When Small Acts Are Multiplied: Assessing Everyday Social Justice Behaviors

Using the Act Frequency Approach, we drew on majority White, U.S. samples to create a new measure of social justice behavior and examine its correlates. Although existing measures of social justice behavior focus on engagement in collective action, participants in Study 1 (n = 137) were encouraged to nominate and evaluate a broad set of acts relevant to their daily lives. The final 17-item Everyday Social Justice Behavior (ESJB) scale reflects a range of global and domain-specific actions rated as prototypical by both 53 undergraduate novices and 20 social justice experts in Study 2. Participants in studies 3 (n = 388) and 4 (n = 613) were then asked to rate how frequently they perform the items. As expected, women and sexual minorities, and those with left political orientation, engaged in more everyday social justice behavior. Moreover, those reporting more everyday social justice behavior also scored higher in structural attributions of social change, intersectional awareness, ratings of the importance of and confidence in taking action, openness to experience, extraversion, and empathy, while being lower in social dominance orientation, system justification, and the need for cognitive closure. In addition, those high in ESJB also reported more progressive activist engagement and intentions. Relations with activism were modest, suggesting social justice activism and ESJB are somewhat distinct forms of social justice behavior. This measure should be of broader use in similar (majority White) samples; the measure development process can also be used to assess such behaviors in other samples and contexts.

People who promote social justice in their personal lives may also engage in public collective action, which has been studied much more extensively.For example, social identity theorists (Reicher, 1984;Tajfel, 1978;Tajfel & Turner, 1979) have studied the relations between common social identifications and social justice behavior, using engagement in collective action (van Zomeren et al., 2008).In addition, Duncan (1999Duncan ( , 2010Duncan ( , 2022) ) studied how group consciousness motivates group actions that undermine institutionalized power inequities.Collective action is a critical factor in social change, but we propose that the actions individuals take that promote fairness in their everyday lives are also important.
Individuals who engage in collective action may not always challenge inequities in their day-to-day lives; their interpersonal interactions may sometimes replicate the power structures that activists work to resist when fighting for a "cause." For example, Evans (1979) chronicled how women in the new left movement of the 1960s and 1970s often faced sexism from their male counterparts, leading some to initiate the second wave of the women's movement.Some of these same women, however, treated sexual minority feminists or feminists of color in the same dismissive manner, reinforcing power relations based on race and sexual orientation (Collins, 1989(Collins, , 1990;;Crenshaw, 1991).Thus, individuals are not always consistent in pursuing social justice.
Ultimately, many kinds of actions can challenge the status quo and promote social justice, though existing measures focus on only a few.A student may stand up to a bully, a friend may share a petition on social media, a colleague may help to make the workplace more inviting to a new employee who is a member of an underrepresented group, or a person might avoid derogatory language across situations.Our goal is to assess social justice behaviors that currently fall outside existing measures.Because these actions may arise in a range of interpersonal situations, and be motivated by a variety of traits, motives, goals, or intentions, we adopted the Act Frequency Approach (or AFA) to assess it.

The Act Frequency Approach
Originally used in the development of a measure for the personality disposition of dominance, the AFA was designed to allow researchers to develop measures for "fuzzy categories" (Buss & Craik, 1983).Over three stages, the AFA is a technique that allows researchers to systematically define constructs that are otherwise difficult to operationalize.Participants are first asked to nominate actions that best represent the construct of interest, which generates a wide range of items, some commonly nominated and others unique to a single participant.The AFA then provides an opportunity to refine the construct through both expert and novice prototypicality ratings.Both are considered valuable and important sources of data, since the understandings of "novices" (that is, ordinary individuals rather than social justice "experts") provide key information about how social behaviors are understood (Latour, 2011).Assessment of the ways that both experts and novices understand concepts are common in developmental (Gelman, 2003), personality (Bröhl et al., 2022;McCrae et al., 1993), social (Reser & Muncer, 2004); and cognitive psychology (Gelman & Legare, 2011).In the AFA approach, after the most prototypical items are selected and a measure created, participants are asked to rate how frequently they themselves perform the actions and complete other measures of interest.The construct validity of the resulting scale can then be evaluated in terms of its relations to other indicators of interest, per the usual approach to construct validation (see, for example, Bröhl et al., 2022;Hogan & Nicholson, 1988;McCrae et al., 1993).
Although the Act Frequency Approach has been criticized for its use in the assessment of personality traits (Block, 1989;Moser, 1989), where specific behaviors may be reflections or expressions of multiple traits, it has been successfully used to identify a range of actions that fall under "fuzzy" behavioral constructs, like everyday social justice behaviors.For example, Ivcevic (2007) used the AFA to define differences between artistic and everyday creativity in terms of specific behavior.Other studies have similarly used the AFA to define the behavioral content of ideal mating strategies (Jonason et al., 2009), patient empowerment (Faulkner, 2001), social intelligence (Willmann et al., 1997), impulsivity (Estrella Romero et al., 1994), generativity (McAdams & de St. Aubin, 1992), and organizational change (Szamosi & Duxbury, 2002).

