Abstract
Many employees possess inconspicuous identities that are stigmatized. At work, a context wherein impression management concerns are salient, these individuals face decisions about when, how, and to whom to disclose their concealable stigmas with important consequences for the way individuals experience work. In the following review, we integrate findings from psychological, sociological, and management literatures pertaining to the management of concealable stigmas. We further push the boundaries of this evidence to develop a multilevel model of workplace identity management behavior. That is, we conceptualize identity management as a within- and between-person phenomenon, accounting for the notion that identity management behaviors will vary as a function of the situation but that people will also exhibit identity management behavioral averages, tendencies, and accumulation that facilitate meaningful comparison among employees. Throughout the review, we highlight common themes, clarify inconsistent findings, and call attention to several fruitful areas we see ripe for future research.
Some attributes are so noxious that their bearers are denigrated, devalued, and detested. In addition to visible characteristics that give rise to stigmatization such as race and gender, unobservable attributes including values and beliefs, sexual orientation, religious affiliation, and personal experiences can also be stigmatized (Goffman, 1963). Unlike individuals with visible stigmas whose main goal is to attenuate interpersonal tension in social interactions, people who possess concealable stigmas have the option of hiding that part of themselves in order to avoid interpersonal derogation, discrimination, or other negative repercussions. That is, the main goal of people with concealable stigmas is to manage stigma-related information—to make decisions about when, how, where, and to whom to disclose their concealable identities (Goffman, 1963). Often referred to as “identity management,” research examining the antecedents, specific behaviors, and consequences associated with this phenomenon has flourished in recent decades.
However, much of the existing research in this area originates in a broader social psychological perspective, meaning considerably less is known about how concealable identity management manifests in workplace settings and the resulting implications for employees and organizations. Thus, our review serves as a critical first step in creating a bridge that allows management researchers to map management constructs onto existing social psychological research on concealable stigma management. We contend that deeper consideration of concealable identity management as it emerges in the context of the workplace is critical for four reasons. First, the current body of empirical research suggests identity management decisions are often psychologically and physiologically destructive, an implication that is relevant to both organizations and their employees. Indeed, disclosure decision-making processes can result in deleterious repercussions via psychological strain, emotional stresses, and stress-related illnesses (Cole, Kemeny, Taylor, & Visscher, 1996; Frable, Platt, & Hoey, 1998; Major & Gramzow, 1999; Pachankis, 2007; Smart & Wegner, 1999, 2000).
Second, the workplace represents a context wherein all employees’ experiences are typified by concerns about interpersonal impressions (see Roberts, 2005). Thus, the linkages of interest here may be even more impactful in workplace than nonwork settings given the salience of impression management concerns for employees with concealable stigmas who must manage the impressions they make on coworkers, clients, subordinates, and supervisors while believing that those same colleagues could reject them if made aware of their hidden identity. Furthermore, these concerns are likely exacerbated in organizational settings given the wide range of high-stakes consequences that could potentially result from a negative interaction with another person at work, including outcomes related to performance evaluations, compensation, and/or promotion decisions. Indeed, it is for this reason existing social psychological literature is not sufficient to guide future research in organizational contexts, and it is our intention to provide a framework from which management scholars may begin to build understanding of the identity management phenomenon as it emerges in workplace settings.
Third, although the specific characteristics that are stigmatized may vary across time and culture, the need to manage interpersonal elements of concealable stigma is constant. A large but immeasurable proportion of employees bear some kind of concealable stigma or work alongside someone who does, whether it be a past experience (e.g., childhood abuse, rape/sexual assault) or an employee’s family background (e.g., child of same-sex parents).
Fourth and finally, there remains a great deal of confusion in current literature surrounding both the conceptualization and measurement of identity management. For example, some research focuses on disclosure as a one-time, dichotomous decision (e.g., “At work, have you disclosed your sexual orientation to (1) no one (2) some people (3) most people (4) everyone”; Ragins & Cornwell, 2001). Measures such as these do not capture the multifaceted process of concealable stigma management, failing to consider ongoing decisions employees must make about concealing, signaling, and revealing their stigmas. Instead, these types of measures assume the process is a simple one-time decision to disclose. In contrast, other studies consider identity management as a more nuanced process involving momentary choices to conceal, reveal, or signal a concealable attribute (Button, 1996, 2001; Chaudoir & Fisher, 2010; Clair, Beatty, & MacLean, 2005), which more adequately captures concealable stigma management as a complex, enduring phenomenon.
It follows that the primary goal of the current article is to review, integrate, and extend existing evidence into a novel, theoretical multilevel model (see Figure 1) that can drive management scholarship in previously unexplored directions. In doing so, we draw largely from highly relevant nonmanagement studies that inform the management literature (while highlighting the areas in which management literature is sparse). Thus, it is not our intent to provide a comprehensive review of the disclosure literature in social and clinical psychology but instead a selective review of articles from those disciplines that may inform our understanding of workplace dynamics in ways that management literature has yet to achieve. We accomplish this goal by first briefly reviewing the theoretical foundations underlying current understandings of concealable stigmas before introducing our novel theoretical model.
A Multilevel Model of Workplace Concealable Stigma Management
Stigma Identity Management Theory
As described in the opening paragraphs, a stigma is an attribute that is devalued in a particular context (Crocker, Major, & Steele, 1998; Goffman, 1963). The concealability of one’s stigma is a central determinant of the nature of the bearer’s experience. Whereas people with observable stigmas cope with the explicit knowledge that others are aware of their identity, people with concealable stigmas have to decide whether, how, when, where, and to whom they should conceal, reveal, or signal their identities. Furthermore, certain stigmas exist on a continuum of concealability wherein they gradually progress from easily hidden to completely conspicuous. Stigmas that change in some way over time have been described as taking on a specific course (E. E. Jones et al., 1984), which undoubtedly plays a critical role in determining concealing, revealing, and signaling decisions over time. One reason for the central importance of these decisions is opposing psychological forces that are more complex than the overarching motives of belongingness and self-enhancement inherent in impression management more generally (Clair et al., 2005).
Motivational Bases for Stigma Management
On the one hand, individuals are motivated to be authentic in their interactions in order to maintain and verify their sense of self (Friskopp & Silverstein, 1996) and build open relationships with others (Creed & Scully, 2000). Furthermore, revealing a stigmatized identity may be necessary to gain benefits afforded to members of a particular social identity group (e.g., same-sex partner benefits, disability accommodations). On the other hand, concerns about how one is perceived by others are particularly salient in workplace contexts where most employees attempt to make positive impressions on supervisors, coworkers, and subordinates (Roberts, 2005). Moreover, the likelihood of experiencing discrimination may increase when a stigmatized identity is confirmed (e.g., Griffith & Hebl, 2002; King & Botsford, 2009). Workers with concealable stigmas must weigh these opposing forces, conscious or not, in deciding how to negotiate the “disclosure dilemmas” presented by each new social situation (Ragins & Cornwell, 2001).
Existing Models of Stigma Management
Two theoretical models of relevance to concealable stigma management have emerged. In the first, Pachankis (2007) developed a process model highlighting the psychological impact of identity concealment, which served as a counterpoint to dominant assumptions that individuals with concealable stigmas would suffer fewer consequences than those with visible stigmas. Indeed, Pachankis highlighted unique aspects of concealable identities—such as the difficulty of identifying similar or supportive others—that might exacerbate the consequences of stigmatization. This includes cognitive outcomes such as preoccupation and vigilance, affective implications such as anxiety and shame, and behavioral implications including impression management.
In the second model, Chaudoir and Fisher (2010) described a disclosure process driven by approach- and avoidance-focused goals that influence a decision to disclose a concealable stigmatized identity. The basic idea is that people can choose whether or not to disclose a stigmatized identity to pursue positive outcomes (e.g., greater intimacy, acceptance, understanding) or to avoid negative outcomes (e.g., social distancing, anxiety). Another important feature of this disclosure model is a feedback loop wherein a single disclosure event can affect later disclosure decisions via “upward spirals toward visibility” or “downward spirals toward concealment.” This implies that a decision to disclose in one situation (and how disclosure is received) likely affects a decision to disclose in a later situation. However, the paper does not go on to identify within-person and between-person sources of variability in these experiences; rather, it serves as a jumping off point for the current model.
