Professional Documents
Culture Documents
a r t i c l e i n f o a b s t r a c t
Article history: This study analyzed the relationship between adolescents' personality types and their preference for online
Received 30 June 2016 or offline self-disclosure. A sample of 1644 Chinese adolescents completed offline/online self-disclosure
Received in revised form 25 September 2016 questionnaire, and personality scale (FFI-60). Latent profile analyses were conducted to get the personality
Accepted 26 September 2016
types. Results revealed that adolescents' personality could be classified into four types: Overcontrollers, resilients,
Available online xxxx
undercontrollers, and the average. Adolescents with a preference for offline self-disclosure showed a high prob-
Keywords:
ability of being resilients, and adolescents with a preference for online self-disclosure showed a high probability
Self-disclosure preference to be undercontrollers and overcontrollers. The implications and limitations are discussed.
Personality type © 2016 Published by Elsevier Ltd.
Adolescents
Person-oriented analysis
1. Introduction Leikas & Salmela-Aro, 2014), and also have been validated consistently
across the cultures and ages (e.g. Specht, Luhmann, & Geiser, 2014).
Personality traits emphasizing single personality dimensions such as Self-disclosure is associated with many aspects of adolescents' psy-
extraversion or agreeableness, are variable oriented, whereas personality chosocial development (Xie, Sun, & Zhou, 2013). Offline self-disclosure
types emphasizing the combination of different personality dimensions is defined as the sharing of thoughts, feelings and experiences with
such as high extraversion and low agreeableness, are person-oriented. close friends in face-to-face settings (Valkenburg, Sumter, & Peter,
Base on Block's (1971) theory of ego resiliency and ego control, 2011). In recent years, the proliferation of computers and smart devices
Robins, John, Caspi, Moffitt, and Stouthamer-Loeber (1996) classified has been accompanied by an increase in online self-disclosure among
the Big Five traits into three personality types such as resilients, adolescents all over the world. As the extension of the definition of tra-
overcontrollers and undercontrollers. The resilients tend to show low ditional offline self-disclosure, online self-disclosure can be defined as a
neuroticism and average or high levels of extraversion, openness to ex- variety of cyber behaviors (e.g. instant communication, post in micro-
perience, agreeableness, and conscientiousness. The overcontrollers are blog) used to convey information and to maintain communication or
characterized by high neuroticism, agreeableness, and conscientious- to satisfy social needs in cyber space (Xie et al., 2013).
ness, and high extraversion. The undercontrollers generally display Disclosure preferences are important for understanding both online
low agreeableness and conscientiousness and high extraversion. Prior and offline social behavior. Adolescents tend to be absorbed in excessive
research demonstrated that resilients were generally well adjusted, online communication given the three most robust adolescent behav-
whereas overcontrollers were vulnerable to internalizing symptoms ioral changes: Increased risk taking, increased sensation seeking, and a
and undercontrollers displayed the most externalizing symptoms move away from parent toward greater peer affiliation (Giedd, 2012).
(Van den Akker, Deković, Asscher, Shiner, & Prinzie, 2013). These Thus, updated theories and research are required to understand how
three types of personality have been established in many studies (e.g. adolescents' high engagement with the virtual world (e.g. too much on-
line disclosure but lack of offline disclosure) may influence their devel-
opment (George & Odgers, 2015). However, as Koo, Woo, Yang, and
Kwon (2015) noted, many studies on social interaction in online space
⁎ Corresponding author.
E-mail address: 943963988@qq.com (M.-Z. Wang).
have not considered the context of social interactions in offline space.
1
These two authors contributed equally to the study. They should be regarded as joint According to Internet paradox (Kraut et al., 1998), excessive use of
first authors. online communication was associated with declines in individuals'
http://dx.doi.org/10.1016/j.paid.2016.09.048
0191-8869/© 2016 Published by Elsevier Ltd.
176 W. Chen et al. / Personality and Individual Differences 105 (2017) 175–178
offline communication with family members or friends. Just as Turkle 2.2.3. Offline self-disclosure
(2011) stated, too much online interaction would interfere with indi- The present study adjusted the items used to measure adolescents'
viduals' (especially young people's) ability to achieve closeness in online self-disclosure (Li et al., 2006) by changing the reference to the
offline relationships. Koo et al. (2015) found that online social behaviors Internet to a reference to in-person communication (see Valkenburg
decreased well-being under conditions of low engagement in offline et al., 2011). We asked adolescents “How much do you usually tell
social behaviors. In contrast, a higher level of online social behavior your close friends about your (personal habits, worst fears, secrets,
predicted greater well-being among individuals with high engagement etc.) in face-to-face meetings (offline)?” Items were rated on a five-
in offline social behaviors. Adolescents who spend too much time on point Likert scale (1 = never tell to others to 5 = tell others in detail).
online disclosure are characterized as spending time “alone together” We found good levels of reliability and validity for the offline self-disclo-
and missing out on important offline disclosure and socialization sure to friends (Cronbach's α = 0.93, χ2/df = 8.03, CFI = 0.98,
experiences. RMSEA = 0.065).