Conceptualizing Everyday Social Justice Behaviors
In this research we were interested in what people think they and others around them do in their everyday lives that promotes fairness, equity, or "social justice." Thus, we hoped to move beyond researchers' and activists' understanding of collective action and social movement behavior to encompass the kinds of "small acts" of inclusion, equity, and even confrontation that we know from experience occur in people's everyday lives to counter the acts of "everyday discrimination" that have been studied alongside racism, sexism, and classism (Bourabain & Verhaeghe, 2021;Hyers, 2010;Showunmi & Tomlin, 2022).
We know that relatively few privileged individuals engage in social movement activism on behalf of less privileged; but some speak up when they witness someone being ignored or overlooked (Rodríguez et al., 2021).Because we believe it is important to understand and promote social justice behaviors among privileged individuals, these actions-the logical corollary of everyday discrimination against marginalized individuals, and a complement to collective or social movement activism that may be adopted by groups on their own behalf-were the focus of our interest.

Personality and Attitudes Related to Social Justice Behavior
Research on social justice behavior in psychology has previously demonstrated some consistent relations between personality dispositions and beliefs about social inequities, commitment to social change, and activism.These include: structural awareness, or the ability to recognize institutionalized power inequities, which has been associated with social justice behaviors (Kluegel, 1990;Lopez et al., 1998); beliefs about the merits of social inequities, or refusal to acknowledge problems with the status quo (Kay & Jost, 2003;Pratto et al., 1994); awareness of the interconnectedness of oppression along different dimensions (Collins, 1989;Crenshaw, 1991;Curtin et al., 2015;Greenwood, 2008); and rec ognition of the importance of reducing prejudice and promoting diversity, as well as confidence about one's individual ability to act, are logical precursors to engaging in social justice behavior (Nagda et al., 2004).
The broader personality trait of Openness to Experience (Curtin et al., 2010;John et al., 1991) has also been linked to political engagement.Need for Cognitive Closure is negatively related to Openness to Experience (Roets & van Hiel, 2011) and may predict lower levels of social justice behavior.Extraversion is also a potential predictor of the kind of interpersonal engagement required.Similarly, empathy (Hoffman, 1990), emotional closeness with stigmatized groups (Fingerhut, 2011), and other affective predictors play a role in motivating social justice action.
One possibility is that collective action requires a stronger basis in a structural analysis and a disposition toward activism than does everyday social justice behavior, which may depend more on other personality dispositions relevant to the interpersonal sphere.This might include aspects of dispositional person perception (like perspective taking and empathy) as well as an understanding of complex social positions held by other people, like intersectional awareness.

Overall Research Questions and Hypotheses
We used the three-study strategy associated with the Act Frequency Approach (Buss & Craik, 1980) to create and validate a new measure of social justice behavior that captures a broad range of everyday social justice actions that are considered normative in samples of activists, experts, as well as members of the public from the United States.This measure is intended to assess behaviors not routinely studied in the literature on political participation or collective action.A fourth study assessed its validity further, using an online sample of adults from the United States.Although the measure was generated and validated using samples from the United States, with a skew toward being more liberal, educated, and white (that is, relatively privileged), we are hopeful that this measure and procedure can be adapted and revised for use in a broader range of contexts.The first study used the Act Nomination Procedure to explore the following research question: R1: How do people promote fairness, inclusion, and equity or challenge unfairness, exclusion, and inequity in their everyday lives?
Using the list of multiple acts generated by Study 1, the second study asked novice and expert raters: R2: Which of the actions promoting fairness, inclusion, and equity are the most prototypical of everyday social justice behaviors?
After producing a final scale from the typicality ratings in Study 2, participants in Studies 3 and 4 were asked to rate how frequently they engaged in each of the social justice behaviors and completed a series of related measures.The following hypotheses were tested: H1: Membership in a marginalized group and left political orientation will relate to scores on the social justice behavior scale.