It is important to note that both models are limited by their relevance to social rather than work situations. This decision inherently prohibits direct consideration of the unique context of the workplace; work creates a number of conditions that are critical for understanding concealable stigma management. First, the American Time Use study showed that most working Americans spend more time at work than they do with their families (U.S. Bureau of Labor Statistics, 2013). This underscores the importance of the topic from a sheer frequency perspective but, more importantly, clarifies why it is important for people to form strong interpersonal relationships with others at work to fulfill their social belongingness needs. Second, the quality of these relationships—which we argue is influenced by identity management behaviors—affects meaningful organizational outcomes such as turnover and performance. Third, the evaluative nature of work, wherein contributions and behaviors are continuously judged, creates a context wherein impression management concerns are salient for all workers (Roberts, 2005). Fourth, there are constructs specific to the workplace—such as perceived organizational and supervisor support, job attitudes and performance, explored below— that have not been considered in these models. Taken together, these factors point to the importance of the workplace as a context within which to study concealable stigma management.
It is also noteworthy that both models take a situational lens to disclosure; that is, the antecedents, consequences, and mechanisms of interest are relevant to events. Yet, virtually all of the existing research summarized in the next section is concerned with chronic experiences rather than specific events. In addition, both models treat disclosure as a “yes” or “no” decision rather than part of a broader set of identity management behaviors ranging from concealing to revealing. The same is true of many empirical models, such as Ragins and Cornwell’s (2001) model of the predictors of disclosure of a gay or lesbian sexual orientation. Though Goffman (1963) identified a variety of identity management strategies in his influential writing on stigma, the full range of strategies has only begun to be incorporated in research.
Concealable Identity Management Strategies
Growing recognition that identity management consists of more than a dichotomous decision of whether or not to disclose is evidenced by recent efforts to assess different types of management strategies (e.g., Anderson, Croteau, Chung, & Distefano, 2001; Button, 2001; Creed & Scully, 2000). For example, Woods (1994) conceptualized identity management for gay and lesbian workers as consisting of three overarching strategies: counterfeiting (i.e., fabricating a false heterosexual identity), avoiding (i.e., eluding questions about personal life), or integrating (i.e., revealing an LGB identity). This formulation emerged from Woods’s (1994) own qualitative work and has since received support from quantitative studies of workplace identity management (Button, 2001). These ideas can be contrasted with Clair and colleagues’ (2005) theoretical classification of a range of identity management behaviors as either “revealing” or “concealing.”
Research on the identity management experiences of pregnant (K. P. Jones et al., in press) and LGB workers (King et al., 2013) has attempted to integrate previous typologies of stigma management; this evidence suggests that identity management strategies involve not only behaviors designed to reveal or conceal one’s stigma but also behaviors that involve information seeking through “testing the waters” or signaling a potential stigma.
Signaling strategies are unique from revealing and concealing strategies in that these indirect, strategic behaviors allow individuals to gauge the confidant’s reaction before actually disclosing and to “backtrack” if it becomes apparent that negative reactions to disclosure are probable. Indeed, signaling strategies entail “[straddling] the line between going public and remaining private about one’s invisible social identity” (Clair et al., 2005, p. 83). However, the indirect hints inherent in signaling may “invite speculation . . . encouraging [others] to read between the lines” (Woods, 1994: 176). Examples of signaling behaviors include dropping hints, providing clues, and sending implicit messages (Clair et al., 2005).
To date, research following existing theories (e.g., Chaudoir & Fisher, 2010) has been limited almost exclusively to global assessments of identity management wherein individuals indicate the extent to which they generally engage in either revealing or concealing strategies (e.g., Anderson et al., 2001; Button, 2001) or have or have not disclosed their identity (Ragins & Cornwell, 2001). Such studies have offered valuable findings about stigma management but should be further supplemented with more nuanced understandings of identity management. Moreover, surprisingly little research has examined the specific interactions in which stigma-related dynamics are expressed (Hebl & Dovidio, 2005). We review and extend this work in the context of an integrative multilevel model of concealable stigma management.
Existing Evidence and Integrative Theoretical Model
Our review of extant identity management research will highlight common themes, describe new evidence to attempt to clarify when particular strategies have positive versus negative outcomes, and call attention to several fruitful areas we see ripe for future research. Based on our integration of extant identity management literature, we advance a theoretical model that conceptualizes identity management as a phenomenon that emerges within discrete disclosure interactions and can give rise to more chronic behavioral averages, behavioral tendencies, and behavioral accumulation. That is, we propose a model that conceptualizes identity management as a within- and between-person phenomenon, meaning that identity management varies as a function of situational characteristics (i.e., within-person) but that people will also manage their concealable stigmas in a certain way on average (i.e., between-person).
Drawing from Chan’s (1998) foundational work delineating a typology of composition models, we define our multilevel model of identity management as a process model wherein the specified impact of identity management at the lower level is analogous to the impact of identity management at the higher level. This is similar to Klein and Kozlowski’s (2000) notion of homologous multilevel models, wherein relationships between constructs hold across multiple levels of analysis. In such models, higher-level constructs do not necessarily simply reflect the means of the lower-level constructs (as is the case in a purely additive model), but the higher-level constructs account for an element of accumulation over time. For instance, Chan (1998) provides the example of team proceduralization (a higher-level construct reflecting the lower-level individual process of proceduralization of task acquisition). The basic notion underlying this example is that the process of proceduralization changes over time as team members gain experience, practice, and routinize procedures. Thus, the specified process accumulates over time and changes in substantively meaningful ways that would not be uncovered by simply aggregating from a lower- to create a higher-level construct.
Consequently, the process of identity management specified at the between-person level reflects three primary components: behavioral averages, behavioral tendencies, and behavioral accumulation. The first component is additive, reflecting an employee’s average identity management experiences aggregated from the lower situational level while ignoring within-person “agreement” or dispersion. The second component incorporates within-person agreement, accounting for the notion that dispersion of identity management cannot be completely attributed to situational influences. Indeed, there will be some degree of consistency within employees in that they will tend to manage concealable stigmas similarly regardless of the situation. The third and final component of between-person identity management incorporates accumulation effects as in Chan’s (1998) process models, wherein the lower level reflects a process that accumulates over time.
Within-Person Identity Management
We begin by discussing our within-person model of identity management wherein we articulate the antecedents and consequences of identity management at the situation level, thereby considering how the precursors and implications of identity management may fluctuate over time from one interaction to the next. The underlying assumption at the within-person level of the model is that we are comparing the same employee at different time points (i.e., in different situations or interactions across time). That is, the within-person level reflects the notion that the same employee may manage a concealable stigma in different ways (and thus, experience meaningfully different outcomes) as a function of situational characteristics of a particular interaction.
We define our unit of analysis at this level as an “identity management interaction,” which reflects any time during the workday when an employee is interacting with at least one other person from work and feels she or he has to make decisions about whether to reveal or conceal her or his concealable stigma to others. An identity management interaction might occur through a variety of mediums including face-to-face interactions, phone interactions, email correspondence, and/or video conferences and depending on how the interaction unfolds, could last anywhere from a minute or two (e.g., a quick email or a brief interaction in the kitchen) to several hours (e.g., a long meeting).
Anticipated Acceptance as a Predictor of Identity Management
In the context of a specific identity management interaction, our model suggests that the extent to which an individual anticipates the person with whom he or she is interacting will be accepting of his or her concealable stigma is the primary predictor of revealing, concealing, and signaling in that particular interaction. That is, the degree to which an employee believes his or her supervisor, subordinate, coworker, and/or client will be accepting of the concealable stigma will dictate episodic identity management behaviors during workplace interactions. We argue that to the degree an individual anticipates acceptance in an interaction, the individual will engage in more revealing and less concealing. Thus, we predict a positive linear relationship between anticipated acceptance and revealing and a negative linear relationship between anticipated acceptance and concealing.
