Possible selves represent how people think about their potential and about their future states (Markus & Nurius, 1986). To explore the cognitive processes involved in the decision to exercise, the possible selves (hoped-for and feared) of 93 undergraduate women at the University of Texas at Austin were examined, as were two self-efficacy constructs – scheduling and barrier self-efficacy. Most important hoped-for and feared possible selves related to exercise were categorized and analyzed. