Differences in Engaging in Sexual Disclosure Between Real Life and Cyberspace Among Adolescents: Social Penetration Model Revisited
- First Online:
- Cite this article as:
- Yang, ML., Yang, CC. & Chiou, WB. Curr Psychol (2010) 29: 144. doi:10.1007/s12144-010-9078-6
- 622 Downloads
To date, relatively few studies have begun to explore adolescents’ sexual self-disclosure in cyberspace. Rare research has taken a close look at differences in adolescents’ sexual self-disclosure occurring in real life and cyberspace. The social penetration model suggests that an individual’s level of sexual disclosure should be in accordance with relationship intimacy in real life. The current study investigated whether the effects of relationship intimacy on adolescents’ willingness to disclose sexual history differ in terms of sex and communication environment (real life vs. cyberspace). A total of 419 Taiwanese adolescents completed a survey about their willingness to communicate on different sexual topics in the contexts of varying levels of relationship intimacy. The results showed that in real life both male and female adolescents showed a parallel relationship between willingness to engage in sexual disclosure and relationship intimacy, supporting predictions according to the social penetration model. However, in cyberspace, male adolescents exhibited a greater willingness to communicate, regardless of degree of relationship intimacy, whereas females revealed a U-shape trend regarding the effect of relationship intimacy on willingness to communicate. These findings indicate that sexual disclosure on the part of adolescents in cyberspace departs from the perspective of the social penetration model.
KeywordsAdolescents Sex differences Sexuality Self-disclosure Social penetration model
Adolescent sexuality in real life has been widely explored in past research, and has focused on sexual attitudes (Werner-Wilson 1998), sexual orientation (Frankowski 2004), homosexuality (Harrison 2003), sexual knowledge (Fisher 1986), sexual behavior (Cubbin et al. 2005; Gowen et al. 2004; Rostosky et al. 2004), and sex education (Somers and Gleason 2001; Song et al. 2000). However, human sexuality, as represented on the Internet, is a growing area of research in the social sciences (Cooper 1998, 2002). Researchers have only recently begun to gather empirical data concerning online sexual activity (OSA). To date, empirical studies have focused on the variety of OSAs (Cooper et al. 1999), their categorization (Cooper and Griffin-Shelley 2002), and online sexual problems (Cooper et al. 2000, 2001; Schneider 2000). Only a few studies have addressed cybersex among adolescents (Cameron et al. 2005; Lo and Wei 2005; Wan et al. 2009). A recent study conducted by Chiou and Wan (2006) explored how sex differences and the interplay between cyberspace and real life affect sexual self-disclosure among adolescents. Chiou (2007) further examined how adolescents disclose their sexuality to cyber friends in a laboratory setting. However, the answers to the questions of whether adolescent sexual disclosure depends on the intimacy of the relationship, and whether sexual disclosure differs between real life and cyberspace, remain unclear. Therefore, the current study investigated differences in the willingness of adolescents to engage in sexual self-disclosure between real life and cyberspace, as related to relationship intimacy.
Social Penetration Model and Corresponding Sexual Disclosure
Some researchers have proposed that self-disclosure is synonymous with an intimate relationship, and that self-disclosure can function as an indicator of intimacy in interpersonal relations (Jourard 1971; Pearce and Sharp 1973). The social penetration model (Altman and Taylor 1973) suggests that interactive partners are more likely to disclose and communicate intimate information when the interpersonal interaction is turning into an intimate relationship. The relationship intimacy effect refers to a parallel relationship between willingness to self-disclose and the intimacy of interactive partners. Based on the social penetration model, it was predicted that adolescents would more willing to communicate sexual disclosures to intimate partners in real life.