H2:
Reporting social justice behaviors will be positively related to indicators of understanding structural power relations and other social-cognitive indicators that have been previously associ ated with progressive social actions.

H3:
Reporting social justice behaviors will correlate moderately with other established measures of progressive activist engagement, suggesting that they are related, but distinct, measures.

Procedure
Participants were asked to nominate examples of social justice actions both "in general" and (to prompt responses that might not arise more generally) in terms of specific domains (e.g., with respect to race, gender, sexuality, etc.).See the full prompt in Supplementary Materials.
Participants were randomly assigned to generate a list of ways in which people may promote fairness, inclusion, and equity related to one of the following interpersonal settings: 1) men and women, 2) people of different racial or ethnic groups, 3) people of different socioeconomic or social class backgrounds, 4) people of different ages, 5) people of different sexual orientations, or 6) people of different physical, mental, or emotional abilities and limitations.Finally, participants were asked to answer a series of demographic questions, their political orientation, and whether they identified as an activist.

Results and Discussion
On average, participants nominated 3.34 acts in response to the Global Justice prompt (n of participants who named at least one act = 137, SD = 1.81 acts).In response to the domain-specific prompts, participants on average nominated 2.68 acts promoting Gender Justice (n = 25, SD = 1.34), 2.41 acts promoting Race/Ethnicity Justice (n = 22, SD = 1.56), 2.33 acts promoting Social Class Justice (n = 30, SD = 0.99), 2.35 acts promoting Age Justice (n = 20, SD = 1.60), 2.92 acts promoting Sexual Orientation Justice (n = 13, SD = 1.50), and 2.52 acts promoting Disability Justice (n = 21, SD = 1.12).Consistent with the norms of AFA research, both commonly named actions (e.g."make it a point to include everyone in activities, " "treat everybody equally") and unique actions (e.g."protest things you believe are wrong") were included in the final list, but responses that did not really provide specific actions (e.g."my mother, " "they are intelligent") were eliminated, as were actions that were redundant.Frequencies for each of the nominated actions are reported in Table A1 of the Supplementary Materials.Although interesting patterns/frequencies could already start to be assessed with the results of Study 1 (e.g., many of the acts nominated were in the Global Justice category, even if they were directed to answer for a domain-specific category), it was important to use an additional study to reduce the number of acts.

Method Participants
Because previous research has indicated the importance of including both expert and novice raters of prototypicality, we recruited both in Study 2. According to previous research (Buss & Craik, 1980;Ivcevic, 2007), 20-100 raters are needed for this stage of the process.Participants in this study included 73 individuals: 53 students recruited for a new sample of undergraduates in the Introduction to Psychology subject pool, and 20 academic and activist "social justice experts" recruited from a variety of activist organizations and relevant academic departments at a large university in the Midwest region of the United States.More specific information about the sample can be found in Supplementary Materials.

Procedure
Participants in Study 2 rated the social justice-related actions from Study 1 in terms of their prototypicality, on a scale from 1 (least) to 5 (most).The original directions recommended by Rosch and Mervis (1975) and used by Buss and Craik (1980) were adapted to eliminate unfamiliar terms like "prototypicality." The precise wording of our directions to participants is included in the Supplementary Materials.Finally, participants were asked to answer the same series of demographic questions as the participants in Study 1.

