. Author manuscript; available in PMC: 2013 Oct 30.

Published in final edited form as: J Aging Health. 2011 Apr 20;23(6):10.1177/0898264311404235. doi: 10.1177/0898264311404235

Table 2. Negative Binomial Regression and Logistic Regression Results.

	Logistic regression (Dependent variable: walking vs. no walking)		Negative binomial regression (Dependent variable: number of blocks walked)

	Main effects model: b (SE)	Model with interaction: b (SE)	Main effects model: b (SE)	Model with interaction: b (SE)
Depressive symptoms	−.074 (.04); p = .066	.051 (.07); p = .495	−.083 (.03); p = .003	−.022 (.04); p = .557
Instrumental social support	−.211 (.08), p = .005	−.073 (.10); p = .457	−.232 (.05); p < .001	−.148 (.06); p = .014
Interaction of Depressive symptoms × Instrumental social support	—	−.036 (.02); p = .044	—	−.018 (.01); p = .058

Note. Control variables: gender, age, education, financial strain, marital status, and body mass index.