The communication of sexual identity images in a self-presentational context

Abstract

We examined the disclosure of sexual attitudes in self-presentational contexts across different levels of accountability. The results indicated that in a self-presentation situation participants expressed more reserved sexual attitudes in a low-accountability (i.e., stranger) compared to a high-accountability (i.e., friend) condition. Conversely, in a non-self-presentation situation sexual attitudes did not differ as a function of accountability conditions. Moreover, participants expressed greater accountability concerns with friends compared to strangers, a finding that heretofore has been absent from the literature. In fact, perceived accountability mediated the interactive effect of self-presentation and accountability on sexual attitudes, again a finding that has yet to be demonstrated in prior work. These findings are the first to illustrate how people convey their sexual attitudes for self-presentational purposes, and they provide a more nuanced and complete view of people's self-presentations.

Keywords:

• Self-presentation
• Sexual identity images

Theorists characterize self-presentation as a purposive, goal-directed activity in which “people control information (consciously or non-consciously) about themselves to influence the impressions formed by others” (Schlenker, 2003, p. 492). Typically, people try to accomplish their self-presentation goals by revealing private self-information that they deem appropriate to an intended audience; their primary intent is to communicate a desired identity image (Derlega, Metts, Petronio, & Margulis, 1993). In general, as people's behavior becomes more public their concern with how it appears to others increases, hence their motivation to manage self-presentation disclosures correspondingly increases (Arkin, Appelman, & Berger, 1980).

These disclosures often include descriptive or evaluative information that can range from the more general facts of one's life (e.g., you have a high GPA) to the more intimate details, which at times may include one's sexual attitudes or experiences (i.e., sexual identity images).1 It is important to note that we are not suggesting people wantonly publicize their sexual attitudes. Rather, this type of information simply conveys one element of people's identities and on occasion people may convey such information to others.

To date there is no extant research directly examining how people utilize information concerning their sexual attitudes as a means to self-present a desirable identity image (see Afifi & Guerrero, 1998; Lucchetti, 1999). Thus the aim of the current study is to fill this gap in the literature by examining the communication of sexual identity images in a self-presentational context.

Self-Presentation of Sexual Identity Images

In broad terms, appropriate disclosures provide social rewards (e.g. increased liking), promote healthy personal functioning, and build close relationships, while inappropriate disclosures incur social costs (e.g., perception of maladjustment; avoidance by others), and leave one vulnerable to criticism, ridicule, and negative judgment (Herold & Way, 1988). For example, revealing personal information that violates social expectations (e.g., revealed too soon in a relationship) can create an uncomfortable situation, and ultimately lead the target other to dislike the disclosing individual (Derlega et al., 1993). Likewise, evidence from studies where intimate disclosures are defined quite broadly (e.g., financial problems, relationship difficulties, mental illness) suggests that people are at times hesitant to reveal the more intimate details of their life (Dolgin & Minowa, 1997; Petronio & Martin, 1986). This reluctance appears to be rather well founded; overall, with all else equal, people who share such information tend to be viewed somewhat negatively (Caltabiano & Smithson, 1983; Chaikin & Derlega, 1974). In short, although not directly focused on self-presentation or sexual identity images per se, evidence suggests a negative relationship between the intimacy levels of a disclosure and liking for the discloser, particularly during the early stages of a developing relationship (Chaikin & Derlega, 1974).

As these potential social risks suggest, people may proceed with caution before they share certain types of personal information (Baxter & Montgomery, 1996; Kelly & McKillop, 1996). An area that is likely to be viewed as particularly risky involves revealing information concerning one's sexual attitudes and experiences (Afifi & Guerrero, 1998). These types of disclosures can involve communicating information about a wide range of sexual topics including likes, dislikes, beliefs, and experiences.

Although the valence of this sexual information may certainly range from conservative (i.e., not permissive at all) to liberal (i.e., very permissive), it is plausible that the majority of people view a somewhat reserved sexual persona as the optimum desired identity image to convey to others. In fact, we have results from pilot work2 indicating that people associate more reserved sexual attitudes with a favorable identity image. However, one may wonder why people would necessarily view a reserved identity as a more desirable image? In a word: stigma. In general, “stigmatized individuals possess (or are believed to possess) some attribute, or characteristic, that conveys a social identity that is devalued” (Crocker, Major, & Steele, 1998, p. 505). Although typically work on stigma is directed at physical issues, research also shows that individual personal characteristics (e.g., pessimistic) can likewise be relatively stigmatizing (Helwig-Larsen, Sadeghian, & Webb, 2002; Tyler & Gill-Rosier, 2009).