Results and Discussion
For both samples, the mean typicality scores were calculated for each of the 71 items (see Supplementary Materials, Table A2) produced by Study 1 1 .The correlation between student and activist typicality ratings on all the items was statistically significant, but moderate (r = .34).
Seventeen acts were rated above the group median 2 (3.8 for students; 4.37 for social justice experts) on typicality by both groups; these items became the final scale, reflecting their joint input.Of course, this risks leaving out rare but important actions.Because we intend to measure common everyday action, we believe our focus on those consensually viewed as most prototypical was justified.Therefore, items that were rated as very typical examples of social justice behavior for only one of the two groups were not included (see Table A2 in Supplementary Materials for more details).Additionally, the item "promote respect" was rated highly by both groups, but was also very similar to "treat people with respect."; we decided to not include this redundant item.
The many items on which students and activists converged included behaviors that promote fairness, equity, and inclusion (e.g., "Stand up for people who are being treated unfairly") and discourage prejudice or discrimination across contexts (e.g., "Speak out against racism").This indicates that the shared view of typical social justice behavior encompasses actions that emphasize respectful interpersonal interaction and promote structural change.We also note that these actions require a sense of entitlement to speak out that in turn may depend on privileged standing.
Study 2 resulted in 17 unique social justice behaviors for inclusion in the final scale.Across the 17 items, there was a strong emphasis on interpersonal actions that promote fairness in one's day-to-day life (e.g., "Stand up for people who are being treated unfairly" "create an environment in which everyone can express their ideas, " "treat people with respect, " etc.).Although some of the items reflect possible participation in collective action (e.g., "stand up for people who are being treated unfairly"), their wording was flexible enough to encompass everyday interpersonal actions.Subject pool studies only provide potential respondents with the IRB number (e.g., HUM 1234000), the study location, and the amount of time required, in order to avoid introducing selection bias.Additionally, no participants from Studies 1 or 2 participated in Study 3. On average, participants in the sample were 19.7 years old, and ranged in age from 18 to 48.On a scale of 1 to 4, the average GPA for the sample was 3.18.Most participants self-identified as female (69.1%), with 30.6%; as male; only one participant self-identified as genderqueer.The sample was predominantly White (73.4%), while 13.6% of participants self-identified as Asian, and smaller percentages with other racial-ethnic groups.Due to the small number of students in several racial categories, the variable was dichotomized into "White" and "non-White" for subsequent analyses.About 5% of the sample (n = 18) identified as international students.Eighty-four percent of the sample self-identified as heterosexual, so sexual orientation was dichotomized into "completely heterosexual" and "not completely heterosexual" for subsequent statistical analyses.Combining both visible and invisible types of disabilities, 3.6% of the sample self-identified as having a disability.Many undergraduate respondents (44.4%) described their family situation as "well to do" or "extremely well to do, " while 31.9% reported having "more than enough to get by." The sample was politically moderate, though somewhat left of center; on a scale of 1 (liberal) to 7 (conservative), the average score on political orientation was 3.29 (SD = 2.10).Finally, most of the undergraduate sample did not identify as activists (79.8%), while a substantial portion of the sample (15.1%) were "unsure." In Study 3 we used case-wise deletion to handle any missing responses to questions.t-tests and chi square analyses between groups did not reveal any systematic differences between those who had missing data and those who did not.

Measures
To avoid priming responses to the personality and attitude measures, participants completed the ESJB scale towards the end of the survey Descriptive statistics for all relevant measures, including their internal consistency, are reported in Table 1.Note.Ns vary due to missing data, as discussed in Method section.
Structural Attributions for Targets of Social Change -Participants were asked to respond to three vignettes modified from those used in Lopez et al. (1998).Each vignette outlined a situation in which an interpersonal conflict could be addressed with a variety of resolutions.For example, in response to an organizational slight, Jiang could "try to be less sensitive" (individual) or make others aware of the conflict "by distributing flyers, writing a letter in the school newspaper, or organizing a workshop on the issue" (structural).Participants received a score ranging from 0 to 3, which corresponded to their structural attributions.Scoring instructions can be found in Supplementary Materials.
Social Dominance Orientation -Social Dominance Orientation was measured with eight items from Pratto et al. (1994).Sample items include: "Some groups of people are simply inferior to other groups" and "We should strive to make incomes as equal as possible" (reverse-scored).
System Justification -System Justification was measured with eight items from Kay and Jost (2003).Sample items include: "Most policies serve the greater good" and "Our society is getting worse every year" (reverse-scored).
Intersectional Awareness -Awareness of intersectionality was measured with 8 items from the measure developed by Curtin and colleagues (2015).A sample item includes: "Black and White women experience sexism in different ways." Action Importance and Confidence -Action Importance and Confidence were measured with eleven items from Nagda and colleagues (2004).Sample items include: "refrain from repeating statements or rumors that reinforce preju dice or bias" and "Make efforts to get to know individuals from diverse backgrounds." For each item, participants were asked how important it is for them do it and how confident they feel about their ability to do it.Openness to Experience and Extraversion -The personality traits of openness (10 items) and extraversion (8 items) were measured using the corresponding sub-scales of the Big Five Inventory (John & Srivastava, 1999).For openness, sample items include: "is inventive" and "likes to reflect, play with ideas." For extraversion, sample items include: "has an assertive personality" and "is reserved (reversed)."