Van den Akker et al. (2013) suggested that overcontrollers were vul-
nerable to internalizing symptoms, and undercontrollers exhibited the 3. Results
most externalizing symptoms. Overcontrollers and undercontrollers
were maladjusted. Cyber space provides an alternative environment 3.1. Descriptive results
for self-disclosure, especially for maladjusted individuals (e.g. low self-
esteem, social withdrawal and anxiety) (Joinson, McKenna, Postmes, & The first step in the analyses was to determine adolescents' self-
Reips, 2007). The properties of online space, such as the absence of disclosure preference. Consistent with Valkenburg et al. (2011), we
physical presence, permission for delayed feedback, and greater infor- subtracted the offline self-disclosure scores from the online self-disclo-
mation manipulation, can help maladjusted adolescents perceive online sure scores, creating a variable that ranged from −4.00 to 4.00. We then
space as a viable, safe, and efficient context to meet friends, thus en- calculated the standard error (SE) of this variable and multiplied this SE
abling them to engage more in online than face-to-face interactions (0.025) by 1.96 to create a 95% confidence interval (CI; − 0.049 to
(Joinson et al., 2007). Therefore, we hypothesize that adolescents who 0.049) for the middle score of the new categorical variable. The newly
prefer online self-disclosure are more likely to be undercontrollers or formed variable had three levels: (a) the lowest score to − 0.049
overcontrollers, whereas adolescents who prefer offline self-disclosure (adolescents who preferred offline self-disclosure), (b) − 0.049 to
are more likely to be resilients. 0.049 (adolescents with no self-disclosure preference), and (c) 0.049
to the highest score (adolescents who preferred online self-disclosure).
2. Method The number for adolescents who preferred offline self-disclosure, online
self-disclosure, and had no preference were 1137 (69.20%), 74 (4.50%),
2.1. Participants and procedure 433 (26.30%), respectively.
The present research recruited participants from 36 classes dis- 3.2. Results of person-oriented analyses
persed over 12 middle schools in southwest China. Participants were
1644 adolescents (754 were girls) whose age ranged from 10 to Before conducting the person-oriented analyses we obtained classi-
16 years old (Mage = 13.21 ± 1.23 years). The investigation was con- fications based on personality traits. According to Van den Akker et al.
ducted in classrooms after informed consent was obtained from school (2013), the scores of each personality trait were first standardized. La-
authorities and the participants. The study was approved by the Ethical tent profile analysis was then executed to analyze the emerging person-
Committee for Scientific Research. ality types. We next compared the 2-class, 3-class, 4-class, and 5-class
models, and the 4-class model was chosen to be the most appropriate
2.2. Measures model of adolescents' personality types. Table 1 showed the results of
the model comparison, and Fig. 1 showed the personality type profile.
2.2.1. The NEO Five-Factor Inventory According to the characteristics of each personality type and tenets
Adolescents' personality was assessed by the Chinese version of the by Van den Akker et al. (2013), we named these four personality
NEO Five-Factor Inventory (NEO-FFI; Nie, Lin, Zheng, Ding, & Peng, types as follows: Overcontrollers (6.20%) had high scores for neuroti-
2008), which was adapted from the original NEO Five-Factor Inventory cism, low scores for extraversion, and average scores for the remaining
(Costa & McCrae, 1992). As the most classic personality questionnaire in three traits (conscientiousness, agreeableness, and openness to experi-
China, NEO-FFI-60 has been widely used in measuring Chinese adoles- ence); resilients (24.80%) had low scores for neuroticism and average
cents' personality (e.g. Nie et al., 2008; Nie, Yang, & Zeng, 2011). NEO- scores for the remaining four traits; the Average group (64.10%) had av-
FFI includes 60 items, with 12 items for each of the five dimensions. erage scores for all the five personality traits; and undercontrollers
Participants rated the items on a five-point Likert scale (1 = completely (4.90%) had low scores for conscientiousness and agreeableness, and
not true to 5 = completely true). In the present study, the Cronbach's α average scores for the remaining three traits (Table 2).
coefficients for extraversion, agreeableness, conscientiousness, neuroti- Multiple logistic regression was conducted to analyze the relation-
cism, and openness were 0.72, 0.64, 0.79, 0.78, and 0.60 respectively. ship between self-disclosure preference and personality types. Because
the largest group of participants had average scores on all the five traits,
2.2.2. Online self-disclosure the average personality type was then taken as the reference group. The
The 10-item Self-disclosure Index questionnaire (Miller, Berg, &
Archer, 1983) was adapted to assess adolescents' online self-disclosure
to close friends (Li et al., 2006). Respondents were asked “How much do Table 1
Criteria for latent profile models of personality types.
you usually tell your close friends about your personal habits, worst
fears, secrets, etc. when you are on the Internet (online)?” Items were Model BIC SSBIC Entropy LMR-LRT (p-value) BLRT (p-value)
rated on a five-point Likert scale (1 = never tell to others to 5 = tell 2-class 22,629.03 22,578.20 0.68 b0.001 b0.001
others in detail), with higher summary scores on the total scale indicat- 3-class 22,561.73 22,491.84 0.71 0.04 b0.001
ing greater levels of self-disclosure. We found satisfactory reliability and 4-class 22,516.57 22,427.62 0.70 0.03 b0.001
validity for the online self-disclosure to close friends (Cronbach's α = 5-class 22,497.45 22,389.45 0.70 0.06 b0.001
4. Discussion
Acknowledgments Leikas, S., & Salmela-Aro, K. (2014). Personality types during transition to young adult-
hood: How are they related to life situation and well-being? Journal of Adolescence,
5, 753–762.