For instance, people evaluate hypothetical descriptions of an individual more negatively if the description depicts the individual's sexual attitudes and experiences as more permissive rather than reserved (Brady & Halpern-Felsher, 2008; Marks & Fraley, 2005; Sprecher & McKinney, 1993). Put differently, a sexually permissive identity image is typically perceived as less socially desirable, whereas a more reserved identity may often be viewed in a more favorable light. Although this evidence is not directly related to self-presentation, it raises a plausible reason why people could be motivated to self-present a more reserved image—there may be some degree of social stigma associated with a permissive identity. Thus, conveying a more reserved sexual image could occur because people are aware that doing otherwise may lead to being stigmatized and socially devalued by others. Hence, when sharing information about their sexual beliefs, people may be motivated to communicate a more reserved sexual identity.

It is also important to note that accountability may influence the self-presentation of any particular identity image (Schlenker & Pontari, 2000). That is, when people perceive that an audience may be privy to accountability information, their self-presentations are typically more cautious. By comparison, when accountability pressures are minimal, self-presentations tend to be more favorably biased (Baumeister, 1982). For example, when interacting with a stranger people's self-presentations tend to be more self-enhancing, whereas when interacting with a friend their self-presentations tend to be less self-enhancing (Tice, Butler, Muraven, & Stillwell, 1995). It is presumed that this occurs because friends know more about the individual than strangers and could potentially dispute overly favorable claims. In other words, accountability concerns are presumably higher with friends than with strangers; however, it is important to point out that this proposition has never been empirically examined.

Thus, although people may be typically motivated to present a more reserved sexual image, it is plausible that the degree to which the expressed identity is more or less reserved may depend on whether one is interacting with a friend or a stranger. In short, the aim of the current study is to examine whether sexual identity images differ as a function of self-presentation and accountability and whether actual accountability concerns influence this effect. We utilized a 2 (self-presentation: public vs private) × 2 (accountability: friend vs stranger) between-participants design. Only female students were recruited in our study for reasons of cross-dyadic control. We expected participants in the public condition to convey a more reserved sexual identity image in the stranger compared to friend condition, whereas in the private condition we expected no significant difference in participants’ sexual identity image across accountability conditions. Moreover, we expected participants’ accountability concerns to mediate the relationship between self-presentation and accountability.

Method

Participants and procedure

A total of 155 undergraduate female participants received extra credit in their courses for participating. The mean age was 20.29 (SD = 1.50). Of the sample 63% were single/not exclusive and 36% were married/engaged/exclusive; 152 participants indicated that they were heterosexual. When registering, participants were instructed to bring a female friend with them to the lab session; participants (and friends) were randomly assigned to an accountability condition (friend vs stranger) and to a self-presentation condition (public vs private). Multiple pairs of participants attended each session.

Upon arrival participants were placed in separate research rooms where they were informed that the purpose of the study was to examine communication patterns across different topics. The experimenter explained that participants would fill out a questionnaire to assess their experiences, attitudes, and beliefs concerning various topics, and then they would have a brief conversation with their friend or another student (i.e., a stranger) (in reality the conversation never occurred). The dependent variable comprised the 10 sexual attitude items from the Brief Sexual Attitudes permissiveness subscale (C. Hendrick, S. S. Hendrick, & Reich, 2006). To disguise the intent of the experiment the 10 items were randomly embedded among a host of other items (e.g., religion, athletic, academic, financial items). All items required participants to indicate their level of agreement for each statement using a 9-point scale (anchored at 1, strongly disagree and 9, strongly agree). The scores on the sexual attitude items were reverse-coded where appropriate, such that low values indicated a reserved identity and high values indicated a permissive identity.

Participants completed the questionnaire after they received the following manipulations. In the public condition, participants were required to write their name and student identification number on the first page of the questionnaire, and they were told that their questionnaire responses would be shared with their conversation partner prior to the interaction. For the private condition, participants were told that their responses would remain completely anonymous and would not be shared with their conversation partner. Specifically, they were assured that their responses would remain totally confidential and that even the researcher would be unable to associate them with their specific response sheet. To reinforce the confidentiality of our procedures, participants’ response sheets were anonymous (coded by condition acronym with no names), and participants were provided with blank envelopes (coded to match post-experimental questionnaire) in which they sealed their response sheets before randomly placing them in a box of other identical envelopes. To induce high accountability, participants were told that their friend would serve as their conversation partner. By comparison, to induce low accountability, participants were told that an unacquainted undergraduate student (female) would be their conversation partner. Across all conditions, participants were instructed that during their conversation they might broadly discuss their responses to the questionnaire items.