Need for Cognitive
Empathy -The personality trait of empathy was measured with 14 items reflecting the Empathic Concern (EC) and Perspective Taking (PT) subscales from the Interpersonal Reactivity Index (Davis, 1980(Davis, , 1983)).Sample item includes: "I often have tender, concerned feelings for people less fortunate than me" (EC).
Collective Action for Social Justice -Previous engagement in Collective Action for Social Justice (CASJ) was assessed using a measure from previous research on activism (Curtin et al., 2010;Duncan, 1999).Participants were asked to indicate which activities they previously pursued for a variety of causes, such as signing a petition and/or attending a rally or demonstration.For each of the six engagement acts, participants could receive a score of 0 (was not active) or 1 (was active).A summed score of 0-6 was then calculated for participation in each cause, and these were averaged for all progressive causes to create the scale score.Seventeen causes were included in the measure of Collective Action for Social Justice (CASJ), including anti-racism and women's rights.See Supplementary Materials for a complete list of causes.
Everyday Social Justice Behavior -Table 2 shows the average frequency for each of the 17 items of the final Everyday Social Justice Behavior (ESJB) scale generated in Study 2, ordered from most frequently endorsed item to least.Although our intent was to measure everyday social justice behavior across different domains, we examined the factor structure of the ESJB items.Principal Components Analysis showed that the first factor accounted for 37.54% of the variance, the next three for 9.85%, 8.14% and 5.88% respectively.Two-to four-factor solutions using varimax rotation were examined.However, these solutions failed to generate a simple factor structure with few cross-loadings, so, consistent with usual practice, the unifactorial solution was selected (Wegener & Fabrigar, 2004).

Results and Discussion
As predicted, women scored higher on the Everyday Social Justice Behavior (ESJB) scale than men (r = .16,p = .005).Individuals who self-identified as more conservative scored lower on ESJB than individuals who self-identified as liberal, r = -.17,p = .002.However, race, sexual orientation, perceived social standing, and ability status were unrelated to ESJB, .02< |r| < .08,p > .05.As predicted, ESJB was not significantly correlated with age, GPA, undergraduate class standing, or type of residence, |.02| < r < |.10|, p > .05.These results only partially support our first hypothesis that Everyday Social Justice Behaviors would be more common among members of marginalized groups, perhaps due to a relative lack of diversity in this homogenous student sample or because the items are better suited to actions by majority group members.Supporting Hypothesis 2, ESJB was significantly, positively related to structural attributions of social change targets, the importance and confidence in taking action, intersectional awareness, openness, extraversion, and empathy, and was significantly negatively related to Social Dominance Orientation (SDO), System Justification, and the Need for Cognitive Closure (NfCC).ESJB was moderately correlated with Collective Action for Social Justice (r = .37,p < .001),supporting Hypothesis 3 (see Table 3 for all inter-item correlations).Given the homogeneity of our sample in Study 3, we wanted to conduct a follow-up and further establish the scales' construct validity.

Study 4: Validation of the Everyday Social Justice Behaviors Scale in a Non-Student Sample
Since we had some concerns regarding external validity based on results from Study 3 we further validated our scale using an online sample of adults from the United States.Using data from a larger study, we examined the correlates of ESJB in a similar fashion to Study 3. The purpose of this larger study was to investigate how personality and social identity factors (many, but not all, of which we discuss here) were related to people's votes in the 2020 presidential election, as well as political activism and trust in social institutions, like the media and government (see Savaş et al., 2021, for details).Since this larger study was not focused on replicating Study 3, some new variables could be examined, including political intentions (activism and radicalism).

