خودارائه گری حفاظتی، منابع اجتماعی شدن و تنهایی در میان نوجوانان و جوانان استرالیا

Abstract This research assessed the relative impact of features of protective self-presentation, key sources of socialization, and social anxiety on individual differences in loneliness among Australian adolescents and young adults. In Study 1, 281 high school students living with parents completed self-report measures of loneliness, self-presentation features (i.e., fear of negative evaluation, social competence), parenting style (overprotection, care), peer relations (attachment, teasing) and social anxiety. Peer and parenting measures were significant predictors of loneliness, independent of self presentation influences. Social anxiety also added to the model after controlling for all other measures. In Study 2, 170 undergraduates living independently completed measures of the same constructs. Paralleling findings from Study 1, protective self-presentation features, peer relations, and social anxiety had unique effects on loneliness. As hypothesized, however, parenting measures did not add to the prediction model for young adults. Implications of findings are discussed in relation to understanding loneliness within high school and university age samples.

1. Introduction Loneliness is a common, distressing experience that has deleterious effects on psychological functioning and physical health (e.g., Cacioppo et al., 2002 and Heinrich and Gullone, 2006). Although self-presentation models (e.g., Arkin et al., 1986 and Schlencker and Leary, 1983) have aided in articulating motivational bases of social anxiety, a cognitive-affective experience triggered by perception of others’ possible evaluations (Leary & Kowalski, 1995) and shyness, a related syndrome characterized by social anxiety and behavioral inhibition in interpersonal situations (Leary, 1983), recent extensions suggest self-presentational factors also contribute to individual differences in loneliness (e.g., Cacioppo and Hawkley, 2005, Dill and Anderson, 1999 and Jackson et al., 2002). Self-presentation theorists posit that people are typically motivated to gain social approval in interpersonal encounters and act in ways that might attract attention and engender recognition (Arkin et al., 1986). In contrast, the socially anxious have doubts about their interpersonal competencies and capacities to create favorable impressions. They adopt a protective self-presentation style, wherein the overarching goal of social interaction is not to win approval from others but to avoid their disapproval. Tendencies to view interpersonal encounters as threats (e.g., Jackson and Ebnet, 2006 and Jackson and Eglitis, 2005), others as rejecting (e.g., Jackson et al., 1997, Meleschko and Alden, 1993 and Wallace and Alden, 1997), and the self as less socially skilled (e.g., Wallace & Alden, 1991) may maintain or increase social anxiety and/or avoidance of social interaction. If this pattern results in having fewer social contacts than one desires, loneliness has emerged (Dill & Anderson, 1999). Indeed, shyness and social anxiety (e.g., Jackson et al., 2000 and Jones et al., 1990), fear of disapproval (e.g., Jackson et al., 2002 and Leary et al., 2001) and perceptions of social incompetence (e.g., Jackson et al., 2002 and Segrin and Flora, 2000), even in the absence of actual deficits (e.g., Christensen & Kashy, 1998), contribute to loneliness. To date, self-presentation models have highlighted actors’ motives and responses on the stages of social life. With few exceptions (e.g., Jackson and Ebnet, 2006 and Meleschko and Alden, 1993), past studies examined perceptions and responses of actors to people in general rather than to specific “audiences” of self-presentation. As a result, little is known about the degree to which particular audiences or socialization influences contribute to loneliness, beyond features of protective self-presentation. Given that parents and peers are key socialization agents in childhood and adolescence (e.g., Shaffer, 2001), and loneliness is more prevalent among adolescents than any other segment of the population ( Perlman & Landolt, 1999), parenting and peer experiences may be particularly relevant to understanding loneliness in this age group. With respect to parenting influences, psychodynamically-rooted researchers argue that disrupted early attachments have adverse effects on mental representations of the self and others; such disturbances interfere with the formation of subsequent attachment relationships and increase risk for loneliness (e.g., Cassidy and Berlin, 1999, Fromm-Reichmann, 1959 and Hojat, 1987). Indeed, lower levels of secure attachment have been observed among lonely children (e.g., de Minzi, 2006) and young adults (e.g., Wiseman, Mayseless, & Sharabany, 2006). Conversely, loneliness has positive correlations with ambivalent and avoidant attachments to parents (Wiseman et al., 2006) as well as parenting behaviours such as overprotection (Terrell, Terrell, & Von Drashek, 2000) and inattention (Antognoli-Toland, 2001). Other studies have highlighted peer group correlates of loneliness. Parker and Asher (1993) found that being poorly accepted by peers, lacking a friend, and having unfulfilling friendships contribute independently to loneliness. Teasing (e.g., Storch et al., 2003 and Strawser et al., 2005), rejection (Crick & Ladd, 1993) and reciprocal distrust (Rotenberg, MacDonald, & King, 2004) are other peer correlates of loneliness. Unfortunately, because these studies examined either parenting or peer influences on loneliness, their relative effects are not clear. Although two studies have directly addressed this issue, each has notable limitations. Chipuer (2001) found that attachment to best friend but not to parents predicted emotional and social loneliness among fifth and sixth grade children, after statistically controlling for other forms of loneliness. The stability of these effects was of concern, however, give that the pattern was not replicated on school, neighborhood, or global measures of loneliness. More recently, Bogaerts, Vanheule, and Desmet (2006) linked loneliness with attachments to peers but not to parental overprotection or care among university students. Use of a single, non-validated item of loneliness also raises potential reliability concerns. If firm conclusions cannot be drawn from these studies, plausible hypotheses about the relative impact of parenting versus peer factors can be generated from other sources. Harris (1995) argues that environmental effects of parenting on later personality development are modest relative to genetic influences, and do not transfer automatically to other contexts and predominate there. Because children enact their adult lives outside the home, experiences in peer groups outside the family of origin have a greater bearing on later personality development than parenting does (Harris, 1995). Although these ideas were not discussed specifically in relation to loneliness, Rapee and Spence (2004) recently proposed that environmental effects on social anxiety, a reliable correlate of loneliness, are minor and temporary. When a relevant environmental influence stops, social anxiety returns to a set point, barring other environmental effects. As such, contributions of parenting to social anxiety in offspring should be strongest when the child is living in the parental home but will diminish after the child has left home. Study 1 attempted to clarify the relative impact of protective self-presentation, parenting, peer relations, and social anxiety on individual differences in loneliness among Australian high school students living with parents. Based on work with American university students (Jackson et al., 1997 and Jackson et al., 2002), facets of protective self-presentation (i.e., heightened disapproval concerns, perceptions of reduced social competence) were expected to contribute to loneliness in the sample. Second, peer measures (increased teasing, low peer attachment) were expected to add to the prediction model, beyond self-presentational influences. Furthermore, given that respondents were living with parents/guardian, it was hypothesized that appraisals of parenting (heightened overprotection, reduced care) would contribute to loneliness, after peer factors had been controlled. Finally, consistent with Jackson et al. (2002), social anxiety was expected to influence loneliness independent of all other measures.