However, self-presentation in cyberspace could be different to that in real life, as it is a highly anonymous and private environment. Personal identification on the Internet is highly anonymous (Wallace 2001). Anonymity can produce a deindividuated state (Diener 1979; Postmes and Spears 1998; Zimbardo 1969). Festinger et al. (1952) described deindividuation as a psychological state in which inner restraints are lost, in a situation where “individuals are not seen or paid attention to as individuals” (p. 283). Many studies have demonstrated that deindividuation is likely to motivate individuals to exhibit antinormative behavior (e.g., Ellison et al. 1995; Rehm et al. 1987; Zimbardo 1975). When Lin and Yeh (1999) investigated Internet sexuality, they found that the anonymity of the Internet ensures the privacy of users; anonymous users can confide secrets to interactive partners when having cybersex. Wallace (2001) also reported that Internet users can feel free of evaluation and criticism from others, prompting them to dare to do things that they might not in real life, due to their deindividuated state in cyberspace. Individuals appear to be more likely to engage in sexual self-disclosure in cyberspace than in real life (Chiou and Wan 2006). Hence, sexual self-disclosure on the Internet might not depend on the intimacy of a relationship as it does in real life. Adolescents might be more willing to make sexual disclosures in cyberspace because a deindividuated state has been induced.
Sex Differences in Sexual Self-Disclosure
Previous studies that have investigated how sex differences affect self-disclosure in real life have consistently found that females are more willing than males to offer full self-disclosure to others (Caldwell and Peplau 1982; Dindia and Allen 1992; Reisman 1990). Female subjects have also reported that they reveal more during self-disclosure and are more likely to be the target of others’ self-disclosure than males (Buhrke and Fuqua 1987; Snell et al. 1988). Research has shown that females self-disclose more often, and share information about more intimate topics, than do males (Davidson and Duberman 1982). However, sexuality is a highly personal and private topic for self-disclosure. Regarding sex differences in sexual self-exposure, Chiou and Wan (2006) found that male adolescents self-disclosed more often and more fully than females in real life. Hence, it was predicted that male adolescents would be more likely to engage in sexual disclosure than females.
Sex differences in relation to Internet sexuality have become a critical issue in the field of Internet psychology. Previous studies have identified sex differences in online sexual activities, sexual exploration, and expressions of sexuality (e.g., Cooper et al. 1999; Weiser 2000). Chiou and Wan (2006) found that male adolescents appeared to be more active in communicating sexuality on the Internet than females. Further, Hsueh (2001) found that self-disclosure during late adolescence differed according to sex: males self-disclosed more fully and to a greater extent than did females in cyberspace. Moreover, male adolescents were also more likely to make sexual disclosures with cyber friends than were females (Chiou 2007). These findings differ from previous studies or theories related to sex differences in real-life self-disclosure. Based on past literature regarding males’ liberal tendencies in relation to Internet sexuality, it could tentatively be argued that male adolescents would be more likely to make sexual disclosures, regardless of relationship intimacy. On the other hand, in a highly anonymous disclosure condition, female adolescents’ willingness to communicate sexual disclosure in cyberspace would be contingent upon relationship intimacy.
The adolescent participant population was stratified into three demographic areas: Northern, Central, and Southern Taiwan. Participants read a paper of agreement which also informed them of the importance of this research and were then asked to rate their willingness to make sexual self-disclosures in both real life and cyberspace, in the contexts of varying degrees of relationship intimacy and different sexual topics. The response rate for the total of 450 questionnaires was 93%. There were 419 adolescents in this sample (211 men and 208 women; aged 15 to 24 years, M = 18.23, SD = 2.31). A quota system ensured proportional representation of sex, age, and the regions of Taiwan. The resulting sample mirrored the national distributions of age and sex among Taiwanese adolescents. The current study was approved by the Institutional Review Board of the authors’ university. Background information was provided by the adolescents before participation. All participants signed consent and assent forms prior to this survey.
The Online Sexual Self-Disclosure Scale for Adolescents (OSSSA) was developed based on the method of Thurstone’s type scale (Thurstone 1925) for obtaining representative sexual topics on the Internet. This method was also utilized by Yang and Huang’s (1980) study on adolescents’ self-disclosure. The OSSSA consists of 15 sexual topics with varying levels of topic intimacy ranging from the lowest to the highest, and opinions on topics on an 11-point scale, including pornography, sexual fantasies, contraception, orgasm, sexual orientation, intercourse styles, venereal diseases, sexual harassment, cybersex, masturbation, one-night-stands, sexual transactions, erotic agents, Paraphilias, and sexual dysfunctions (See Chiou and Wan 2006, for details).