Participants also filled out a post-experimental questionnaire that included three items (adapted from Kray, Thompson, & Lind, 2005) to assess the degree to which participants experienced accountability concerns using a 7-point scale anchored at 1, not at all accountable/concerned and 7, very accountable/concerned (e.g., how accountable are you for the responses that you made; how accountable did you feel when you made your responses; are you concerned that your responses would be viewed by others). Participants were told the packet would not be shared with the partner. The packet also included four items to measure the frequency of participants’ religious activities (on an 8-point scale with 1 indicating a very high frequency and 8 a very low frequency) and three items to measure their overall religiosity (on 4-point scale with 1 indicating very religious and 4, not very religious). In addition, participants indicated how close they were to their friend using a 9-point scale (anchored at 1, not at all close and 9, very close) and the length of the friendship (in months).

Results

Manipulation checks

To assess the degree of friendship, participants indicated how long (in months) they had known the friend (M = 33.20, SD = 46.17) and the closeness of their relationship with the friend (M = 7.41, SD = 1.64). The length and closeness of the friendship suggests that participants brought a friend to the study rather than just a mere class acquaintance. In addition, analysis of both measures indicated no significant differences across the self-presentation conditions, ts < 1.50.

Accountability concerns

An ANOVA indicated that participants viewed accountability concerns (average of three accountability items: α = .82) differently across both the self-presentation and accountability conditions, respectively, F(1, 151) = 34.75, p < .001, η ²= .19; F(1, 151) = 18.04, p < .001, η ²= .11. Specifically, accountability concerns in the public condition (M = 6.37, SD = 1.77) were significantly greater than in the private condition (M = 5.14, SD = 1.00). Likewise, participants reported significantly greater accountability concerns in the friend condition (M = 6.21, SD = 1.78) compared to the stranger condition (M = 5.31, SD = 1.18). These main effects were qualified by a significant interaction effect, F(1, 151) = 18.83, p < .001, η ²= .11. Planned comparisons showed that participants’ accountability concerns in the public condition were significantly greater in the friend (M = 7.27, SD = 1.86) compared to stranger condition (M = 5.47, SD = 1.16) t(151) = 6.13, p < .001, d = 1.16, whereas in the private condition there was no significant difference in participants’ accountability concern across the friend (M = 5.12, SD = .75) and stranger conditions (M = 5.15, SD = 1.19) (t < 1).

Sexual attitudes

An ANOVA was conducted using the average of the Brief Sexual Attitudes permissiveness subscale (α = .94) as the dependent variable (recall that low values indicate a reserved identity and high values indicate a permissive identity). In line with expectations, the analysis revealed significant main effects for both the self-presentation and accountability conditions, respectively, F(1, 151) = 8.11, p < .01, η ²= .05, F(1, 151) = 12.27, p < .001, η ²= .08. However, these effects were qualified by the predicted interaction between the self-presentation and accountability conditions, F(1, 151) = 8.56, p < .01, η ²= .05. Planned comparisons showed that participants in the public condition expressed a significantly more reserved sexual identity image in the stranger condition (M = 2.07, SD = 0.97) compared to the friend condition (M = 3.81, SD = 1.82) t(151) = 4.59, p < .001, d = 1.19, whereas in the private condition there was no significant difference in participants’ sexual identity image across the stranger (M = 3.63, SD = 2.21) and friend conditions (M = 3.79, SD = 1.49) (t < 1).3 When the religious frequency (M = 5.80, SD = 1.46, α = .76) and overall religiosity measures (M = 2.59, SD = 0.86, α = .80) were employed as covariates, the hypothesized effects of self-presentation and accountability on sexual attitudes remained significant.

Mediation analysis

We utilized an SPSS macro designed by Preacher, Rucker, and Hayes (2007) that uses path analysis to compute mediated moderation tests. The aim was to examine whether accountability concerns mediate the interactive effect of self-presentation and accountability on sexual attitudes. The results indicated that the relationship between the predictor term (self-presentation and accountability interaction) and the mediator (perceived accountability concerns) was significant, (β = −0.46, t = −4.34, p < .001), as was the association between the mediator and the dependent variable (sexual attitudes), (β = 0.25, t = 2.45, p < .01). A Sobel test (1982) showed that the indirect effect of the self-presentation and accountability interaction on sexual attitudes via perceived accountability concerns was significant (z = 2.14, p < .05) (Brandt & Reyna, 2010).

General Discussion

Although there is a well-developed literature documenting self-presentation efforts in general, there is no extant research explicating the communication of sexual identity images for self-presentational purposes. The current study aimed to fill this gap in the literature by examining how such identity images differ as a function of self-presentation and accountability, and whether people's accountability concerns mediate this effect. The contributions are straightforward.

