. Author manuscript; available in PMC: 2021 Dec 13.

Published in final edited form as: Behav Ther. 2014 Sep 18;46(2):257–271. doi: 10.1016/j.beth.2014.09.006

Table 3.

Generalized Estimating Equations Analysis Using a Gamma Distribution of Social Comparisons (i.e., Body, Eating, and Exercise) and Body Surveillance as Predictors of Contemporaneous Body Dissatisfaction in a Single Model Controlling for Negative Affect

Parameter Estimates	B	SE	Wald χ²	df	p
Outcome variable: body dissatisfaction, N_{Level 1} = 8742, N_{Level 2} = 232
Intercept	1.97	.25	61.10	1	<.001
Body-related Social Comparison (person-mean centered)	.003	.001	32.97	1	<.001
Body-related Social Comparison (person means)	.04	.01	34.60	1	<.001
Eating-related Social Comparison (person-mean centered)	.003	.001	22.81	1	<.001
Eating-related Social Comparison (person means)	−.01	.01	2.35	1	.126
Exercise-related Social Comparison (person-mean centered)	.003	.001	16.86	1	<.001
Exercise-related Social Comparison (person means)	.001	.01	.05	1	.823
Body Surveillance (person-mean centered)	.10	.01	60.75	1	<.001
Body Surveillance (person means)	.19	.06	9.43	1	.002
Time	.001	.0002	31.19	1	<.001
Negative Affect (person-mean centered)	.004	.003	2.33	1	.127
Negative Affect (person means)	−.01	.02	.29	1	.587
(Scale)	.90
Pseudo-R² = .228