Furthermore, we expect employees with concealable stigmas to engage in particularly high levels of signaling in ambiguous situations when more information is needed to determine the relative risk associated with disclosure. High and low levels of anticipated acceptance reflect high-clarity situations in which the employee perceives acceptance will either definitely occur or definitely not occur. Therefore, situations characterized by high and low anticipated acceptance do not necessitate signaling to the same degree as moderate anticipated acceptance. Thus, we expect anticipated acceptance will have a curvilinear impact on signaling such that signaling will be peak when anticipated acceptance is moderate.
Support for this notion can be found in both management and social psychological literatures. For example, an experience sampling study of LGB employees indicated that more acceptance (and less rejection) cues perceived by individuals prior to an identity management interaction led to increased revealing (and decreased concealing) behaviors within that interaction (King et al., 2013). Indeed, these results highlighted the importance of situational cues within these types of interactions, revealing a significant amount of both within-person and between-person variability in concealing, revealing, and signaling; LGB workers’ experiences differed from interaction-to-interaction as much as from person-to-person. Furthermore, recent empirical evidence provides support for the proposed curvilinear effect of anticipated discrimination on signaling for pregnant employees. Using a longitudinal event-sampling methodology, K. P. Jones (2013) found that identity management interactions wherein a pregnant employee anticipated moderate levels of discrimination, triggered more signaling as compared interactions in which the same employee anticipated very high or very low acceptance.
Findings from social psychological research further suggest anticipated acceptance is indeed an important predictor of identity management behaviors. For example, Clark and colleagues (2003) examined perceptions of stigma, degree of disclosure, and psychological functioning at four points in time over the course of 6 years in a sample of HIV-infected African American women. Findings suggested that as perceptions of stigma increased, disclosure decreased. Similarly, Zea and colleagues (2007) found perceived positive consequences of HIV status disclosure were associated with greater disclosure, whereas perceived negative consequences of HIV status disclosure were associated with less disclosure. One study that examined reasons for delay of disclosure in 218 victims of childhood sexual abuse found fear of negative consequences of disclosure was significantly associated with delay of disclosure (Goodman-Brown et al., 2003). Furthermore, findings from a study that followed 442 women from the day after their abortion until 2 years later suggested insofar as women anticipated incurring stigma if others found out about their abortion, they tended to conceal the abortion (Major & Gramzow, 1999). Additionally, Troster (1998) found that persons with epilepsy indicated more willingness to disclose their physical stigma insofar as they anticipated more positive (and less negative) social consequences of disclosing. Finally, meta-analytic evidence also indicates an overall negative correlation between perceived stigma and HIV disclosure (Smith, Rossetto, & Peterson, 2008). Taken together, extant empirical research across several literatures supports our proposition that employees with concealable stigmas base situation-specific identity management behaviors in large part on the degree to which they expect others at work to be accepting (or unaccepting) of the stigma.
Signaling as a Precursor to Revealing and Concealing
When signaling does occur within a specific situation, the presumed purpose of the behavior is to aid individuals in making a more informed decision about how to proceed with disclosure decisions. At the within-person (i.e., situation) level of analysis, then, we argue that signaling reflects a proximal antecedent to subsequent revealing and concealing. However, as detailed above, signaling is most likely to occur in ambiguous situations (when anticipated acceptance is unclear). Thus, in high-clarity situations, employees may bypass signaling in favor of immediate revealing and concealing. However, when signaling does indeed occur, it will be exhibited immediately prior to revealing and concealing. Though empirical evidence on signaling behaviors remains scant, extant qualitative evidence supports this conceptual notion of signaling. Specifically, Black and Miles’s (2002) qualitative study of 48 African American women diagnosed with HIV suggested participants engaged in a recursive cycle of signaling, described by the authors as a “calculus of disclosure” wherein women had to constantly weigh the costs and benefits of disclosure. Participants’ motivations to reveal were primarily characterized by the need for social support, whereas their motives to conceal were fueled by fear of stigma and feelings of shame. We reason that employees with concealable stigmas proceed through a similar process, first weighing the costs and benefits of disclosure (i.e., signaling) to inform the ultimate decision to reveal or conceal (especially in low-clarity situations). Overall, these findings are consistent with prior theoretical reasoning from the management literature conceptualizing signaling as a strategy entailing a cost-benefit analysis of disclosure (Clair et al., 2005; Ragins, 2008).
Consequences of Identity Management
In the next section, we articulate how identity management behaviors within a specific situation influence proximal outcomes immediately following the interaction. In general, we expect revealing to have positive, and concealing negative, consequences. Furthermore, we expect signaling to be interpersonally beneficial (e.g., solicit helping behaviors from others, decrease perceived discrimination); however, we reason the cognitive and emotional load created by signaling will detract from important workplace outcomes including performance, job attitudes, and well-being.
To address potential skepticism a single interaction would substantially impact an individual, we provide the reader with an example. Imagine a gay employee interacting with his supervisor, with whom he has a positive working relationship and values his opinion. The employee decides to signal his sexual orientation to gauge how his boss would react if he learned of his stigmatized identity, and much to the employee’s disappointment, his boss spouts off some derogatory statements about gay men, dispelling the employee’s intentions to disclose. The employee is likely negatively impacted both immediately following the situation and perhaps days, weeks, even months later. As illustrated by our example, we contend that a single interaction undoubtedly impacts proximal outcomes of importance to employees and likely has enduring effects as well. Indeed, this notion has received support from experience sampling studies that have examined employees’ immediate reactions following from a single identity management interaction (K. P. Jones, 2013; King et al., 2013). We do not mean to imply, however, that a single identity management interaction will eliminate the within-person variability in identity management behavior in future interactions. Indeed, the outcomes associated with one interaction likely shape expectations and behaviors in future interactions; however, these outcomes only account for a piece of the variance in future interactions. As illustrated by our model, anticipated acceptance of the interaction partner is another key predictor of identity management behavior, a predictor that will inevitably fluctuate depending on the person with whom the stigmatized individual is interacting.
Task performance
Identity management behaviors likely have differential effects on employee task performance as a function of the varying levels of cognitive and emotional load imposed by each strategy. Whereas concealing and signaling require relatively high attentional resources (Clair et al., 2005), revealing behavior is less cognitively demanding. Indeed, within a specific interaction, once the concealable stigma has been revealed to a colleague, concealing and signaling with that colleague are no longer realistic options. Consequently, the question of how to manage one’s stigma becomes less relevant and, thus, less cognitively and emotionally consuming. However, if one has not revealed to a colleague, the individual may spend a great deal of attentional resources ensuring “suspicious” stigma-related information is not accidentally exposed, considering if revealing might be a good idea, and/or trying to gauge how the colleague would react to potential disclosure. For example, a gay employee who is concealing to a client must be cautious about referring to his partner in conversation or having photographs of his partner in plain sight in his office. These necessary precautions and increased attention required of concealing and signaling likely detract from an employee’s focus on the work task at hand, ultimately resulting in decreased performance.
Experimental evidence has shown that preoccupation with impression management concerns can deplete cognitive resources among stigmatized individuals and ultimately negatively impact task performance. For example, within the context of an elite university, students of low socioeconomic status (SES) experienced concerns about academic fit (Johnson, Richeson, & Finkel, 2011). Students who were instructed to address these concerns through a self-presentation task ultimately showed depleted cognitive resources on subsequent self-regulation and Stroop tasks. These findings suggest that employees with concealable stigmas may have concerns about their fit in the organization. Furthermore, when these person-organization (P-O) fit concerns are addressed through cognitively loaded identity management strategies, performance may ultimately suffer. This notion is echoed by Goffman, who states, “He who passes will have to be alive to aspects of the social interactions which others treat as uncalculated and unattended” (1963: 88).
Job attitudes
Identity management behaviors likely influence the way employees experience work and, thus, impact employees’ feelings and attitudes toward the job. This notion is consistent with theory posited by Roberts (2005), suggesting that outward attempts to manage others’ impressions of oneself may be especially detrimental to job attitudes when those outward impression management behaviors conflict with internal desires to behave authentically. Indeed, empirical research has shown that employees whose outward emotional displays conflict with their internal emotional states experience decreased job satisfaction and increased intention to turnover (Abraham, 1999; Côté & Morgan, 2002; Hochschild, 1983; Morris & Feldman, 1996; Pugliesi, 1999; Rafaeli & Sutton, 1987). Concealing behaviors, then, inevitably remind employees that they are behaving in an inauthentic manner that is inconsistent with internal desires and likely contribute to decreased job attitudes. Consequently, employees may view their working environment as a place where they cannot “be themselves,” which may ultimately trigger dissatisfaction.
