Authors
Matthew Grindal
Department of Sociology and Anthropology
University of Idaho
|
Melanie Kushida, Tanya Nieri, and Jennifer Rodriguez-Trujillo
Department of Sociology
University of California, Riverside
|
Abstract
Past work in identity theory has examined the role of resources in the identity verification process. However, much of this work has only examined role identities even though some of these resources are intrinsically tied to social identities (e.g., status). In the current study, we examined how resources influence the verification dynamics of three social identities (ethnic, gender, and sexual orientation), and thus indirectly influence self-esteem. Drawing on survey data from a diverse college student sample (N = 554), we tested the influence of three resources implicated by prior identity research as possibly relevant to the verification of social identities: status, social bonds, and private regard. Consistent with our hypotheses, people with stronger social bonds and greater private regard were more likely to experience identity verification and thus enhanced self-esteem. However, a similar indirect effect was not found for status, which was not associated with identity verification. This research extends the understanding of resources in identity theory and provides theoretical elaboration to the relationships of private regard and social bonds with self-esteem by establishing the causal mechanism of identity verification.
|
Introduction
|
Within the identity theory tradition, one of the strongest predictors of positive self-esteem is identity verification, or the perception that one’s own self-image in an identity is congruent with how other people see him/her (Burke and Stets 2009). Given the central motivating drive for identity verification, identity theorists have examined how people draw upon personal, interpersonal, and structural resources to aid in the verification process (Burke 2008; Stets and Cast 2007). This work has typically found that resources enable people to better verify their identity.
|
|
Most of the prior work studying the role of resources in the verification process has examined role identities, with little prior work looking at social identities (e.g., ethnic identity, gender identity, sexual orientation identity). Unlike role identities (e.g., teacher and student), which are structurally anchored to the positions one occupies in society (Burke and Stets 2009), social identities are tied to the social groups to which people belong (e.g., ethnicity and religion). Many of these social groups are implicitly ranked within a broader social dominance hierarchy (Sidanius and Pratto 2001), where one may be viewed positively or negatively based on their particular social identity as different groups struggle for power. These distinct social dynamics of social identities may influence the types of resources available to verify a social identity as well as how those resources operate in the verification process.
|
|
Rooted in social identity theory’s assumption of a self-enhancement motivation where people strive to see themselves positively (Tajfel and Turner 1986), prior work in the developmental psychology literature (Phinney 1989; 1990; 1992; Rosario, Schrimshaw, and Hunter 2008) has called attention to this hierarchical component of social identities. This research indicates that through a process of social exploration, people develop strong and stable social identities, which may buffer against the harmful effect negative evaluations such as discrimination and prejudice have on one’s self-esteem (Phinney 1989). Two central dimensions of a strong social identity, which are often viewed as outcomes of this exploration process, are positive feelings for one’s social identity (i.e., private regard) and strong social bonds with other members of one’s social identity (Ashmore, Deaux, McLaughlin-Volpe 2004; Cameron 2004; Sellers et al. 1998). While prior research indicates that strong social bonds and private regard directly enhance psychological health outcomes (Kiecolt, Momplaisir, and Hughes 2016), no known work has examined how these dimensions of social identity might operate as resources in the identity verification process, and thus indirectly enhance psychological health and self-esteem.
|
|
Drawing on a diverse college student sample, we examine how three resources informed by this prior literature (status, social bonds, and private regard) directly enhance self-esteem. Further, we examine how they potentially operate as resources that help people navigate the verification process of three social identities, which are marked by status boundaries and the potential for ingroup/outgroup hostility (ethnic identity, gender identity, and sexual orientation identity). By integrating theoretical insights from the social identity and identity theory literatures, this study extends identity theory by developing a greater understanding of how resources influence the verification of social identities. Further, it extends the social identity literature by providing theoretical elaboration, through a consideration of identity verification as a causal mechanism, of why social bonds and private regard enhance psychological health, and specifically self-esteem.
