com/2022/03/08/world/europe/ukraine-refugees-poland-russia.html

35.

Pettigrew

T. F.

Tropp

L. R.

(2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90(5), 751–783. https://doi.org/10.1037/0022-3514.90.5.751

36.

Radke

H. R.

Kutlaca

Siem

Wright

S. C.

Becker

J. C.

(2020). Beyond allyship: Motivations for advantaged group members to engage in action for disadvantaged groups. Personality and Social Psychology Review, 24(4), 291–315. https://doi.org/10.1177/1088868320918698

37.

Reimer

N. K.

Becker

J. C.

Benz

Christ

Dhont

Klocke

Hewstone

(2017). Intergroup contact and social change: Implications of negative and positive contact for collective action in advantaged and disadvantaged groups. Personality and Social Psychology Bulletin, 43(1), 121–136. https://doi.org/10.1177/0146167216676478

38.

Rohrer

J. M.

Murayama

(2021). These are not the effects you are looking for: Causality and the within-/between-person distinction in longitudinal data analysis [Preprint]. https://doi.org/10.31234/osf.io/tg4vj

39.

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(3), 539–560. https://doi.org/10.1111/bjso.12095

40.

Selvanathan

H. P.

Techakesari

Tropp

L. R.

Barlow

F. K.

(2018). Whites for racial justice: How contact with Black Americans predicts support for collective action among White Americans. Group Processes & Intergroup Relations, 21(6), 893–912. https://doi.org/10.1177/1368430217690908

41.

Sengupta

N. K.

Reimer

N. K.

Sibley

C. G.

Barlow

(2022). Does intergroup contact foster solidarity with the disadvantaged? A longitudinal analysis across seven years [Preprint]. https://psyarxiv.com/c6r29/

42.

Simon

Klandermans

(2001). Politicized collective identity: A social psychological analysis. American Psychologist, 56(4), 319–331. https://doi.org/10.1037/0003-066x.56.4.319

43.

Skitka

L. J.

(2010). The psychology of moral conviction. Social and Personality Psychology Compass, 4(4), 267–281. https://doi.org/10.1111/j.1751-9004.2010.00254.x

44.

Stefaniak

Winiewski

(Eds.) (2018). Uprzedzenia w Polsce 2017. Oblicza przemocy międzygrupowej [Prejudice in Poland. 2017 The shades of intergroup violence]. Liberi Libri.

45.

Tausch

Becker

J. C.

(2013). Emotional reactions to success and failure of collective action as predictors of future action intentions: A longitudinal investigation in the context of student protests in Germany. British Journal of Social Psychology, 52(3), 525–542. https://doi.org/10.1111/j.2044-8309.2012.02109.x

46.

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(1), 75–88. https://doi.org/10.1177/1368430211413619

47.

Thomas

E. F.

McGarty

Louis

W. R.

Wenzel

Bury

Woodyatt

(2021). It’s about time! identifying and explaining unique trajectories of solidarity-based collective action to support people in developing countries. Personality and Social Psychology Bulletin, 48, 1451–1464. https://doi.org/10.1177/01461672211047083

48.

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. https://doi.org/10.1177/1088868309341563

49.

Thomas

E. F.

Zubielevitch

Sibley

C. G.

Osborne

(2020). Testing the social identity model of collective action longitudinally and across structurally disadvantaged and advantaged groups. Personality and Social Psychology Bulletin, 46(6), 823–838. https://doi.org/10.1177/0146167219879111

50.

Usami

Murayama

Hamaker

E. L.

(2019). A unified framework of longitudinal models to examine reciprocal relations. Psychological Methods, 24(5), 637–657. http://doi.org/10.1037/met0000210

51.

Van Zomeren

Kutlaca

Turner-Zwinkels

(2018). Integrating who “we” are with what “we”(will not) stand for: A further extension of the social identity model of collective action. European Review of Social Psychology, 29(1), 122–160. https://doi.org/10.1080/10463283.2018.1479347

52.

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(2), 180–199. https://doi.org/10.1177/1088868311430835

53.

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(4), 504–535. https://doi.org/10.1037/0033-2909.134.4.504

54.

Van Zomeren

Postmes

Spears

(2012). On conviction's collective consequences: Integrating moral conviction with the social identity model of collective action. British Journal of Social Psychology, 51(1), 52–71. https://doi.org/10.1111/j.2044-8309.2010.02000.x

55.

Waddell

J. T.

King

S. E.

Corbin

W. R.

(2022). Dynamic longitudinal relations among solitary drinking, coping motives, & alcohol problems during emerging adulthood. Drug and Alcohol Dependence, 238, Article 109576. https://doi.org/10.1016/j.drugalcdep.2022.109576

56.

Selig

J. P.

Little

T. D.

(2013). Longitudinal data analysis. In Little

T. D.

(Ed.), The Oxford handbook of quantitative methods: Vol. 2. Statistical analysis (pp. 387–410). Oxford University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.55 MB

Dynamic,Yet Stable: Separating Within- and Between-Person Components of Collective Action in Support of a Disadvantaged Outgroup and its Antecedents

Abstract

Keywords

Collective Action for Disadvantaged Outgroups

The Dual Nature of Collective Action

The Current Research

Method

Sample

Measures

Collective Action Intentions3

Personal Intergroup Contact 4

Cross-Group Friendship

Politicized Identity

Moral Convictions

Results

Analytic Strategy

Random Intercept Cross-Lagged Panel Model6

Stationarity Test

Between-Person Level

Within-Person Level

Discussion

Supplemental Material

sj-docx-1-spp-10.1177_19485506221133882 – Supplemental material for Dynamic, Yet Stable: Separating Within- and Between-Person Components of Collective Action in Support of a Disadvantaged Outgroup and its Antecedents

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

Supplemental Material

Notes

Author Biographies

References

Supplementary Material

Personal Intergroup Contact⁴