One possible motive for disclosing a concealable stigma at work is to gain access to available resources and accommodations that may alleviate job stress. For example, pregnant employees may be motivated to reveal in order to negotiate maternity leave or flexible scheduling around doctor appointments (K. P. Jones et al., in press; King & Botsford, 2009). Similarly, LBG employees may reveal in order to access same-sex partner benefits. According to the job demands-resource model, lack of job resources leads to employee disengagement from the job (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001). Thus, employees who conceal will have limited access to available resources at work, which will in turn contribute to disengagement from work and subsequently worsened job attitudes. In contrast, revealing will facilitate greater access to available resources, which may ultimately facilitate more positive attitudes toward work.
Existing research from the management literature supports the notion that openness about one’s concealable stigma is associated with increased job satisfaction. For instance, degree of disclosure of one’s sexual orientation at work has been linked to increased job satisfaction, career commitment, affective commitment, promotion rates, and belief in the support of top management as well as decreased job anxiety, role conflict, work-family conflict, role ambiguity, and intentions to turnover (Day & Schoenrade, 1997, 2000; Griffith & Hebl, 2002; Ragins & Cornwell, 2001; Ragins, Singh, & Cornwell, 2007). Furthermore, one study found that concealment of sexual orientation at work was associated with increased retirement financial planning in a sample of gay and bisexual men (Mock, Sedlovskaya, & Purdie-Vaughns, 2010). The authors reasoned that the burden of concealing one’s stigmatized identity at work “tends to diminish the appeal of the workplace and thus may increase motivation to plan for life beyond work” (Mock et al., 2010: 123). Finally, empirical research has linked identity-consistent behaviors with increased job satisfaction and enjoyment of the job (Wrzesniewski, Dutton, & Debebe, 2003). Taken together, theoretical and empirical evidence supports the notion that concealing behaviors will contribute to decreased and revealing, improved, job attitudes.
To our knowledge, no existing research has examined the impact of signaling behavior on job attitudes. However, signaling requires an employee to be vigilant to potential acceptance and rejection cues from others at work, reminding the employee that she or he must take precautionary measures prior to being completely authentic with colleagues, which likely takes a substantial emotional toll. In other words, signaling behavior serves to remind the employee that “being yourself” is not “a given.” Thus, we propose signaling behavior will have a damaging effect on job attitudes.
Helping behavior
Being open and effusive about a concealable stigma has been theorized as an avenue through which to elicit helping behaviors from others including instrumental and psychosocial support (Chaudoir & Fisher, 2010; Clair et al., 2005). Indeed, working relationships must be characterized by some exchange of personal information in order to be effective (Kronenberger, 1991; Schneider, 1986). In light of a large body of disclosure literature suggesting that disclosing secrets can foster interpersonal liking (Berg & Wright-Buckley, 1988; Collins & Miller, 1994; Miller, 1990), intimacy (Laurenceau et al., 1998, 2005; Manne et al., 2004), and trust (Turner, Hewstone, & Voci, 2007), we reason that identity management behaviors play a substantial role in eliciting or inhibiting supportive helping behaviors from others at work.
Extant empirical research examining a variety of concealable identities substantiates this notion. For instance, Capodacqua (2007) found that to the extent LGB employees reported being more open about their sexual identities, they also reported receiving more psychosocial support from coworkers and supervisors. Furthermore, sexual assault victims who reported greater extent of disclosure also reported more positive and less negative social reactions to disclosure (Ullman & Filipas, 2001). In another study examining psychosocial support as a consequence of disclosure, Zea and colleagues (2005) found disclosure of HIV status was associated with greater quality of social support. Finally, a meta-analysis of relationship between perceived stigma, HIV disclosure, and social support found evidence of a positive overall correlation between disclosure and social support. In light of these findings, we contend a similar process occurs within the context of the workplace such that revealing behavior elicits, and concealing behavior undermines, others’ ability to help stigmatized employees via the provision of instrumental and psychosocial support.
Though little empirical research exists on the interpersonal implications of signaling including how signaling might impact the amount of psychosocial and instrumental support received from others at work as well as subsequent perceptions of discrimination, we expect that someone who tends to engage in high levels of signaling is in tune with the potential risks and benefits of disclosure as they relate to interpersonal consequences. Given the inherent purpose of signaling is to gather relevant information to gauge the confidante’s potential reaction to disclosure and backtrack if necessary, we argue that engaging in more of this behavior will facilitate better disclosure decisions on average. Thus, high signalers will use this additional information to inform their disclosure decisions and, as a result, will benefit from more positive interpersonal interactions. Implied in this reasoning is strategic decision making: high signalers will decide to express the identity when disclosure would result in positive interpersonal outcomes, and they will choose to suppress the identity when disclosure would result in negative interpersonal outcomes.
Perceived discrimination
The above reasoning would suggest that just as revealing elicits supportive helping behavior from others, these behaviors will also foster more positive interpersonal treatment and thus lower perceptions of discrimination from others. Because concealing may lead others to perceive the individual as antisocial, aloof, or even suspicious (Chrobot-Mason et al., 2001), situations that are high in concealing are likely not as positive relative to those that are low in concealing. That is, concealing behaviors may elicit negative reactions and, thus, more discriminatory treatment, from others because when employees conceal, they are likely isolating themselves in the process for fear the stigma would “slip,” which colleagues may perceive as cold, disinterested, and dismissive.
Evidence for the link between identity management and perceptions of discrimination can be found in the management literature, especially with regard to LGB employees. Indeed, Button (2001) found that to the extent LGB employees reported being more open at work about their sexual identities, they reported less discriminatory treatment toward sexual minorities at their workplaces. Similarly, Ragins and Cornwell (2001) found that LGB workers perceived less discriminatory treatment at work insofar as they were more open about their sexual identities at work. Further supporting the notion that concealing negatively impacts working relationships, concealment of an LGB identity has been linked to negative workgroup functioning (Chrobot-Mason et al., 2001). Finally, King and colleagues (2013) examined antecedents and consequences of identity management interactions among a sample of LGB employees and found that interactions were rated as being more comfortable, meaningful, and satisfactory to the extent that revealing strategies were used and concealing strategies were not used.
This phenomenon has also been observed in pregnant employees. For example, pregnant workers reported lower perceptions of discrimination when they indicated integrating (i.e., a revealing strategy) their pregnancy at work rather than an avoiding or counterfeiting their pregnancy (i.e., concealing strategies; K. P. Jones & King, 2010). Furthermore, a separate study examining identity management strategies used by pregnant employees suggested a negative relationship between revealing and experiences of discrimination at work (K. P. Jones, 2013). Taken together, existing theory and evidence indicates concealing can be detrimental to interpersonal working relationships and may actually elicit discriminatory treatment from others, whereas revealing likely fosters more positive interpersonal relationships, thereby reducing perceptions of discriminatory treatment. Based on our reasoning above, we argue signaling will foster beneficial interpersonal outcomes, including lower perceptions of discrimination.
Psychological well-being
Concealing one’s true identity runs counter to the need or desire to have one’s identity viewed as valuable and can therefore cause increased psychological strain (Clair et al., 2005; Goffman, 1963). While early theorists asserted that possessing a concealable (rather than visible) stigma was preferable since these individuals could “pass for normal” (E. E. Jones et al., 1984), recent stigma-based theories have highlighted the damaging psychological consequences of hiding one’s true identity from others (Clair et al., 2005; Ragins, 2008). Indeed, concealing triggers tremendous anxiety and fear as a result of “living a life that can be collapsed at any moment” (Goffman, 1963: 87). Individuals who choose to hide their concealable stigmas incur deleterious repercussions via psychological strain and emotional stresses (Cole et al., 1996; Frable et al., 1998; Major & Gramzow, 1999; Pachankis, 2007; Smart & Wegner, 1999, 2000). As Smart and Wegner state, “Concealing a stigma leads to an inner turmoil that is remarkable for its intensity and its capacity for absorbing an individual’s mental life” (2000: 221).