|
THEORY
Identity Theory, Identity Verification, and Resources
|
Identity theory examines how people’s identities influence their behaviors, thoughts, and emotions (Burke and Stets 2009). Identities are conceptualized as conventional and idiosyncratic sets of meanings that define how people see themselves when they occupy a structural position such as a parent or worker (i.e., role identity), are a member of a social group such as one’s religious group or ethnic group (i.e., social identity), or when they identify with certain personal characteristics that are unique to them such as being intelligent (i.e., person identity).
|
|
For example, imagine an African American person who defines their ethnic identity in terms of being proud and aware of their ethnic group heritage as well as engaging in behaviors reflecting similar cultural awareness. After canceling a long-planned engagement to attend a cultural festival promoting awareness and knowledge of African American culture, this person receives reflected appraisals from other people questioning their ethnic pride and the degree of commitment to issues relevant to African Americans. These discrepant reflected appraisals would lead to non-verification of their ethnic identity, which would generate negative emotions. They would then be motivated to adjust their behavior, possibly by assuring the other people of their ethnic pride and providing them a reassurance to attend future events, which would allow them to obtain verifying reflected appraisals.
|
|
Given the primacy of the verification motivation within the identity theory framework, identity scholars have studied the resources people call upon to assist them in the verification process (Burke 2008; Cast 2003; Stets and Cast 2007). Resources have been studied at the personal level (e.g., self-esteem), the interpersonal level (e.g., role-taking and trust), and the structural level (e.g., status) (Stets 2018; Stets and Cast 2007). Resources help people better control situational meanings and obtain verifying reflected appraisals (Cast 2003; Stets and Harrod 2004). In addition, they provide resilience, which mitigates the harmful impact that non-verification has on adverse psychological outcomes (e.g., negative emotions) (Burke 2008). Most prior work examining the role of resources has studied them in relation to role identities. However, the nature of social identities suggests that they may possess unique sets of resources deserving of study.
|
Social Identities and Resources
|
Social identities are those identities that tie us to social groups (Burke and Stets 2009), including but not limited to one’s ethnicity, gender, religion, and sexual orientation identity. Unlike role identities, which typically complement each other in social structure (e.g., parent and child, teacher and student) (Burke and Stets 2009), membership in a social group often engenders an ingroup/outgroup mentality, where people who do not belong to one’s group are viewed less favorably (Tajfel and Turner 1986). If outgroups are perceived as threatening, these less favorable attitudes might develop into feelings of hostility (Brewer 1999), which are then accentuated when different groups compete for resources and power within a social dominance hierarchy (Blumer 1958; Sidanius and Pratto 2001). In competition for social power, these groups will often develop beliefs and attitudes that are favorable of their ingroup and unfavorable of outgroups to either justify or displace current power arrangements between groups. Because of these dynamics, social identities may be defined differently depending on whether someone is in the ingroup or outgroup. This differs from role identities, which have meanings that are more likely to be consensually shared across all society (e.g., friends being supportive, mothers being nurturing). These unique social elements of social identities have implications for the types of resources available to people when they seek verification of these identities, which we discuss below.
|
|
Prior work in the social identity and developmental literature has noted that social identities develop through an interactive process (Phinney 1989). Rooted in the developmental work of Erikson and Marcia (Erikson 1968; Marica 1980), people who belong to some social groups may experience an identity crisis when they first encounter negative stereotypes and treatment due to their social group membership (Cross 1991). This sparks an exploration process wherein they interact with other social group members to develop a greater understanding and knowledge of their social identity. This leads to identity achievement, or a stable commitment to their social identity and social group membership, which is buttressed by this understanding. The achievement process can be understood as acquiring the cultural and social capital necessary to confront the negative meanings ascribed to one’s social group and replacing them with positive meanings.