This work was supported by Education Youth Project of National Li, D., Yu, M., Wang, C., Xie, X., Zhou, L., & Zhu, X. (2006). Self-disclosure and self-conceal-
Social Science Foundation of P. R. China [Project name: The impact of ment in adolescence and the relationship between them and subjective well-being.
Psychol. Dev. Educ., 22(4), 83–90.
mobile social network on adolescent self-control; project number: Miller, L. C., Berg, J. H., & Archer, R. L. (1983). Openers: Individuals who elicit intimate self-
CBA150157]. No competing financial interests existed. disclosure. Journal of Personality and Social Psychology, 44, 1234–1244.
Nie, Y., Lin, C., Zheng, X., Ding, L., & Peng, Y. (2008). The relationship between adolescent
social adaptive behavior and the “big five” personality. Psychological Science, 31(4),
References 774–779.
Nie, Y., Yang, A., & Zeng, M. (2011). The relationship among adolescents' meta-cognition,
Block, J. (1971). Lives through time. Berkeley: Bancroft.
big five personality and academic adaptive behavior. Psychol. Dev. Educ., 27(2),
Caplan, S. E. (2005). A social skill account of problematic internet use. Journal of
151–157.
Communication, 55, 721–736.
Peterson, R. A., & Kim, Y. (2013). On the relationship between coefficient alpha and com-
Chang, L., Mak, M. C. K., Li, T., Wu, B. P., Chen, B. B., & Lu, H. J. (2011). Cultural adaptations
posite reliability. Journal of Applied Psychology, 98(1), 194–198.
to environmental variability: an evolutionary account of east–west differences.
Robins, R. W., John, O. P., Caspi, A., Moffitt, T. E., & Stouthamer-Loeber, M. (1996). Resil-
Educational Psychology Review, 23, 99–129.
ient, overcontrolled, and undercontrolled boys: Three replicable personality types.
Costa, P. T., Jr., & McCrae, R. R. (1992). The NEO-PI-R: Professional manual. Odessa, FL: Psy-
Journal of Personality and Social Psychology, 70, 157–171.
chological Assessment Resources.
Specht, J., Luhmann, M., & Geiser, C. (2014). On the consistency of personality types across
De Clercq, B., Rettew, D., Althoff, R. R., & De Bolle, M. (2012). Childhood personality types:
adulthood: Latent profile analyses in two large-scale panel studies. Journal of
Vulnerability and adaptation over time. Journal of Child Psychology and Psychiatry,
Personality and Social Psychology, 3, 540–556.
53(6), 716–722.
Turkle, S. (2011). Alone together: Why we expect more from technology and less from each
George, M. J., & Odgers, C. L. (2015). Seven fears and the science of how mobile technol-
other. New York, NY: Basic Books.
ogies may be influencing adolescents in the digital age. Perspectives on Psychological
Valkenburg, P. M., Sumter, S., & Peter, J. (2011). Gender differences in online and offline
Science, 10(6), 832–851.
self-disclosure in pre-adolescence and adolescence. British Journal of Developmental
Giedd, J. N. (2012). The digital revolution and adolescent brain evolution. Journal of
Psychology, 29, 253–269.
Adolescent Health, 51, 101–105.
Van den Akker, A. L., Deković, M., Asscher, J. J., Shiner, R. L., & Prinzie, P. (2013). Personal-
Joinson, A. N., McKenna, K. Y. A., Postmes, T., & Reips, U. (2007). The Oxford handbook of
ity types in childhood: Relations to latent trajectory classes of problem behavior and
internet psychology. Oxford: Oxford University Press.
over reactive parenting across the transition into adolescence. Journal of Personality
Koo, H. J., Woo, S., Yang, E., & Kwon, J. H. (2015). The double meaning of online social
and Social Psychology, 4, 750–764.
space: Three-way interactions among social anxiety, online social behavior, and
Xie, X. C., Sun, X. J., & Zhou, Z. K. (2013). The type, function and influencing factors of on-
offline social behavior. Cyberpsychology, Behavior and Social Networking, 9, 514–520.
line self-disclosure. Advances in Psychological Science, 2, 272–281.
Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T., & Scherlis, W. (1998).
Internet paradox. A social technology that reduces social involvement and psycholog-
ical well-being? American Psychologist, 53(9), 1017–1031.