The impression management literature further echoes the potential toll inauthentic behavior can take on psychological well-being. According to Roberts (2005), our efforts to manage others’ impressions of us are driven by motives to behave in both a credible and authentic manner; however, these motives often conflict and engender psychological distress. For example, identity management interactions in which pregnant employees engaged in high levels of concealing triggered increased anxiety and depression relative to interactions in which the same employee concealed to a lesser extent (K. P. Jones, 2013).
Though relatively less research exists that directly examines the influence of identity management on psychological health in workplace settings, empirical research across a variety of other settings supports the psychologically damaging impact of concealing behaviors. For instance, Broman-Fulks and colleagues (2007) examined implications of disclosure among a sample of 319 adolescent victims of sexual assault, and the findings suggested youth who had disclosed to at least one person within a month of the sexual assault incident were less likely to report major depressive symptoms relative to their counterparts who concealed the assault. Furthermore, longitudinal data examining HIV disclosure among a sample of African American women indicated a negative link between disclosure and psychological functioning (Clark et al., 2003). Additional evidence demonstrated that women who concealed their abortion reported greater suppression of abortion-related thoughts, more intrusive abortion-related thoughts, and, thus, increased psychological distress over time (Major & Gramzow, 1999). Similarly, recent evidence suggests that certain identity management strategies reflective of revealing behaviors (e.g., community involvement, humor, and positive in-group stereotyping) related to increased self-esteem, whereas strategies embodying concealing (e.g., secrecy, selective disclosure) related to decreased self-esteem among a sample of 355 persons with mental illness (PWMI; Ilic et al., 2012). In light of the above evidence and theory suggesting the importance of authenticity for psychological coherence (Swann, 1987), we suggest situations in which employees conceal a concealable stigma will result in worsened psychological well-being immediately following the event as compared with situations in which employees are expressive of their stigma. Similarly, we expect revealing behavior will result in improved psychological well-being.
Though signaling behavior will promote positive interpersonal outcomes by providing more information on which individuals can base disclosure decisions, we contend that signaling will result in more negative intrapersonal outcomes. Our reasoning for this contention derives from the fact that signaling in and of itself serves as a cue that confidantes may be unaccepting and, thus, increases the salience of the stigma. Indeed, Pachankis’s (2007) disclosure model of concealable stigmas incorporated salience of stigma and stigma-related cues, arguing that insofar as a stigma is salient, psychological distress will ensue. Providing empirical support for the notion that degree of stigma salience triggers distress, one study examining 235 participants with concealable stigmas found that increased stigma salience positively predicted psychological distress (Quinn & Chaudoir, 2009). Thus, we extend this argument by incorporating the idea that increased salience of a stigma is one of the mechanisms by which signaling leads to distress. Indeed, recent empirical evidence suggested pregnant employees experienced increased anxiety following identity management interactions characterized by high-frequency signaling that led to increased anxiety relative to interactions in which those same employees signaled their pregnancies to a lesser extent (K. P. Jones, 2013).
Physical well-being
Similarly, our model posits a negative impact of concealing and signaling, and positive impact of revealing, on employee physical well-being. Indeed, Cole and colleagues (1996) found that concealment of sexual identity triggered subsequent increases in susceptibility to infectious diseases among gay men over a 5-year follow-up period. Similarly, disclosure of both sexual orientation and HIV status have been linked to improved immune functioning over time (i.e., within-person) among a sample of HIV-positive psychiatric outpatients (Strachan et al., 2007).
Extant evidence on this proposed link is somewhat mixed, however. For instance, some evidence suggests disclosure can be a potential risk to physical health well-being. For instance, K. P. Jones and colleagues (in press) examined identity management and physical health over time among a sample of pregnant employees. Analyses of leading and lagged indicators suggested concealing led to subsequent decreases in the frequency of physical health symptoms, whereas revealing led to subsequent increases in the frequency of physical health symptoms, pointing to discrimination avoidance rather than need or desire to have one’s identity viewed as valuable as the dominant process at play. That is, within-person variability in revealing and concealing behaviors led to different health outcomes across situations within individuals. Furthermore, among a sample of lesbian and gay employees, more than a third indicated experiencing verbal or physical harassment at work as a function of their sexual identity (Ragins & Cornwell, 2001). In addition, one study examining disclosure experiences among gay and lesbian individuals indicated that 75% of the sample incurred physical attacks or threats of physical violence as a direct result of disclosing their sexual identity (D’Augelli & Grossman, 2001). These findings suggest that at least in some cases, concealing may serve a protective purpose, safeguarding individuals from denigration.
In short, the evidence is mixed. Some research suggests that expressing a stigmatized identity will be beneficial for well-being via authenticity and the affirmation that one’s identity is seen as valuable; however, other evidence indicates revealing a stigmatized identity may put individuals at a disadvantage, making them increasingly vulnerable to discrimination and, thus, more susceptible to negative mental and physical health consequences. The divergent findings of the consequences of identity management likely depend on the boundary conditions discussed in the next section.
Boundary Conditions of the Identity Management-Consequences Relationship
In the following section, we discuss boundary conditions of the effect of identity management on mental and physical health well-being as well as interpersonal outcomes. In other words, we propose factors to explain when and why disclosing a concealable identity may lead to favorable proximal outcomes and under what conditions disclosure may foster adverse consequences.
Confidante’s reaction
Extant evidence suggests the extent to which a confidante’s reaction communicates acceptance subsequent to disclosure influences the effect of disclosure on mental and physical health well-being. Specifically, the proposed link between expressing a concealable stigma and improved psychological and physical health will be enhanced to the degree that confidantes’ reactions are accepting. This reasoning is consistent with the decision-making model for revealing personal secrets advanced by Kelly and McKillop (1996), which suggests that although revealing is beneficial if keeping a personal secret hidden is causing distress, the benefits of revealing will only be realized if the confidante’s reaction is positive and accepting. Though workplace research examining this question is limited, Griffith and Hebl (2002) found that coworker reactions (assessed as the extent to which coworkers displayed comfort, fairness, inclusiveness, and acceptance toward LGB workers) mediated the relationship between disclosure behaviors at work and improved job attitudes.
The importance of the confidante’s reaction to disclosure for ultimate well-being has been more extensively examined in the social psychological literature. For example, qualitative interview data from rape survivors found that negative social reactions to initial disclosure served as a silencing function discouraging women from disclosing in the future (Ahrens, 2006). These negative social reactions led to increased self-blame, uncertainty regarding the legitimacy of classifying their experience as rape, and increased doubt regarding the effectiveness of future disclosures. Furthermore, interview data from women who incurred sexual abuse during childhood suggested positive reactions from their current partners related to improved psychological (i.e., anxiety/depression) and psychosomatic health, yet negative reactions from friends during adulthood related to worsened health (Jonzon & Lindblad, 2005). Similarly, sexual assault victims reported more severe PTSD symptoms to the extent they had received negative social reactions upon disclosing their experience to others (Ullman & Filipas, 2001). Finally, Zea and colleagues (2005) found the link between disclosure of HIV status and improved psychological functioning was mediated by increased social support.
In addition to these survey studies, limited experimental research supports the notion that the acceptance of a confidante is critical to the well-being of the discloser. For instance, Rodriguez and Kelly (2006) instructed participants to write about a personal secret and imagine disclosing that secret to an accepting confidante, an unaccepting confidante, or to no confidante. In an 8-week follow-up, the participants in the accepting group reported fewer illnesses relative to those in the nonaccepting group. As we mentioned earlier, we reason that a confidante’s reaction to stigma disclosure may be even more impactful in a workplace setting given the wide range of high-stakes consequences that could potentially result from a negative interaction with someone at work, including outcomes related to performance evaluations, compensation, and/or promotion decisions. Thus, we expect that when an employee expresses a concealable stigma, she or he will benefit from improved mental and physical health well-being insofar as colleagues’ reactions to disclosure are positive.