Method Participants and Procedure
Participants in Study 4 were recruited from Amazon Mechanical Turk (MTurk) as part of a larger study (see Savaş et al., 2021 for more details).Participants were surveyed in two waves, which occurred prior to and after the 2020 presidential election (n = 613).See Supplementary Materials for details about drop-out rates, data integrity checks, and ethical compliance.Since the larger study was focused on voting behaviors in the 2020 U.S. election, we targeted U.S. citizens, who were registered to vote.The sample was demographically diverse, especially in terms of race-ethnicity.This was because we also used tools available through the CloudResearch platform (Litman et al., 2017) to explicitly target a diverse sample for the first wave of the study (Goodman et al., 2013).This resulted in a final sample that was 24.8% Black (n = 152), 18% Latino/Latina/Hispanic (n = 111), 24.6% Asian/Pacific Islander (n = 151), 3.1% Native American (n = 19), and 35.2% White (n = 216).Additionally, 19 individuals (3.1%) identified as bi/multi-racial, and one individual wrote in that they identified as Arab.In terms of gender, 47% (n = 288) identified as men and 51.5% (n = 316) identified as women, with 1.5% (n = 9) identifying as transgender or gender non-conforming/non-binary. In terms of sexual orientation, 3.4% (n = 21) identified as gay or lesbian, 9% (n = 55) identified as bisexual, 1% (n = 6) identified as none of the provided options and wrote in identities like "asexual, " "pan, " or "queer." Most of the sample, 86.6% (n = 531), identified as straight.The sample was also diverse in terms of socio-economic variables.About one quarter of the sample identified as working class (n = 154), with 20% as lower-middle (n = 124), 46% as middle (n = 282), 8.3% as upper middle (n = 51), and less than 1% (n = 2) as upper class 3 .
Finally, in terms of political party affiliation, a slight majority (n = 312, 50.9%) identified as Democrats, with 116 (18.9%)Republicans, 24.5% Independents (n = 150), and a few others.Rated on a 7-point scale from very liberal (1) to very conservative (7), they reported an average of 3.28 and a standard deviation of 1.79.We examined the correlates of our student-based measure of everyday social justice behaviors to assess their applicability to a larger, much more diverse sample of U.S. adults.

Measures
Participants completed all measures in the order below.All demographics, individual difference measures, and the ESJB measure were completed in Wave 1, while the traditional activism measures and candidate-related measures were completed in Wave 2. Participants were allowed to skip any questions.See Table 4 for descriptive statistics, as well as reliability coefficients.a Openness to Experience and Extraversion were each measured using the mean of two items.The corresponding items were moderately correlated with each other (r = .43and .61,respectively) and significant (p < .001),indicating acceptable internal reliability.

Openness to Experience and Extraversion -
The Big Five, including openness to experience and extraversion, was measured using the Ten Item Personality Inventory (TIPI) (Gosling et al., 2003).Participants rated the extent that each item characterized how they thought of themselves (i.e., "I am…"), using a 7-point agreement scale.
Social Dominance Orientation -Social Dominance Orientation was measured using an 8-item scale, developed by Ho and colleagues (2015).Participants rated their level of agreement with each item using a 7-point agreement scale.Higher values indicated more agreement with SDO-related attitudes.
Intersectional Awareness -The Intersectional Awareness Scale used the same 8-items as in Study 3 (Curtin et al., 2015).
Everyday Social Justice Behaviors -Everyday Social Justice Behaviors were measured as in Study 3, except participants rated their frequency of engaging in each of the behaviors on a 7-point scale from 1 "never" to 7 "always" with a mid-point (4) of "sometimes, " instead of the 5-point scale used previously.
Collective Action for Social Justice (CASJ) -Collective Action for Social Justice (CASJ) was measured as in Study 3, but some slightly different "causes" were listed than in Study 3, as different collaborators with different particular interests were engaged in this project, and the group focused on a specific time-period, where some issues were more salient; items included: civil rights, prison reform, criminal justice reform, Black Lives Matter, defund the police, sanctuary movement (for immigrants), immigrant rights, support for dreamers/DACA, abolish I.C.E., women's rights, treatment of women, #MeToo, pro-choice, disability rights, transgender rights, LGBT rights, workers' rights, and environmental issues.

Activism and Radical Intentions Scale (ARIS) -
We used a modified version of the ARIS scale (Moskalenko & McCauley, 2009), comprising two subscales, activism and radical intentions, analyzed separately.Each participant was asked to think about a political group or cause to which they felt especially connected.They were asked to name the group and indicate their agreement with 10 items, using a 7-point Likert scale.An example item for the radical intentions sub-scale is "I would participate in a public protest even if I thought the protest might turn violent, " while an example item for the activism sub-scale is "I would travel for one hour to join in a public rally, protest, or demonstration in support of my group." The scoring procedure (discussed in Supplementary Materials) produced scores of -7 (strong right-wing activism/radical intentions) to 7 (strong left-wing activism/radical intentions).