|
|
A consequence of an achieved identity is that along with the increased commitment to their social group, people develop stronger social bonds to their social group and hold more positive feelings about their social group membership (i.e., private regard) (Phinney and Ong 2007; Sellers et al. 1998), which has positive implications for enhanced well-being. Past research employing this developmental model has found that an achieved social identity contributes to improved psychological health outcomes, including heightened self-esteem. Some of this work has measured this developmental process directly (Phinney and Chavira 1992; Rosario, Schrimshaw, and Hunter 2008), whereas other work has established direct links of social bonds and positive feelings for one’s social group with improved psychological health (Crocker et al. 1994; Kiecolt, Momplaisir, and Hughes 2016).
|
|
In addition to social bonds and private regard directly enhancing psychological health, these outcomes of an achieved identity might also act as interpersonal and personal resources, respectively, which people use to help verify their social identities. Social bonds to other group members provide people with a reliable source of social support (McCall and Simmons 1978), which prior work in identity theory has suggested is comparable to verification (Burke and Stets 2009). In the face of negative treatment from outgroup members, strong bonds to one’s social group can offer a consistent source of identity verification. Private regard may also operate as a resource. While this construct has not been examined as a resource in identity theory, its global equivalent of self-esteem has been found to operate as a resource that helps increase identity verification (Cast and Burke 2002; Stets and Cast 2007). Stets and Cast (2007) argue that high self-esteem (i.e., self-worth and self-efficacy) provides people with resilience to overcome the day-to-day difficulties of verifying their identities. A similar argument can be made for private regard as a resource in verifying a social identity. The internalized belief that one is a good person because of their social identity might be a resource when helping people overcome identity discrepancies in everyday interactions that specifically pertain to negative evaluations of their social identity membership (i.e., prejudice).
|
|
We also suggest that status may operate as a structural resource when helping people verify their social identities. Status is one of the more frequently studied resources in identity theory (Burke 2008; Stets and Harrod 2004; Stets and Cast 2007). High status positions and identities carry with them expectations about access to resources and knowledge of resources (Burke 2008), which develop into beliefs people hold about those who occupy high status positions and identities (Berger and Webster 2018). Compared to those in low status positions and identities, high status people are often assumed to be more intelligent, capable, and worthy. This allows high status people to exert more influence in interactions and obtain verifying reflected appraisals (Cast 2003). Based on this past work, we hypothesize:
|
• H1: Status is positively associated with identity verification.
• H2: Social bonds are positively associated with identity verification.
• H3: Private regard is positively associated with identity verification.
|
|
Last, given prior work’s support for the link between identity verification and self-esteem (Cast and Burke 2002), we hypothesize:
|
|
• H4: Identity verification is positively associated with self-esteem.
|
METHODS
|
The data for this study come from a cross-sectional survey examining social identity dynamics, which was administered to students at a diverse Southwestern university. The university has an ethnic breakdown of approximately 40% Latino, 40% Asian, 10% white, 5% black, and 5% identifying with another ethnic group or with multiple ethnic groups. The survey was administered online via surveygizmo.com in 2017 to students in five sociology classes. After an in-class announcement was made to the students, students were sent a link where they could fill out the survey if they chose to participate. In four of the five classes, survey participation was incentivized with course extra credit. In the remaining course, participation was incentivized with entrance into a raffle to win one of three $50 Amazon gift cards. Six-hundred and seventy-six students participated in the survey.
|
Sample
|
Of the students who participated in the survey, 554 provided complete data and were thus included in the operational sample. The sample was diverse in terms of the three social identities examined in this study. The ethnic breakdown of the sample was 56% Latino, 20% Asian, 10% white, 7% black, and 7% identifying with another or multiple ethnic groups. The gender breakdown of the sample was 77% female, 22% male, and 1% identifying as another gender. In terms of sexual orientation identity, 86% identified as straight/heterosexual, 4% identified as gay or lesbian, 7% identified as bisexual, and 3% identified with another sexual orientation. The respondents’ mean age was 20.97 (SD = 2.67), their median level of parental education was high school, and the median level of parental income was $37, 500.
