Category: Adult Depression / Dysthymia
Recent findings have shown social support garnered in online spaces, such as social media and online gaming, can have positive effects on psychosocial outcomes in college and community adult samples. Established in-person theories purport that one way social support confers positive effects on these outcomes is by offsetting the ill effects of stressors via a main effect. In a two-wave, two-month longitudinal study of 162 adults aged 18-40, we tested the following hypotheses in a series of multiple regressions: 1) online social support offsets the effects of online and in-person victimization on depressive thoughts and self-esteem cross-sectionally via a main effect, in the same manner as in-person social support, and 2) each type of social support predicts improvements in depressive thoughts and self-esteem longitudinally. At time one, less frequent online social support (standardized b=-.223, p =.002) and more frequent in-person victimization (b=.397, p=.000) predicted greater depressive thoughts over and above each other and in expected, opposite directions (R2=.185, p=.000). In addition, more frequent online social support (b=.110, p=.000) and less frequent online victimization (b=-.294, p=.000) predicted greater self-esteem over and above each other in expected, opposite directions (R2=.110, p=.000). Findings were the same, though larger, for in-person social support: less frequent in-person social support (b=-.399, p=.000) and more frequent in-person victimization (b=.288, p=.000) predicted greater depressive thoughts (R2=.289, p=.000), while more frequent in-person social support (b=.376, p=.000) and less frequent online victimization (b=-.200, p=.005) predicted greater self-esteem (R2=.198, p=.000). At time two, six of eight of these findings replicated. To further explore which aspects of online support are most impactful, we placed all four subtypes of online social support into a multiple regression with each type of victimization. Subtypes are esteem/emotional support (caring and validation), social companionship (spending time together), informational (providing information), and instrumental (providing services or help with tasks). In both waves, esteem/emotional support (b=-.328, p=.010; b=.-334, p=.006) demonstrated incremental predictive utility over and above in-person victimization in the prediction of depressive thoughts (change in R2=.039, .037), as well as over and above online victimization in the prediction of depressive thoughts (b=-.289, p=.027; b=-.348, p=.005; change in R2=.037, .036). Also in both waves, instrumental support (b=.281, p=.018; b=.239, p=.051) trended towards incremental predictive utility over-and-above online victimization in the prediction of self-esteem (change in R2=.025, .039). Longitudinal main effects of time one online or in-person social support on time two depressive thoughts or self-esteem, over and above prior levels of each outcome, were not found. These findings largely replicate past cross-sectional findings, and perhaps suggest the effects of everyday forms of in-person and online social support on depressive and self-esteem outcomes are more momentary than long-lasting in community adults.
Elizabeth Nick– Graduate Student Researcher, Vanderbilt University, Nashville, Tennessee
David Cole– Professor, Vanderbilt University
Darcy Smith– Graduate Student Researcher, Vanderbilt University
Grace Carter– Undergraduate Research Assistant, Vanderbilt University