Timing of disclosure
Another attribute of an identity management situation that we argue will influence the proposed linkages between revealing, concealing, and signaling and proximal outcomes via its influence on the confidante’s reaction is timing of disclosure. Generally, we expect later disclosures to foster more favorable confidante reactions relative to earlier disclosures in first-time interactions. In contrast, earlier disclosures will be more beneficial when the concealable stigma has some bearing on the work experiences of the confidante. Extant empirical findings suggest disclosing a concealable stigma earlier in a first-time interaction may disrupt initial impression formation processes, not allowing enough time for the benefits of individuating information to occur (Buck & Plant, 2011; Goodman, 2008).
For instance, Buck and Plant (2011) found male confederates who disclosed sexual orientation early in an interview were perceived and treated more negatively by male participants relative to male confederates who disclosed later in the interview, an effect that was fully mediated by the formation of stereotypic impressions. Similarly, Goodman (2008) experimentally manipulated timing of disclosure (early/late) of a visible (congenital skin condition with unattractive scarring) or a concealable stigma (concealable congenital skin condition) and found that in the concealable stigma condition, early disclosure led to more negative perceptions of the target relative to later disclosure. Specifically, when confederates with a supposed concealable stigma disclosed earlier, participants rated them as less likeable and less psychologically healthy. These findings suggest individuating information prior to disclosure can foster less biased evaluations of a target with a concealable stigma. Therefore, employees with concealable stigmas should delay disclosure in first-time interpersonal interactions to allow impressions to be formed based on individuating information.
Though early disclosure may disrupt impression formation in first-time interpersonal encounters, existing evidence from the management literature suggests that it may be beneficial under certain conditions. For example, early disclosure may be important when the confidante perceives he or she will be impacted in some way at work by the concealable stigma. Supporting this notion, Hebl and Skorinko (2005) found that disclosing a concealable disability earlier in an interview led to more positive evaluations of the interviewee as compared with those who waited to disclose. In this situation, interviewers may have perceived individuals who disclosed later as withholding important information that might potentially impact job performance (Goodman, 2008).
Similarly, King and Botsford (2009) advanced a model of pregnancy disclosure wherein they propose that delay of disclosure will foster more negative intrapersonal and interpersonal outcomes. Specifically, disclosing earlier would result in colleagues feeling more included in the process since they may likely be affected by accommodations granted during pregnancy and redistribution of work during maternity. Thus, earlier disclosure allows more time for planning and likely increases perceptions of fairness. Taken together, extant theoretical and empirical evidence suggests later disclosure is preferable unless the concealable stigma impacts the confidante’s workplace experience in some notable way.
Visibility
Given that concealable stigmas exist on a continuum of visibility, we briefly discuss how the extent to which a concealable stigma is readily observable can influence the impact of concealing, revealing, and signaling in meaningful ways, noting the focus of our review remains on the management of concealable (as compared with visible) stigmas at work. Some stigmas, though not definitively obvious, may be more visible and, thus, more likely to create suspicion than others. This is particularly relevant for stigmas with a specific course (E. E. Jones et al., 1984), or those that are dynamic and changing in some way over time. For example, pregnancy follows a specific course, starting as completely concealable, and gradually becoming more conspicuous over time. A similar process likely unfolds in employees diagnosed with HIV/AIDS and various forms of cancer such that, as the stigma becomes increasingly visible, it is more and more difficult to hide from others (Hays et al., 1993). Thus, we expect visibility of a stigma at the time of a specific situation to moderate the proposed linkages between identity management and interpersonal outcomes such as the provision of supportive helping behaviors and perceptions of discrimination. Specifically, we expect that to the degree a stigma is visible in a given situation, the beneficial impact of revealing and signaling will be attenuated and the damaging impact of concealing will be exacerbated.
The implication of this reasoning is that revealing an identity will be particularly meaningful to the confidante to the extent the target felt he or she had a choice in the matter. Revealing a stigma that is completely concealable communicates trust in the confidante and suggests this trust was the sole reason for disclosure. In contrast, revealing a stigma that is somewhat visible (but still ambiguous) may still convey trust in the confidante, but could also be attributed to increased pressure to explain oneself due to the increased “evidence” of the stigma. Similarly, concealing a stigma will be more negatively perceived to the degree that others are suspicious of the identity because it communicates distrust in the confidante. However, when then the stigma is completely concealable, others will not respond as negatively to concealing because they will not as likely be aware the target is hiding something. Supporting this view, one study found that participants rated fictitious pregnant employees as more likable, more committed, and more likely to be promoted when targets were expressive of the pregnancy and not showing relative to when they were expressive of the pregnancy and showing (Botsford Morgan, Singletary Walker, Jones, & King, 2012). That is, others perceived a revealing strategy as effective only when they were unsuspecting of the target’s pregnant status.
Feedback Loop
Finally, our model posits feedback loops from the proximal outcomes of an identity management interaction to anticipated acceptance in the next identity management interaction. Indeed, critical antecedents to expectancy judgments are past successes and failures related to engaging in the particular behavior in question (Bandura & Locke, 2003). That is, employees will base their expectations about future identity management interactions on the outcomes associated with previous identity management interactions. To the degree an employee experienced an identity management interaction positively, she or he will be more likely to anticipate acceptance in future interactions. Similarly, negative identity management experiences are likely to foster negative expectations of future encounters.
Indeed, extant research has found support for this notion in workplace settings. For instance, perceptions of past discrimination positively predicted fears about disclosing among a sample of LGB workers (Ragins et al., 2007). Furthermore, qualitative data on 228 lesbian workers suggested that prior loss of job due to disclosure of sexual identity impacted subsequent decisions and concerns about revealing their identity to coworkers (Schneider, 1986). Thus, based on theoretical reasoning and empirical evidence, we expect employees’ expectations about identity management interactions are in large part shaped by previous experiences.
Between-Person Identity Management
As can be seen in Figure 1, both identity management (i.e., concealing, revealing, signaling) and meaningful workplace outcomes exist at the within-person and the between-person levels, meaning that these constructs will vary as a function of the situation (as described above) but also as a function of the employee. Thus, our between-person model incorporates behavioral averages, behavioral tendencies, and behavioral accumulation to facilitate the comparison of identity management processes between employees. That is, we consider employees’ average identity management behaviors (i.e., the aggregation or average of identity management behaviors across specific interactions), their general identity management behavioral tendencies (i.e., the degree to which they tend to rely on certain strategies more than others), and finally, accumulation effects (i.e., the degree to which specific identity management interactions at the situation level accumulate over time to produce meaningful differences among employees). The underlying assumption, then, of the between-person model is the comparison of employees to one another.
Antecedents of Identity Management
Our model identifies several factors specific to an individual employee that will impact how that person manages a concealable stigma at work on average relative to other employees. These antecedents of person-level identity management tendencies reflect stable characteristics of an individual employee.
Perceived organizational and supervisor support
Individuals burdened with managing concealable stigmas will base identity management decisions in part on the degree to which they perceive their organizations as supportive. Perceptions of support may be gleaned from various cues or signals stigmatized employees observe via the climate or culture of the organization, organizational policies and procedures, or numerical representation of their group at their organization. Insofar as these aspects of the organization communicate support and acceptance, individuals perceive their working environments as more psychologically safe, and thus tend to reveal more and conceal less as compared with individuals working in less supportive environments. Indeed, extant evidence suggests lesbian and gay employees report greater disclosure of their sexual identities at work insofar as they perceive their organization as gay supportive (Griffith & Hebl, 2002) and encompassing an LGB-affirming climate (Chrobot-Mason et al., 2001).
Furthermore, supportive organizational policies tend to promote disclosure of concealable stigmas. For instance, research suggests more supportive LGB policies and practices are associated with greater degree of openness about sexual identity at work among LGB employees (Button, 2001; Griffith & Hebl, 2002; Ragins & Cornwell, 2001). Similarly, K. P. Jones and colleagues (in press) found that pregnant employees who worked in organizations with more supportive work-family cultures engaged in less concealing on average relative to pregnant women working in organizations with less supportive work-family cultures. Finally, Murphy and colleagues (2007) found evidence suggesting that contextual cues such as numerical underrepresentation of one’s group may signal to employees that their social group is devalued in the organization, which may ultimately undermine perceptions of support and lead to subsequent declines in performance. Thus, employees’ perceptions of organizational support are critical in determining the degree to which they feel comfortable expressing their concealable stigmas at work.