Results and Discussion
Descriptive Statistics, Correlations, and Group Differences in ESJB Descriptive statistics are reported in Table 4. Correlations between ESJB and social identity group memberships were consistent with those discussed in Study 3, again offering mixed support for Hypothesis 1. Women, r = .15,p < .001,and people with a more liberal political orientation, r = -.20,p < .001,were again significantly more likely to engage in Everyday Social Justice Behaviors.Race, F(3, 584) = 1.94, p = .12,and social class standing, r = -.04,p = .29,were not significantly associated with ESJB.There was a trend for sexual orientation, such that sexual minorities were more likely to engage in these behaviors, but this did not reach traditional levels of significance, r = .07,p = .07.Again, although we had a much more diverse sample, in terms of race, some of these lower correlations could have been a result of a lack of diversity in these areas.It is also possible that contextual factors or the items themselves could have produced this lack of effect (discussed later).

Correlates of the Everyday Social Justice Behavior (ESJB) Scale
Supporting Hypothesis 2 and consistent with Study 3, ESJB scores were significantly, positively related to the personali ty traits of Openness to Experience, Extraversion, and Intersectional Awareness, and significantly negatively related to Social Dominance Orientation (see Table 5).Supporting Hypothesis 3, ESJB was positively correlated with endorsement of the pro-justice pole of the activism and radical intentions subscales of the ARIS measure.However, unlike for Study 3, ESJB was, surprisingly, not correlated with Collective Action for Social Justice.
In general, the results of Study 4 were consistent with the findings from Study 3, using a more diverse, non-student, adult sample.There were also some differences in the means and correlations for some of the variables assessed in the same way.

General Discussion
In four studies, a new measure of everyday social justice behavior in U.S. samples was created and validated using the Act Frequency Approach (Buss & Craik, 1980), as well as conventional validation procedures.The final 17-item Every day Social Justice Behavior (ESJB) scale reflects a range of global and domain-specific actions rated as prototypical by both undergraduate novices and social justice experts.The items focus on interpersonal, everyday domains, addressing inclusion, respectful treatment, and equality of access.
In the last two studies we examined how these everyday social justice behaviors were related to other variables in expected ways.ESJB scores were, overall, positively related to structural attributions of social change, intersectional awareness, beliefs about the importance and confidence in taking action, openness to experience, extraversion, and empathy, while being negatively related to social dominance orientation, system justification, and the need for cognitive closure.These results match our expectations for Hypothesis 2, given that previous research which has found similar patterns of correlations between these and more traditional measures of social justice-related collective action.Previous research has consistently found that members of marginalized groups are more likely to engage in actions aimed at achieving social justice, with mechanisms ranging from classic social identity theory to enhanced group consciousness (Reicher, 1984;van Zomeren et al., 2008).Therefore, we expected that holding these identities would be similarly related to our new measure of everyday social justice behaviors.We, however, found mixed support, as membership in some marginalized groups (e.g., women) were positively related to ESJB, while membership in other marginalized groups (e.g., racial-ethnic minorities) was unrelated to ESJB in our studies.Perhaps our measure did not include enough relevant items in the domains of social class, sexual orientation, and race, or perhaps individuals marginalized in terms of these identities do not feel in a position to take these kinds of actions.Some previous research has noted that interpersonal social actions, such as "standing up against racism, " are not only rare, but also highly context-dependent, even among members of marginalized groups (Ashburn-Nardo & Abdul Karim, 2019).The propensity to engage in such actions is influenced by a myriad of factors, including personal goals (e.g., desiring respect versus being liked) that one holds during potentially contentious interactions (Mallett & Melchiori, 2019).
ESJB was also correlated moderately with progressive activist engagement (though, surprisingly, only in Study 3), as well as future activism and radical intentions, suggesting that these constructs are related, yet distinct.The differences between studies 3 and 4 could be a result of the peculiarities of the sample.It is possible that our students (often more activist than older adults) in study 3 had more privilege, time, resources, or had recently-developed a critical consciousness, which prompted more traditional collective action than in our adult, non-college sample.
Overall, our studies present a clear contribution to the field, in the form of a measure of Everyday Social Justice Behaviors (ESJB), which was related to existing activism measures, while remaining distinct in its focus on those behaviors that occur in everyday contexts.

