BackgroundAlthough there are many known benefits to exercise, only 10–30% of older adults report regular exercise. Understanding the factors that explain exercise behavior in older adults will help structure interventions that motivate these individuals to initiate and adhere to regular exercise.ObjectiveThe purpose of this study was to test the impact of components from two theoretical perspectives that explain exercise behavior, social cognitive theory, and the transtheoretical model.MethodsThis was a descriptive study using a sample of 179 older adults living independently in an East Coast continuing care retirement community (CCRC). A single one-time interview was completed and included: (a) stage of change for exercise, (b) self-efficacy and outcome expectations, (c) health status, (d) fear of falling, and (e) exercise activity. Model testing used structural equation modeling.ResultsTesting of the hypothesized model showed that 11 of the 22 paths were statistically significant. Health status and social support influenced self-efficacy and outcome expectations which directly influenced stage of change and exercise. Social support likewise directly influenced stage of change. Together the variables in the model explained 64% of exercise behavior in older adults. There was a poor fit of the hypothesized model to the data with a χ2of 144.7, degrees of freedom (df) of 23 and a ratio of 6.3. The revised model fit the data better χ2= 80.5,df= 19,p< .05, χ2/df= 4.2) and there was improved fit when compared to the hypothesized model (χ2difference of 64.0,dfdifference of 4,p≤ l.DiscussionThe combined testing of the mediating components of the theory of self-efficacy as well as the incorporation of the stage of change to explain exercise behavior in older adults suggests that both of these approaches are useful. Stage of change may be particularly useful to help determine the most appropriate intervention to increase activity and exercise in older adults.