6. Results Loneliness was not related to gender, t = 1.20, p < .27, age, r = −.03, p < .67 or employment status, t = 1.67, p < .10. Correlations between loneliness and other measures were highly significant (p < .005 or above) except for Parental Care and Overprotection ( Table 3). In this sample, the seven predictors combined for Adj. R2 = .55 of the variance in loneliness [F (7, 162) = 30.08, p < .0001]. Features of protective self-presentation were highly related to loneliness. Peer factors added to the model, albeit effects were attenuated ( Table 4). In contrast, parenting was not related to loneliness. Social anxiety had a modest effect, beyond other factors. In an alternate regression model, parenting factors had little effect on loneliness even when entered in Step 2 of the equation, Adj. R2 = .005 [F Ch. (2, 165) = .69, p < .50]; in contrast, peer influences were undiminished, Adj. R2 = .16 [F Ch. (2, 163) = 18.66, p < .001]. Table 3. Intercorrelations between loneliness and other research measures for the university sample (N = 170) 2 3 4 5 6 7 8 1. Loneliness .59⁎⁎ −.55⁎⁎ −.53⁎⁎ .23⁎ −.19 .17 .56⁎⁎ 2. Fear of negative evaluation −.48⁎⁎ −.28⁎⁎ .31⁎⁎ −.15 .21⁎ .59⁎⁎ 3. Social competence .30⁎⁎ −.29⁎⁎ .17 −.23⁎ −.59⁎⁎ 4. Peer attachment −.22⁎ .23⁎ −.15 −.28⁎⁎ 5. Peer teasing −.31⁎⁎ .28⁎⁎ .30⁎⁎ 6. Parent care −.40⁎⁎ −.17 7. Parent overprotection .29⁎⁎ 8. Social anxiety – ⁎ p < .01. ⁎⁎ p < .001. Table options Table 4. Predictors of loneliness in the university sample Step Measure β sr2 t p 1 Fear of negative evaluation .43 .38 6.55 .001 Social competence −.34 −.25 5.21 .001 Adj. R2 = .44 F Ch. (2, 167) = 63.34, p < .001 2 Peer attachment −.35 −.33 −6.01 .001 Peer teasing .04 .04 .78 .44 Adj. R2 = .54 F Ch. (2, 165) = 19.61, p < .001 3 Parent care −.02 −.02 −.370 .713 Parent overprotectiona −.02 −.02 −.369 .712 Adj. R2 = .53 F Ch. (2, 163) = .10, p < .90 4 Social anxiety .17 .12 2.35 .02 Adj. R2 = .55 F Ch. (1, 162) = 5.52, p < .02 a Note: The negative β value for Overprotection in the regression contrasts with its positive bivariate relation to loneliness. Results were consistent within steps in alternate prediction equations that removed various combinations of variables. Hence, there was no evidence for suppression of error variance in the prediction model. Table options In Study 2, effects of protective self-presentation, peer relations, and social anxiety were replicated among university students. On this basis, the hypothesis that these factors influence loneliness across the lifespan is worthy of pursuit. In contrast, parenting factors did not add to the prediction of loneliness among university students was also supported. On the one hand, this finding may illustrate how effects of parenting on loneliness do not transfer automatically to environments outside one’s family of origin. Conversely, differential effects of parenting on loneliness within the two samples may reflect differences in the recall of parenting due to cognitive and interpersonal changes of the self between adolescence and young adulthood wherein attachment-related thoughts and feeling shifted in emphasis from parents to peers.