Further Research
In addition to our main goal across our four studies, which was to construct and validate a novel measure of social justice behaviors in everyday contexts, our findings also contribute to the broader literature on social justice actions, while also opening new areas of inquiry.
First, our findings expand on previous research, indicating that individual differences matter in collective action (Kay & Jost, 2003;Sidanius & Pratto, 2001;Snyder et al., 2008;Thomas et al., 2022).We found that the judgments most tied to evaluating causes of discrimination and activism were more strongly related to collective action, while personality dispositions that were more focused at the individual level were more related to everyday social justice actions.These results could be examined further in tests of models used to predict engagement in collective action (e.g., MOBILISE, Thomas et al., 2022).Similarly, microaggression researchers (e.g., Sue & Spanierman, 2020;Williams et al., 2021) could explore whether different forms of everyday discrimination (e.g., ignoring someone vs. using hostile or stereotyping language) are related to different scores on ESJB.
As is typical of all non-activist samples, very few participants in these studies engaged in formal activism.It would be useful to test the patterns of relations among self-identified activists.Future research could explore other ways in which individuals who engage in everyday social justice behavior may differ from those who engage in collective action.
Finally, although it was not the main focus of our efforts, we conducted additional exploratory analyses (see Supplementary Materials) for Studies 3 and 4.These Supplementary Analyses further probed how our individual difference and identity measures predicted our outcome measures, compared ESJB to the other activism measures, and used moderation with Hayes' PROCESS Macro for SPSS (Hayes, 2021) to examine whether the relationship between ESJB and other outcomes was different across racial and socio-economic groups.The results, although exploratory, proved interesting and a place where future research could investigate further.

Limitations and Future Directions
As we have noted, the measure developed in this set of studies may be most useful for studying samples that are like the relatively privileged groups we collected.Future studies could further interrogate the utility of this scale with other types of samples by directly recruiting those groups and exploring other potential correlates of the ESJB scale that are relevant to the experiences of minority group members, such as group consciousness (Duncan, 1999(Duncan, , 2010)), collective self-esteem (Luhtanen & Crocker, 1992), or awareness of social identity privilege (Montgomery & Stewart, 2012), in order to better understand how individuals with different social identities than our participants might engage in these kinds of behaviors differently or for different reasons.Along these lines, future researcher could further interrogate the construct validity and reliability of our measure by conducting cross-group invariance testing (e.g., race-ethnicity; see Matsuzaka et al., 2022 for an example).Other research could assess how experiences particular to certain minority groups are correlated with our measure, to further address how it is similar and different from existing measures of collective action and social justice behaviors.For instance, past encounters with microaggressions (Sue & Spanierman, 2020), selective incivility (Cortina et al., 2013), and overt discrimination (Tropp et al., 2012) have been shown to affect levels of traditional forms of activism.It is reasonable to expect that these could also predict everyday social justice behavior in similar ways, particularly if individuals perceive discrimination against their group as pervasive, but addressable (Foster, 2009).
Finally, these studies have many common limitations, including reliance on self-report measures and the lack of ability to make causal inference.Future studies could include behavioral/observational outcomes or experimental designs to address these common methodological concerns (Podsakoff et al., 2003).For example, after self-reporting ESJB, participants could be approached with an opportunity to donate their incentive payment to a worthwhile cause.ESJB might be associated with contributing more to such causes.Experimental or longitudinal study designs would help address the lack of causal inference.In general, we hope that future research will use this measure in a variety of contexts and in conjunction with a variety of methods.

Conclusions
Despite the real limitations of our research, we offer this measure as a potential tool for studying everyday social justice behaviors in samples like ours, that is, in student and U.S. citizen samples, particularly when they are majority White.We hope that others can use our scale in other contexts and establish its utility in a wider range of samples.Increasing the number of social justice actions occurring in our everyday interactions is worth pursuing, as is understanding the relationship between this sort of activity and other more directly political, activist engagements.Future efforts to research (and promote) these small, everyday actions toward social justice have the potential to create real progress toward justice, since "Small acts, when multiplied by millions of people, " (Zinn, 2007) can truly change the world.
students-overrepresented among those who engage in social justice activism of any kind-recruited via the undergraduate Introduction to Psychology subject pool and graduate student email lists of a large university in the Midwest region of the United States.Details about the specific demographics are included in the Supplementary Materials.

Study 3 :
Validation of the Everyday Social Justice Behavior (ESJB) Scale in a Student Sample Method Participants Participants were a new sample of 333 college students recruited via the undergraduate subject pool of an introductory psychology courses at a large university in the Midwest region of the United States.
Closure -Need for Cognitive Closure was measured using 15 items fromRoets and van Hiel (2011).Sample items include: "I dislike questions which could be answered in many different ways" and "I feel irritated when one person disagrees with what everyone else in a group believes."

Table 2
Frequency of Everyday Social Justice Behaviors -17 Items

Table 3
Bivariate Correlations of Variables, Study 3

Table 5
Bivariate Correlations of Variables, Study 4