We posit that the present study offers the first evidence illustrating how people convey sexual identity images for self-presentational aims. Specifically, with all else equal, participants self-presented a more reserved image. The exception to this pattern emerged in the public high accountability condition; that is, when participants in the public condition interacted with a friend they expressed a less-reserved sexual identity image.4

Following this latter finding, participants in the public condition also expressed greater accountability concerns with friends compared to strangers, a finding that heretofore has been absent from the literature. In fact, perceived accountability mediated the interactive effect of self-presentation and accountability on sexual attitudes, again a finding that has yet to be demonstrated in prior work. In short, these findings extend earlier work by demonstrating that accountability concerns underlie participants’ motivation to differentially self-present to friends and strangers. Although Tice et al. (1995) speculated that accountability may be the cause for such a difference, there was no empirical support for this proposition until the present results.

Moreover, sexual attitudes/experiences offer a unique area for self-presentation research that substantively extends Tice et al.'s (1995) earlier work, which involved very broad self-presentations that focused on people's general self-enhancements. Why different? Because sexual information is more intimate in nature, is strongly influenced by social norms, and is personally meaningful for most people. Indeed, Tice et al. noted that future work would do well to examine how self-presentations outside of general self-enhancements may differ as function of audience. In short, we undertook this challenge in the current study and aimed our examination at an area that is not typically considered to be in the purview of self-presentation and, to our knowledge, one that has never been empirically examined in a self-presentational context.

Some may argue that the current results do no more than simply offer evidence for something that people can already deduce based on their individual social experiences. In other words, people should be aware that their responses would differ depending on whether they were interacting with a friend or a stranger. However, we have pilot data that suggest otherwise. Simply put, people were directly asked to rate how they would express their sexual attitudes to a friend or stranger, and the results indicated no differences. Thus, although people may presume that their sexual attitudes stay constant irrespective of the situation, the current results indicate the influence of self-presentational motives, in that people adjust their sexual identity images as a function of whether the audience is a friend or a stranger. Hence we propose that our results go beyond people's intuitive understanding or beliefs concerning how they would self-present sexual identity images.

Although we utilized female participants to examine the self-presentation and accountability process, it will be valuable for future research to extend the current effect by including male participants, who were excluded from this study for reasons of control. In addition, future work would be advised to examine cross-sex dyads of both friends and strangers. In this case, one could make an argument that males and females may self-present sexual identity images differently not only as a function of whether they are talking to a friend or a stranger, but also whether the interlocutor is male or female. Doing so will allow us to examine potential boundary conditions where people may feel more or less constrained with sharing certain types of information with the same or opposite sex partners.

Moreover, because sexuality is one of the most morally and culturally regulated topics, it will be important to examine the current effect in light of potential cross-cultural differences. Although the relationship between culture and the self-presentation of sexual identity images was not the focus of the current study, some of our data suggest that there may be some similarity in sexual attitudes for females from different cultures. For example, although the sample of cultures was limited, the results from the present study and earlier pilot work indicated that there was no significant mean difference on the sexual attitudes permissiveness scale for female participants from the U.S., Western Europe, and Asia. This finding is consistent with data from earlier research (S. Hendrick & C. Hendrick, 1987), which also reported no mean differences on the sexual attitudes permissiveness scale between U.S. and international students. Thus it may be reasonable, given that there were no overall mean differences, to expect that a pattern similar to the current effect might emerge in different cultures. That is, females may convey a more permissive sexual identity to their friends compared to strangers. Clearly we are speculating and further research is needed to more closely examine the issue of potential cultural differences. Likewise, there would be benefit in examining whether the current effects vary as a function of sexual orientation, age, and audience (i.e., a more distinct audience, for example, a potential romantic partner, rather than the more global notion of friends versus strangers).

In conclusion, the current results offer initial evidence showing that people at times self-present a more reserved sexual identity image, although they tend to do so when accountability is relatively minimal; in fact, perceived accountability mediated the interactive effect of self-presentation and accountability on sexual identity.

Notes

¹ We focused on the self-presentation of attitudes toward sex in general, rather than on more specific or emotionally laden issues (e.g., sexual orientation, HIV disclosure, sexual abuse).

² Details available from corresponding author.

³ We suspect that similar means for the public:friend, private:friend, and private:stranger conditions were due to a relative ceiling effect where there is a modest limit to how “permissive” people actually are.

⁴ Although intimacy motives may have been at play, we posit that in the current context such motives were less salient, with the results suggesting that accountability concerns were much more prominent.