Finally, research has shown that individuals with concealable stigmas who perceive their supervisors as supportive tend to be less fearful of disclosure and engage in less concealing as compared with individuals who do not benefit from supportive supervisors (K. P. Jones, 2013; K. P. Jones et al., in press; Ragins et al., 2007). Less empirical research has examined the influence of such forms of support on signaling behaviors; however, given the strategic nature of signaling and its likelihood of occurring in ambiguous or uncertain contexts (Clair et al., 2005; Woods, 1994), we expect individuals who perceive moderate levels of support at work (i.e., moderately supportive supervisors, climates, policies) will engage in more signaling on average as compared with invisibly stigmatized workers who perceive their workplaces as extremely low or extremely high in terms of support.
Individual differences
In addition to perceptions of workplace support, individual differences will also play a role in determining the extent to which a person generally reveals, conceals, and signals a concealable stigmatized identity at work. Specifically, the personality traits of extraversion and neuroticism, stigma centrality, and hierarchical level in the organization will determine an individual’s average identity management tendencies.
Neuroticism is associated with chronic negative affect (Watson & Clark, 1984), the prevalence of psychiatric disorders (Zonderman, Stone, & Costa, 1989), and the adoption of avoidance goals (Elliot & Thrash, 2002; McAdams & Pals, 2006). Furthermore, people who tend to pursue avoidance goals are generally more likely to attend to negative cues and to interpret neutral or ambiguous stimuli in a negative light (Derryberry & Reed, 1994; Strachman & Gable, 2006). Thus, individuals with concealable stigmas who possess chronically activated avoidance systems tend to be more sensitive to the possibility of social rejection and, thus, opt to conceal their stigma as a method of coping (Chaudoir & Fisher, 2010). Indeed, a longitudinal study of 200 individuals diagnosed with HIV/AIDS revealed that individuals who cited avoidance-related disclosure goals were consistently less likely to disclose their stigma to a chosen confidante (Chaudoir, 2009). These findings suggest employees high in neuroticism may be simply less likely to disclose because they are chronically worried about potential negative consequences of disclosing, such as social rejection.
Given signaling behaviors are likely to “invite speculation . . . encouraging peers to read between the lines” (Woods, 1994: 176), individuals high in neuroticism will be less likely to engage in a signaling strategy relative to those who are lower in neuroticism. In other words, individuals high in neuroticism will go to great lengths to keep others from asking questions and possibly “finding them out” because of their fears associated with the concealable stigma becoming public. Consequently, these individuals would be less likely to engage in signaling behavior since it may lead to increased speculation and unwanted questions from others.
Another dimension of personality that likely plays an important role in predicting identity management decisions is extraversion. Indeed, extant evidence suggests extraversion is positively associated with self-disclosure (Cozby, 1973) and the adoption of approach goals (Elliot & Thrash, 2002; McAdams & Pals, 2006). Furthermore, empirical evidence supports the notion that extraverts are simply happier and more comfortable than introverts in social situations (Diener et al., 1984; Moskowitz & Cote, 1995; Pavot, Deiner, & Fujita, 1990). In light of research suggesting people who tend to adopt approach goals are generally more likely to attend to positive cues and interpret neutral or ambiguous stimuli in a positive light (Derryberry & Reed, 1994; Strachman & Gable, 2006), we expect extraverted employees will be more likely to reveal, and thus less likely to conceal, a concealable stigma. Furthermore, given their focus on the benefits associated with disclosing, it seems less likely that extraverted employees would feel the need to signal their concealable stigmas since signaling strategies imply uncertainty with regard to whether or not to disclose.
We further argue that stigma centrality—the extent to which a person’s stigmatized identity is central to his or her self-identity—will influence the extent to which an employee reveals, conceals, and signals a concealable stigma to others at work on average. That is, insofar as the concealable stigma is integral to a person’s self-concept, the individual will tend to reveal and signal more but to conceal less as compared with individuals with less central stigmas. Indeed, to the extent the concealable stigma is central to and valued by an employee, self-verification motives will drive the individual to be authentic with others and expressive of the identity (Ragins, 2008). For instance, extant research supports the notion that LGB workers with highly central sexual identities are more likely to be expressive of their sexual orientations at work relative to LGB workers lower in sexual identity centrality (Chrobot-Mason et al., 2001; Griffith & Hebl, 2002).
Given the value attached to a highly central stigma, employees will be motivated to express the identity even in ambiguous situations. Thus, when it is unclear if a confidante will be accepting, high-centrality employees will be more motivated to gather the information necessary to discern whether its safe to express the identity even if it does invite increased speculation from others. Indeed, the benefit that high-centrality individuals would gain from expressing the identity outweighs any potential consequences associated with triggering suspicion from others. Therefore, employees whose concealable stigmas are highly central will tend to engage in more signaling as compared with their counterparts whose stigmas are less central.
Finally, drawing from power-dependency theory (Emerson, 1962), we argue that hierarchical level of the organization will influence how employees manage concealable stigmas. Specifically, we expect that employees at higher levels of the organization will reveal more, and conceal and signal less, on average. Inherent in this argument is the assumption that employees at higher organizational levels possess more power and, thus, greater control of resources and meaningful outcomes for others in the organization. Consequently, others at work may be highly dependent upon these individuals in various ways, thereby motivating others to make favorable impressions and foster positive working relationships with the target employee. Thus, employees with concealable stigmas who work at higher levels of the organization will likely hold fewer fears about being themselves with others given others high dependency upon them. This power acts as a sort of “just-in-case” leverage if negative consequences ensue. In contrast, employees at lower levels of the organization do not have the same kind of leverage as higher-status employees and thus would likely be more cautious about disclosing a stigma, concealing and signaling more on average, given their relative lack of power if negative reactions followed disclosure. Therefore, as organizational level increases (i.e., as power increases), revealing increases and concealing and signaling decrease.
Consequences of Identity Management
In the within-person section above, we articulated the proximal consequences of identity management at the situation level. That is, we described how engaging in concealing, revealing, and signaling in specific situations would lead to varying implications for workplace outcomes immediately following the situation. Drawing from these ideas, we expect these relationships will hold at the between-person level given that our model reflects a process model wherein the impact of identity management specified at lower levels is analogous to the impact of identity management specified at higher levels (see Chan, 1998). Specifically, we expect mean concealing will lead to lower average performance, job attitudes, helping behavior, well-being, and higher average perceptions of discrimination. Furthermore, we anticipated mean revealing will lead to higher average performance, job attitudes, helping behavior, well-being, and lower average perceptions of discrimination. Finally, we anticipate mean signaling will lead to more positive interpersonal consequences on average (i.e., more helping behavior, lower perceptions of discrimination) but worsened average job attitudes, performance, and well-being as a function of its high cognitive and emotional load.
Furthermore, as described earlier, the between-person level reflects the accumulation of situations over time. Indeed, recent empirical research indicates that the physical health impact of identity management experiences can accumulate markedly over time, ultimately producing substantial differences among employees. Specifically, K. P. Jones et al. (in press) examined the relationship between workplace identity management and physical health symptoms over the course of pregnancy. Their findings showed that assuming employees’ physical health symptoms scores were at the mean (2.33 on a 4-poing scale) at the beginning of the study, an employee initially one standard deviation below the mean on concealing would report much more frequent physical health symptoms by the end of the 22-week study (3.18 on a 4-point scale) as compared with an employee who initially started at the concealing mean (2.77 on a 4-point scale) or an employee who initially started one standard deviation above the concealing mean (2.37 on a 4-point scale), demonstrating that while changes may be relatively smaller if considered in a vacuum, the accumulation of these smaller changes can be quite meaningful over time.