For measuring participants’ willingness to make sexual disclosures, participants were asked to rate their willingness to communicate on each topic of the OSSSA after receiving a sexual disclosure from interactive partners with varying levels of relationship intimacy (i.e., initial acquaintance = low intimacy; friends = medium intimacy; lovers = high intimacy). Their responses were assessed on an 11-point scale that ranged from 1 (very likely) to 11 (least likely). The 15 topics of the OSSSA were randomly arranged. The order of participants’ evaluations from the two kinds of communication environments was counter-balanced. Half of the participants responded on their willingness to communicate on the 15 sexual topics in real life followed by those topics in cyberspace, whereas the other half responded in the opposite order.
Participants’ mean scores of willingness to communicate across the 15 sexual topics were later computed in terms of relationship intimacy and communication environment. Higher scores indicated that participants exhibited greater willingness to communicate in tandem with an interactive partners’ sexual disclosure. In real life, the internal consistencies (Cronbach’s Alpha) were .91 for responses in low-intimacy relationships, .90 in the medium-intimacy relationships, and .94 in the high-intimacy relationships, respectively. In cyberspace, the internal consistencies were .89 for responses in the low-intimacy relationships, .92 in the medium-intimacy relationships, and .93 in the high-intimacy relationships, respectively.
Participants’ willingness to engage in sexual disclosure by sex, communication environment, and relationship intimacy
Communication environment by sex
Female (n = 208)
Male (n = 211)
A significant three-way interaction was found between sex, communication environment, and relationship intimacy (F(2, 834) = 61.32, p < .001, η2 = .13). This interaction was independent of the order of presentation of communication environment (F(2, 830) = 0.51, p > .05). Subsequent simple interaction analyses revealed that a two-way interaction of communication environment and relationship intimacy was significant for males (F(2, 420) = 60.79, p < .001, η2 = .22) and the interaction was also significant for females (F(2, 414) = 429.51, p < .001, η2 = .68).
For male participants’ willingness to communicate in real life, follow-up trend analysis showed that the ascending linear trend of relationship intimacy effect (Mreal life-low intimacy = 4.70, SD = 2.08; Mreal life-medium intimacy = 6.02, SD = 2.27; Mreal life-high intimacy = 6.84, SD = 2.14) was prominent when communicating sexual disclosures in real life (F(1, 210) = 116.88, p < .001, η2 = .36). Specifically, male participants were more likely to engage in communicating sexual disclosure with interactive partners with greater degrees of relationship intimacy. On the other hand, the relationship intimacy effect was not significant (F(2, 420) = 1.96, p > .05) when male participants engaged in communicating sexual disclosure in cyberspace (Mcyberspace-low intimacy = 7.52, SD = 2.17; Mcyberspace-medium intimacy = 7.26, SD = 2.19; Mcyberspace-high intimacy = 7.60, SD = 1.97). Regardless of relationship intimacy, male participants were more likely to engage in sexual disclosure with interactive partners in cyberspace than in real life at low levels of relationship intimacy (t(210) = 15.92, p < .001, Cohen’s d = 1.10), at medium levels of topic intimacy (t(210) = 10.11. p < .001, Cohen’s d = 0.69), and at high levels of topic intimacy (t(210) = 5.87p < .001, Cohen’s d = 0.41). The effect size between two means is reflected by the d statistic. Cohen refers to a d value of .20 as a small effect size. A medium effect size is one for which d = .50, and a larger effect size is one for which d = .80 (Cohen 1992).
For female participants, the ascending linear trend of relationship intimacy effect (Mreal life-low intimacy = 2.07, SD = 1.45; Mreal life-medium intimacy = 3.30, SD = 1.29; Mreal life-high intimacy = 4.23, SD = 1.27) was pronounced in communicating sexual disclosure in real life, a finding similar to that observed in male participants (F(1, 207) = 693.38, p < .001, η2 = .77). Female participants were more willing to make sexual disclosures with interactive partners and with greater relationship intimacy. Both male and female adolescents showed that the link of relationship intimacy with willingness to communicate sexual disclosure in real life, supporting the hypothesis from the social penetration model. However, a quadratic trend of the effect of relationship intimacy on female participants’ willingness to communicate sexual disclosure in cyberspace (Mcyberspace-low intimacy = 6.17, SD = 1.98; Mcyberspace-medium intimacy = 3.76, SD = 1.92; Mcyberspace-high intimacy = 6.58, SD = 2.43) was observed (F(1, 207) = 2495.64, p < .001, η2 = .92). Compared to their willingness to communicate in real life, female participants appeared more likely to engage in sexual disclosure when relationship intimacy was either at a low or high level (t(207) = 22.66 for low-intimacy relationships, p < .001, Cohen’s d = 1.57; t(207) = 12.20 for high-intimacy relationships, p < .001, Cohen’s d = 0.85). Differential impact of relationship intimacy on willingness to communicate sexual disclosure in cyberspace suggests that females tend to adopt a different self-presentation strategy.
