Moderating effects of sociotropy and autonomy on affective and self-esteem reactivity to daily stressors

Based on psychoanalytic theory, Blatt (1974) proposed the depression vulnerability factors of dependency and self-criticism, which are similar to Beck’s factors of sociotropy and autonomy. The Depressive Experiences Questionnaire (DEQ; Blatt, D’Afflitti, & Quinlan, 1976) was developed to measure these constructs. The respective scales on the PSI and DEQ are highly intercorrelated (about .70) (Zuroff & Fitzpatrick, 1995).

Participants also rated their worst problem of the day on six appraisal dimensions (e.g., undesirability, control) and then completed 16 items from the brief COPE (Carver, 1997) to report their use of various coping strategies (e.g., emotional support, planning) to deal with their worst problem of the day.

For exploratory purposes, we conducted multilevel analyses to rigorously examine the unique effects of sociotropy and autonomy on exposure to different types of daily events, that is occurrence of interpersonal, achievement, and total negative events and positive events. (See Bolger & Zuckerman, 1995 for a discussion of Level 2 analyses of stress exposure.) There was only one effect, and it was a trend: Sociotropy was positively associated with number of interpersonal positive events, when neuroticism, gender, and autonomy were controlled, γ₀₃ = .007, p = .068.

To estimate the magnitude of the effect of sociotropy and autonomy on the reactivity slopes for daily stressors, we calculated the proportion reduction in mean squared prediction error when sociotropy or autonomy was entered into the Level 2 equations. Incremental variance explained by sociotropy was: (a) 1.9% for sadness reactivity to interpersonal stressors; (b) 1.4% for sadness reactivity to total stressors; (c) 2.9% for self-esteem reactivity to interpersonal stressors; and (d) 1.3% for self-esteem reactivity to total stressors. Sociotropy did not reduce the percentage of variance in self-esteem reactivity to achievement stressors. Incremental variance explained by autonomy was: (a) 5.7% in self-esteem reactivity to interpersonal stressors; and (b) 2.9% in self-esteem reactivity to total stressors.

We conducted a number of partial correlations involving sociotropy and autonomy and variables related to participants’ worst event of the day, controlling for gender, neuroticism, and the other PSI variable. None of these correlations were significant for percentage of worst events that were interpersonal, achievement, or other. There were significant positive correlations between sociotropy and appraisals of importance for interpersonal stressors and perceived coping efficacy for achievement stressors. There were a large number of significant partial correlations for the coping variables. In general, autonomy was associated with increased coping efforts regardless of whether interpersonal or achievement events were nominated as the day’s most stressful.

We also examined differences in affective and self-esteem reactivity to different types of negative events (i.e., interpersonal and achievement) nominated as the worst event of the day, and how reactivity to those events might be moderated by sociotropy and autonomy. Analyses were conducted first using a dummy-coded interpersonal event variable (1 = interpersonal, 0 = noninterpersonal), and then using a dummy coded achievement event variable (1 = achievement, 0 = nonachievement). Whether the worst event of the day was interpersonal or noninterpersonal had a significant effect on end of day sadness, γ₁₀ = 1.888, p < .001, and positive affect, γ₁₀ = −2.357, p < .001, such that end-of-day sadness was higher, and end-of-day positive affect was lower, when the negative event was interpersonal. Neither sociotropy nor autonomy had a significant moderating effect on these relationships. Whether the worst event of the day was interpersonal or noninterpersonal also had a significant effect on end-of-day self-esteem, γ₁₀ = −.985, p < .001, such that end of day self-esteem was lower when the event was interpersonal. Sociotropy did not have a significant moderating effect on this relationship, but autonomy had a near-significant moderating effect, γ₁₀ = .030, p = .078, such that autonomy was associated with weaker reactivity to type of event. Overall, individuals reacted to interpersonal stressors, compared with noninterpersonal stressors, with a greater decrease in self-esteem, but this pattern was weaker for autonomous individuals. Whether the worst event of the day was achievement or nonachievement did not have a significant effect on end-of-day sadness, positive affect, and self-esteem.