|
Measures
|
Self-Esteem: We adopted self-esteem as our latent measure of well-being given its prevalence in both the social identity and identity theory literatures (e.g., Cast and Burke 2002; Stets and Burke 2014; Umana-Taylor, Diversi, and Fine 2002). We measured self-esteem with three observed scales, reflecting a three-factor conceptualization developed in identity theory research: self-worth, self-efficacy, and self-authenticity (Stets and Burke 2014). Self-worth refers to the strength of the positive feelings people hold about themselves; self-efficacy refers to the extent that people feel they can control their environment; self-authenticity refers to the extent that people’s expressions and behaviors reflect who they think they really are. The three scales are adapted from prior measures (see Stets and Burke 2014 for a full description). The self-worth scale was measured with seven items (e.g., “I feel that I am a person of worth, at least on an equal plane with others.”) and showed strong internal consistency (α= .94). The self-efficacy scale was measured with seven items (e.g., “I certainly feel hopeless at times.” – reverse-coded) and showed strong internal consistency (α= .85). The self-authenticity scale was measured with seven items (e.g., “I find that I can almost be myself.”) and showed strong internal consistency (α= .80).
|
|
Social Identity Dimensions: For each of the three identities, we measured the dimensions of social bonds, private regard, and verification, with nine overall measures. The dimensions of social bonds and private regard were measured with previously validated scales adapted from the social identity theory literature (Cameron 2004). The language in the social bonds and private regard scales were adapted to fit the particular identity. The social bonds scale contained six items assessing the respondents’ degree of attachment and connectedness to other members sharing their social identity (ethnicity, gender, sexual orientation) (“When I am around other members of my [social identity], I feel that I really belong.”). The items had six response options ranging from strongly disagree (0) to strongly agree (5). These items were averaged with a range of 0-5 creating social bonds scales for the ethnic identity (α =.91), gender identity (α = .88), and sexual orientation identity (α = .88). The private regard scale contained five items assessing the respondents’ positive or negative evaluation of their social identity (ethnicity, gender, sexual orientation) (“In general, I am glad to be someone with my [social identity].”). The items had six response options ranging from strongly disagree (0) to strongly agree (5). These items were averaged with a range of 0-5 creating private regard scales for the ethnic identity (α =.87), gender identity (α = .83), and sexual orientation identity (α = .82).
|
|
The dimension of identity verification was measured with one question for each of the three identities. The item is adapted from the identity module of the 2014 GSS data set (Smith et al. 2018). Respondents were first asked to think about how they see themselves as a member of their [social identity]. They were then asked, thinking about how they see themselves as a member of their [social identity], how much their friends see them in the same way that they see themselves. There were ten response options ranging from “not at all” (1) to “completely” (10), ranging from 1-10 with higher scores indicating higher levels of identity verification.
|
|
Status: Status was measured with three items assessing the respondents’ self-identification of their ethnicity, gender, and sexual orientation. Respondents were provided with open-ended and close-ended questions to self-report their ethnic, gender, and sexual orientation identification. For each identity, we then dichotomized the responses into two categories: high status (1) and low status (0). For ethnicity, if the respondents identified as white, they were coded as high status, while all other ethnicities were coded as low status. For gender, if respondents identified as a man, they were coded as high status, while other genders were coded as low status. For sexual identity, if respondents identified as straight/heterosexual, they were coded as high status, while other sexual identities were coded as low status. These dichotomized items were then summed creating an index ranging from 0-3 with a higher score indicating greater status.
|
ANALYTICAL STRATEGY
|
We tested the validity of the theoretical model in two steps employing the structural equation model module in Stata 15. Using maximum likelihood estimation, we first ran a confirmatory factor analysis to test the validity of the measurement model, specifically the latent constructs of self-esteem and the three social identity dimensions. We then added the observed measure of status and the pathways between the latent constructs, which enabled us to test the four hypotheses. We tested the goodness of fit of the models with maximum likelihood chi-square and the root mean square error of approximation (RMSEA). Given the large sample size, the maximum likelihood chi-square is a less reliable measure of goodness of fit. The RMSEA provides a more parsimonious measure, which takes into account the complexity of the model and sample size. Last we report standardized path coefficients to report the magnitude and significance of the hypothesized effects.