Supplementary analysis with regard to sex differences in adolescents’ corresponding sexual disclosure revealed that male participants’ willingness to communicate was significantly greater than that of females, regardless of communication context and relationship intimacy. In real life, male participants were more willing to communicate sexual disclosure than females (t(417) = 14.99 for the condition of low-intimacy relationships in real life, p < .001, Cohen’s d = 0.72; t(417) = 15.09 for the condition of medium-intimacy relationships in real life, p < .001, Cohen’s d = 0.74; t(417) = 15.14 for the condition of high-intimacy relationships in real life, p < .001, Cohen’s d = 0.75). A similar pattern of sex differences was also found in cyberspace data (t(417) = 6.62 for the condition of low-intimacy relationships in cyberspace, p < .001, Cohen’s d = 0.33; t(417) = 17.31 for the condition of medium-intimacy relationships in cyberspace, p < .001, Cohen’s d = 0.86; t(417) = 4.72 for the condition of high-intimacy relationships in cyberspace, p < .001, Cohen’s d = 0.23).
According to the social penetration model, the individuals’ level of sexual disclosure with their interactive partners should be parallel to relationship intimacy in real life. As predicted, both male and female participants were more likely to communicate sexual disclosure as the relationship increased in intimacy, which was congruent with the social penetration model. The findings suggest that the social penetration model is corroborative with adolescents’ corresponding sexual disclosure in real life.
However, based on the deindividuated condition induced in cyberspace, it was assumed that the effects of relationship intimacy on adolescents’ willingness to engage in sexual disclosure would differ between real life and cyberspace. Data on the adolescent participants’ sexual self-disclosure in cyberspace revealed a different pattern to that in real life. Independent of the effects of relationship intimacy, male participants showed a greater willingness to communicate sexual disclosure. This finding indicates that male adolescents were more likely to communicate sexual disclosures in cyberspace than in real life, and further suggests that male adolescents appear to adopt a liberal and noncontingent strategy of self-presentation with regard to Internet sexuality when engaging in sexual disclosure. For male adolescents, the relationship intimacy effect proposed by the social penetration model could not account for their level of sexual self-disclosure in cyberspace. Chiou (2007) found out that the higher the level of anonymity, the higher adolescents’ level of intent to make sexual disclosures would be. Male adolescents’ corresponding sexual disclosure could be strongly related to the deindividuated state induced by cyberspace.
Like male participants, female participants also exhibited different patterns with respect to communicating sexual disclosure in real life and the cyberspace. The effects of relationship intimacy on their willingness to communicate showed a U-shape trend in cyberspace. Female participants were more likely to correspond with interactive partners with low and high intimacy levels than those with medium intimacy. More interestingly, the major difference in female participants’ sexual self-disclosure between the two kinds of communication environment was observed at low levels of relationship intimacy. Anonymity and the deindividuated state induced by cyberspace could account for these findings. Female participants’ greater willingness to communicate with initial acquaintances (i.e., the low-intimacy relationship) could result from the anonymity the feel in cyberspace. When interacting with friends (i.e., the medium-intimacy relationship), the attenuated or nonexistent anonymity could result in reducing their willingness to communicate sexual disclosure. As for corresponding with lovers (i.e., the high-intimacy relationship), the deindividuated state induced by cyberspace could motivate them to be more likely to engage in sexual disclosure on the Internet than in real life. Similar to data for male participants, female participants’ willingness to communicate in cyberspace challenges the social penetration model, which suggests a parallel relationship between relationship intimacy and willingness to engage in sexual disclosure. Female adolescents’ sexual self-disclosure in cyberspace was congruent with the arguments about the effects of anonymity and the deindividuated state on Internet users’ self-presentation in cyberspace (Wallace 2001).