Conclusions and Future Directions
Given our comprehensive review of psychological, sociological, and management literatures on concealable stigma management, we can draw several conclusions about the state of the literature. First, it is clear that it is insufficient to view identity management and disclosure decisions as a dichotomous phenomenon. That is, it is not enough to simply ask whether someone is “in” or “out” at work. Such a dichotomous perspective does not allow for the possibility (and likelihood) that employees with concealable stigmas are expressive of their identities to varying degrees with different people at work. Instead, identity management processes continuously develop and unfold over time, with experiences from one interaction shaping expectations and behaviors in future interactions. However, stable person-level attributes shape employees’ average identity management behaviors, on which employees can be compared with one another. Furthermore, these average behavioral tendencies reflect the aggregation of identity management interactions across time.
Second, we conclude that revealing a stigmatized identity tends to result in favorable interpersonal and intrapersonal outcomes, whereas concealing a stigma leads to more negative outcomes. Furthermore, signaling will generally foster positive interpersonal relationships while detracting from performance, job attitudes, and physical and psychological well-being. However, there are specific boundary conditions surrounding these effects. The beneficial impact of expressing a stigmatized identity will be more fully realized when the stigma is completely concealable, whereas the increasing visibility of the stigma will exacerbate the damaging interpersonal effects of stigma suppression.
Finally, it is important to consider boundary conditions surrounding the intrapersonal impact of identity management behaviors. For example, the beneficial effect of revealing on physical and mental health well-being will be attenuated to the extent the confidante is unaccepting in his or her reaction to disclosure. Similarly, concealing and signaling will be intrapersonally detrimental insofar as confidantes’ reactions are negative. Finally, later disclosure appears more beneficial in situations in first-time interactions wherein first impressions are forming; however, earlier disclosure seems the most advantageous option when the stigmatized identity has some bearing on confidante’s work experiences.
Implications for Practice
Our integration of the extant identity management literature into a multilevel model of identity management yields several implications for organizational practitioners. First, it is clear that supervisor and organizational perceptions of support can function to improve stigmatized workers’ experiences. Insofar as workers with concealable stigmas perceive their organizations and supervisors as supportive, they will be more likely to express their stigma at work, thereby enhancing performance, job attitudes, and well-being and fostering more positive and supportive interpersonal relationships. Furthermore, employee perceptions of support may function to buffer the unfavorable effects of negative identity management interactions. Thus, practitioners should ensure the culture and climate emanate support to all employees starting with upper management. In addition, reevaluation and potential revision of organizational policies to better communicate support to employees burdened by a variety of concealable stigmas should be considered.
Second, supportive supervisors also play a critical role in positively shaping the work experiences of invisibly stigmatized employees. That said, practitioners could design supervisory training that specifically addresses many of the issues faced by workers who are burdened by concealable stigmas. This goal of such training would be to increase awareness among supervisors of the importance of providing support to all subordinates, including and perhaps especially those who may not appear in need of support.
Third and finally, dispositional traits of employees burdened by concealable stigmas in part dictate how they manage their identities on average. Extant evidence and our multilevel model suggest highly neurotic individuals may be at higher risk for more negative identity management experiences, whereas highly extraverted individuals might be more prone to experience identity management positively. From a managerial perspective, this information could help identify “high-risk” individuals who may be experiencing an especially difficult time and take extra steps to ensure these individuals feel supported. Furthermore, the notion that individuals might manage and experience diversity differently as a function of individual level traits could be incorporated into general diversity training.
Future Directions
First and foremost, the literature review described in the preceding sections unfortunately suggests that much of the extant scholarship on concealable stigma management has been conducted without a specific focus on the workplace context. In light of this gap and the importance of concealable stigma management for working relationships, performance, job attitudes, and employee well-being, it is our hope that this review will inspire organizational scholars to pursue research in this area.
Furthermore, the current review clarifies future research must examine identity management from a multilevel perspective, taking into account stable aspects of the person, variable characteristics of the situation, as well as the interaction between the two. To date, most research has focused on the between-person level; less has taken a longitudinal, within-person focus; and finally, almost no research has taken an integrative person-situation perspective. Future research should work toward filling this important gap in the literature, developing measures that accurately capture this multilevel perspective of concealable identity management. Measures should not only assess how identity management changes over time (i.e., within-person), but also identity management behavioral averages, tendencies, and accumulation effects to facilitate more meaningful comparisons among employees.
Indeed, one critical implication of taking a multilevel perspective on identity management and disclosure is the consideration of cross-level moderators of within-person linkages. Indeed, employees respond to situations based in part on their predispositions (and here, we conceptualize our between-person level antecedents as dispositional constructs). Thus, their dispositional tendencies will exacerbate or alleviate the effects of situational variables. Below, we detail some of our preliminary ideas about potential cross-level moderators of these relationships (delineated by dotted arrows in Figure 1), which we hope management scholars will pursue further in future research.
First, we expect that high perceptions of organizational and supervisor support, high extraversion, low neuroticism and low stigma centrality, and high hierarchical level in the organization will function to alleviate the effects of situational variables. That is, under these conditions, there is generally less “at stake” in a specific interaction, so situation-specific linkages are not as meaningful. For example, if individuals perceive high levels of support in their workplace, the anticipated acceptance of one particular person is not as critical since they have other support systems to fall back on if the interaction does not unfold positively. Similarly, the consequences associated with a particular identity management interaction are less impactful in light of the larger overarching support system.
In contrast, low perceptions of organizational and supervisor support, low extraversion, high neuroticism and high stigma centrality, and low hierarchical level in the organization function to exacerbate the associations between situational predictors and consequences of a particular identity management interaction. That is, employees who perceive little support or who are particularly sensitive to rejection use acceptance cues to a greater extent when deciding how to manage their stigmas at work and will be more impacted (either negatively or positively) by how the specific interaction unfolds. Taken together, the above reasoning suggests the stable person-level aspects described in our model shape the larger context in which specific identity management interactions unfold, and in doing so, they function to constrain or exacerbate within-person, situation-specific model linkages. Given the preliminary nature of these ideas, we strongly encourage organizational scholars to pursue exploration of cross-level moderators in future research.
Additionally, further consideration of identity management of visible stigmas is warranted. Future research should examine how visible stigma management differs from the processes discussed in this review. Indeed, complex processes are involved even when a stigma can be immediately observed; in such situations, individuals are still faced with deciding when to acknowledge the stigma and when to suppress it. It is important to further consider how the effectiveness of identity management strategies differs as a function of the visibility of the stigma.
Another area ripe for future research is the important yet understudied identity management strategy of signaling. Our model theorizes signaling will have beneficial interpersonal implications but may have intrapersonal costs; however, more research needs to be done to further substantiate this notion. Furthermore, we propose when signaling occurs within an interaction, it functions as an immediate precursor to subsequent revealing and concealing behaviors. Future research should generate creative measurement techniques to tease apart the temporal ordering of revealing, concealing, and signaling within identity management interactions as well as their antecedents and consequences. Finally, given the strategic nature of signaling, it seems reasonable to assume some individuals may be more skilled at signaling than others. Thus, consideration of the individual-level traits that characterize a “good” signaler and a “bad” signaler is an important area for future inquiry.
As described above, our model incorporates stable individual differences as person-level predictors of average identity management behaviors and as cross-level moderators of within-person linkages of identity management. These individual difference variables include two personality traits we argue are particularly relevant to identity management—extraversion and neuroticism—in addition to stigma centrality and hierarchical organizational level. Clearly, these four individual difference variables will not always align perfectly with our main effects predictions, and interactions among these traits in their prediction of identity management surely exist. For example, an employee could be high in neuroticism and extraversion, or low in extraversion but in a high-level position. Indeed, there are several possible profiles that could result from different configurations of these four variables. Though these possibilities were outside the scope of our review, we encourage future researchers to consider the complex relationships that likely emerge between different combinations of individual difference traits and identity management behaviors.
Finally, a fruitful area for future research is the examination of identity management processes in individuals burdened by multiple stigmatized identities. For instance, does an HIV-positive lesbian employee manage her LGB identity and her HIV identity separately or together as an integrated identity? Furthermore, does an African American employee who is pregnant face different challenges in managing her pregnant identity at work relative to a Caucasian employee who is pregnant? It is our hope that rich questions such as these will spark a new era of stigma identity management research driven by a framework that conceptualizes identity management as an ongoing and complex process with both intrapersonal and interpersonal antecedents and consequences.