|
|
The descriptive statistics and correlations of the observed variables are displayed in Table 1. For the multivariate analyses, we first ran a confirmatory factor analysis (CFA) with maximum likelihood estimation to test the measurement validity of the four latent constructs: the dependent variable of self-esteem as well as the identity dimension of social bonds, private regard, and identity verification. We specified three observed indicators for the latent construct of social bonds (ethnic identity bonds, gender identity bonds, and sexual orientation identity bonds), three observed indicators for the latent construct of private regard (ethnic identity private regard, gender identity private regard, and sexual orientation identity private regard), three observed indicators for identity verification (ethnic identity verification, gender identity verification, and sexual orientation identity verification), and three indicators for self-esteem (self-worth, self-efficacy, and self-authenticity).
|
|
We anticipated that the error terms of the observed indicators of the different identity dimensions (bonds, private regard, and verification) would be correlated by type of identity. We ran a preliminary measurement model without these error terms correlated, and the model was a poor fit to the data (χ² = 561.77 (df = 48), p < .001, RMSEA = .139). We examined the modification indices for error terms of different identity dimensions that were correlated because they shared a common identity (e.g., gender verification and gender private regard). We iteratively correlated these error terms where they significantly improved model fit. This re-specified model fit the data well (χ² = 91.32 (df = 42), p < .001, RMSEA = .042). We then added the observed measure of status, specified the pathways enabling us to test the hypotheses, and reran the analyses of the full structural equation model.
|
|
The test of the full structural equation model was an adequate fit to the data (χ² = 168.96 (df = 50), p < .001, RMSEA = .064). Table 2 reports the standardized coefficients of the measurement model, which were moderate to strong (λ > .40). Table 3 reports the standardized coefficients of the structural model. Model 1 shows the path coefficients for status, social bonds, and private regard on identity verification. We hypothesized that status, social bonds, and private regard should operate as resources that enable people to increase their levels of identity verification. Contrary to the first hypothesis, status was not associated with identity verification. However, the other two hypotheses were supported. Stronger social bonds were positively associated with identity verification (β = .41, p <.001) as was stronger private regard (β = .30, p <.01). Model 2 examines the direct effects of status and the identity dimensions on self-esteem. Identity verification was associated with increased self-esteem (β = .26, p <.001) providing support for our fourth hypothesis. In addition, status (β = .14, p <.01) and private regard (β = .21, p <.01) were both positively associated with self-esteem, while social bonds was not associated with self-esteem. Taken collectively, these results suggest that the direct effect of status on self-esteem was not mediated by identity verification, while the direct effect of social bonds on self-esteem was fully mediated by identity verification, and the direct effect of private regard on self-esteem was partially mediated by identity verification.
|
|
Given the relatively large percentage of missing cases in our sample, we ran a robustness check by testing the full structural equation with maximum likelihood estimating the missing values, thus including everyone in the sample who responded to the status question (N = 658). The goodness of fit statistics were comparable to the initial model (χ² = 205.27 (df = 50), p < .01, RMSEA = .069), and the statistical significance of the findings in the structural model were replicated
|
DISCUSSION
|
This study examined the role of resources in the identity verification process as they pertain to social identities. We suggested that the ingroup/outgroup dynamics of social identities and their potential for intergroup hostility lead to specific challenges when people work to verify these identities. Drawing on work from the identity theory and social identity theory literatures, we tested the extent to which three resources help in the verification of social identities (social bonds, private regard, and status), and thus may indirectly enhance self-esteem. Consistent with our hypotheses, we found that people with strong social bonds and positive feelings for their social identities were more likely to experience identity verification, which in turn led to enhanced self-esteem. Consistent with prior research, it appears that these resources afford people better control of situational meanings as well as greater resilience when working to overcome non-verifying feedback (Stets and Cast 2007), both of which increase identity verification, and thus enhance self-esteem.