According to the social penetration model, the individuals’ level of sexual disclosure with their interactive partners should be parallel to relationship intimacy in real life. As predicted, both male and female participants were more likely to engage in sexual disclosure as the relationship increased in intimacy, which was congruent with the social penetration model. The findings suggest that the social penetration model is corroborative with adolescents’ level of sexual disclosure in real life.
In terms of sex differences, male participants exhibited a greater willingness than female participants to engage in sexual disclosure regardless of relationship intimacy, which was in accordance with previous findings related to sex differences in Internet sexuality (Chiou and Wan 2006; Chiou 2007; Cooper et al. 2000). Previous studies have revealed that, compared to males, females are more willing to engage in a deeper level of self-exposure, particularly about intimate topics (e.g., Dindia and Allen 1992; Reisman 1990; Yang and Huang 1980). However, our data indicate that, in contrast, the male adolescents made sexual disclosures in a more liberal manner than did the females. In addition, the findings on sex differences suggest that male adolescents are more active than females in exchanging sexual self-disclosure with cyber partners. Female adolescents tend to be more sexually constrained or repressed than males, even in the deindividuated and private world of cyberspace.
Limitations and Future Directions
This study had certain limitations. First, interpretation of results requires caution because of the cohort or generation effect. Our research sample consisted of a particular cohort, which could limit the ability to generalize our results. In addition, all the participants were Taiwanese adolescents, and cultural differences across countries and ethnicities may affect the development of sexual attitudes and practices. Any attempt to apply or generalize these findings should consider cultural differences in relation to sexual concepts and values. Finally, a cross-sectional survey was not able to reveal the dynamic processes of the relationships between partners’ self-disclosure and the development of relationships.
The current study found sex differences in relation to the effects of relationship intimacy on adolescents’ level of sexual self-disclosure. Future research may clarify how social factors affect these differences (e.g., gender roles, sex stereotypes, sexual repression, religiosity, sexist attitudes, etc.). Second, a recent study conducted by Liau et al. (2008) indicated that parents tend to underestimate adolescents’ engagement in risky Internet behaviors. Parental awareness and monitoring of adolescent Internet sexuality is worthy of further investigation. Third, interpersonal self-disclosure can be influenced by the development of a relationship (Derlega et al. 1993). A longitudinal design combined with dyad analysis could reveal how relationship factors (e.g., relationship satisfaction and relationship expectations) affect the dynamics of adolescents’ sexual self-disclosure in cyberspace. Finally, previous studies have revealed that females are more likely than males to self-disclose in real life (Collins and Miller 1994; Reis et al. 1985; Snell et al. 1988). The female-female dyad was the most frequently observed pair to engage in self-disclosure, whereas the male-male dyad was least likely to engage in self-disclosure on intimate topics (Dolgin and Minowa 1997; Dolgin et al. 1991). Future research may investigate how the sex composition of interactive pairs affects adolescent sexual self-disclosure. This may provide insights into whether the self-presentation strategies of adolescents in relation to Internet sexuality have distinct patterns based on the sex composition of the interacting pair.
In sum, the present study showed that adolescents’ sexual self-disclosure differs in real life and in cyberspace. The relationship intimacy effect proposed by the social penetration model was observed in both male and female adolescents’ willingness to engage in sexual disclosure in real life. However, this effect was not found in cyberspace. Male adolescents consistently exhibited a greater willingness to engage in sexual disclosure, independently of the degree of relationship intimacy. Female adolescents showed a U-shape trend in the effects of relationship intimacy on their willingness to communicate. Our findings in relation to adolescents’ sexual self-disclosure in cyberspace reveal that anonymity and deindividuation, rather than the relationship intimacy proposed by the social penetration model, appear to affect adolescents’ willingness to exchange sexual self-disclosure with cyber partners.
The authors would like to thank the National Science Council of the Republic of China for financially supporting this research (Project No. NSC 92-2815-C-328-006-S). We also thank Wang Siang for data collection.