Socially Anxious Individuals Get a Second Chance After Being Disliked at First Sight: The Role of Self-Disclosure in the Development of Likeability in Sequential Social Contact.

"Self-Disclosure Behaviour Scale The Self-disclosure Behaviour Scale was composed out of four scales which were rated by all three video-raters on a 9-point Likert scale: (1) positive self-disclosure (e.g., “How much did you get to know about the participant concerning superficial but positive topics, “To what extent did the participant show positive emotions?”) (2) negative self-disclosure (e.g., “How much did the participant tell about superficial but negative topics”, “In what extent did the participant share about things she had difficulties with”, “To what extent did the participant show negative emotions”); (3) eliciting self-disclosure in the confederate (e.g., “When the participant asked questions: to what extent did the participant show genuine interest when she asked the confederate questions”); (4) responding to self-disclosure of the confederate (e.g., “How does the participants handles information the confederate displays: to what extent does the participant genuinely listens to the confederate when he displays information”, “How empathic did the participant respond to information the confederate displayed?”). See Table 1 for the Cronbach’s alphas and Intraclass Correlation Coefficients of the subscales of the Self-disclosure Behaviour Scale and Table 2 for correlations between these subscales. The consistency and interrater reliability of the positive self-disclosure, eliciting self-disclosure and responding to self-disclosure was good and the correlations between these three subscales were high. However, for the negative self-disclosure the consistency was only low to good and the interrater reliability for the getting-acquainted task was low (see Table 1). In addition, the correlations of the negative self-disclosure scale with the others scales were moderate in the waiting room and low in the getting-acquainted task. Inspecting the means of the negative self-disclosure scale revealed that the video-raters observed only few negative self-disclosures in the social tasks (waiting room M = 2.6; SD = 1.5; getting-acquainted task: M = 4.0; SD = 1.1). This seems to indicate that negative self-disclosure did not occur with enough frequency to gain reliable ratings for this subscale. Therefore, this subscale was excluded from further analyses. One self-disclosure behaviour mean score was then constructed from the three reliable subscales: positive self-disclosure, eliciting self-disclosure and responding to self-disclosure.Table 1Cronbach’s alphas for each video-rater and Intraclass correlation coefficient (ICC) for each of the subscales of the self-disclosure behaviour scaleWaiting roomGetting-acquainted taskSubscales self-disclosure behaviour scaleCronbach’s alphaICCSubscales self-disclosure behaviour scaleCronbach’s alphaICCRater 1Rater 2Rater 3Rater 1Rater 2Rater 3Positive self-disclosure0.940.950.930.93Positive self-disclosure0.830.910.830.86Negative self-disclosure0.710.870.740.85Negative self-disclosure0.610.790.290.57Eliciting self-disclosure0.920.990.980.86Eliciting self-disclosure0.900.990.860.78Responding to self-disclosure0.930.980.900.92Responding to self-disclosure0.880.940.830.82 Table 2Simple correlations of the mean subscales of the self-disclosure behaviour scale rated by the three video-raters during the waiting room situation and getting-acquainted taskWaiting roomGetting-acquainted taskSubscales self-disclosure behaviour scale2.3.4.Subscales self-disclosure behaviour scale2.3.4.1. Positive self-disclosure0.52*0.87*0.92*1. Positive self-disclosure0.090.62*0.59*2. Negative self-disclosure0.41*0.43*2. Negative self-disclosure−0.11−0.213. Eliciting self-disclosure0.95*3. Eliciting self-disclosure0.94*4. Responding to self-disclosure4. Responding to self-disclosure*Correlation is significant at the 0.01 level Self-Disclosure Behaviour First, to investigate the self-disclosure behaviour for each group in the waiting room and in the getting-acquainted task, we conducted a repeated measures analyses with task (waiting room vs. getting-acquainted task) as a within subjects variable and group (high vs. low SAs) as a between subjects variable. Means and standard deviations of the self-disclosure behaviour are displayed in Table 3. A main effect of task appeared, F (1, 42) = 32.4, p < 0.001, η p2 = 0.44, indicating that more self-disclosure behaviour was observed in the getting-acquainted task than in the waiting room. Moreover, a main effect of group appeared, F (1, 42) = 14.9, p < 0.001, η p2 = 0.26, indicating that the high SAs showed less self-disclosure behaviour than the low SAs. Furthermore, a borderline interaction effect between task and group was present, F (1, 42) = 3.7, p = 0.061, η p2 = 0.08. A t test was conducted to test the difference (getting acquainted - waiting room) in self-disclosure behaviour between both groups. This specified that the high SAs showed a slightly greater increase in self-disclosure behaviour than the low SAs, t (42) = 1.9, p = 0.061, indicating that the difference between the groups was slightly more pronounced in the waiting room compared to the getting-acquainted task. Prediction of Change in Likeability by Group and Self-Disclosure Behaviour To examine the simple main effects, two t tests were conducted to test the difference between high and low SA in the change of likeability for both tasks (see Table 3 for the mean residual change scores). These showed a significant effect for the residual change score of the waiting room, t (47) = 2.55, p = 0.014, d = 0.74, but not of the getting-acquainted task, t (47)d = −0.33, p = 0.741, d = 0.11. Second, pearsons-r correlations showed that there is a positive relation between self-disclosure behaviour and the residual change scores in both tasks, rwaiting room = 0.60, p < 0.001; rgetting-acquainted task = 0.60, p < 0.001). Two regression analyses were conducted to test if group, the self-disclosure behaviour and the interaction between these two variables can explain the residual change scores of likeability during (1) the waiting room and (2) during the getting-acquainted task. Regression coefficients are displayed in Table 4. Both regression analyses showed a significant effect of self-disclosure, but no main effect of group and no interaction between group and self-disclosure. Thus, independent of social anxiety, the more self-disclosure participants displayed, the more they increased in likeability.Table 4Standardized regression coefficients of regression analysis to predict the residual change of likeability during the waiting room (i.e., from first impression to the DFI after the waiting room) and during the getting-acquainted task (i.e., from DFI after the waiting room to after the getting-acquainted task)ΒβR2Waiting room0.37 Group0.210.21 Self-disclosure behaviour0.17*0.71 Interaction between group and self-disclosure behaviour−0.04−0.19Social task0.44 Group1.00.97 Self-disclosure behaviour0.44**0.84 Interaction between group and self-disclosure behaviour−0.11−0.63 *p < 0.005. ** p < 0.001 "