|
|
While higher status people did have greater self-esteem, contrary to prior identity theory research (Cast 2003), status was not positively associated with identity verification, thus suggesting that status may not operate as a resource to help people verify social identities. This finding could be an artifact of how identity verification was measured in our study. The reference group specified in our identity verification measure was friends. Prior research indicates that friendships tend to exhibit a homophily bias and have significantly fewer status distinctions (MacPherson and Smith-Lovin 1987). With fewer status distinctions, status may not be easy to leverage in securing identity verification among friends. It is possible that with the use of a different reference group in the verification measure, status would be positively associated with identity verification.
|
|
This study extends identity theory by examining the verification dynamics that are specific to social identities, and how these dynamics are influenced by resources specific to social identities. Our research suggests that private regard and social bonds operate as resources that help in the verification of social identities. The resources of social bonds and private regard are similar to concepts previously studied by identity theorists. We consider each of these below and how our findings can inform the study of these concepts in future work.
|
|
Social bonds is similar to the concept of emotional attachment, which has been studied in prior identity theory research on spouses (Burke and Stets 1999). In this prior study, emotional attachment was measured as the emotional closeness between two spouses. Identity verification was found to indirectly enhance the emotional attachment between spouses by fostering greater feelings of trust. These findings combined with our results suggest that the relationship between identity verification and social bonds may be reciprocal. Greater social bonds may not only be an outcome of the verification process, but also a resource that helps people verify their social identities. Stronger ties to other social group members may provide a constant network of individuals who provide support and identity verification. Relatedly, private regard is similar to the concept of specific self-esteem (Rosenberg et al. 1995) and collective self-esteem (Crocker and Luhatenen 1990), the self-esteem tied to a specific identity, which prior work in identity theory has studied in relation to global self-esteem (Ervin and Stryker 2001). We suggest that identity theory scholars should also theoretically disentangle the relationship between private regard and global self-esteem when understanding their role as resources and outcomes in the identity verification process.
|
|
Implied in these two parallels, it is worth noting that prior identity theory research has documented the reciprocal relationship between resources, identity verification, and the affective outcomes of identity verification (Stets 2018; Stets and Cast 2007). For instance, Cast and Burke (2002) found that identity verification increased self-esteem, but also that self-esteem was then used as a resource to buffer against the harmful consequences of non-verification. That is, not only do resources help people obtain identity verification, but successful identity verification increases access to resources (Stets and Cast 2007). Given the cross-sectional nature of our data, we were unable to examine the reciprocal nature of these relationships. Future research using longitudinal data could extend our results by examining the extent to which the verification of social identities increases private regard, social bondss. With respect to self-esteem, it would also be instructive to examine how global self-esteem operates alongside specific self-esteem (private regard) to influence identity verification. Private regard might be a more effective resource at securing identity verification given the greater relevance that the specific identity meanings have to the non-verifying reflected appraisals.
|
|
Our findings also help to elaborate on prior research examining how achieved social identities, marked by social bonds and high private regard, influence psychological health, and in particular self-esteem. The literature is replete with studies indicating how these dimensions of social identities (as well as achieved and strongly held social identities generally) positively influence psychological health (Crocker et al. 1994; Kiang et al. 2006; Kiecolt, Momplaisir, and Hughes 2016; Martinez and Dukes 1997; Phinney and Chavira 1992; Rosario, Schrimshaw, and Hunter 2011). Our findings mirror this past work. Private regard and social bonds to one’s social group were both directly associated with enhanced self-esteem. However, we add to this prior work by noting identity verification as a potential causal mechanism in this relationship. People with strong social bonds to their social group and greater levels of private regard experienced enhanced self-esteem, completely or in part, because they were more likely to have their identities verified. The dimensions of social bonds and private regard appear especially essential to members of low status social groups who may not be able to benefit from status as a resource.
|
|
In closing, this study integrated and extended identity theory and the social identity literature to provide insight into how people maintain positive self-esteem while holding social identities, which are distinctively marked by ingroup/outgroup dynamics and the possibility of outgroup hostility. Social bonds and private regard enhance psychological health, and this occurs in part because people use them as resources to secure identity verification.
|
REFERENCES
|
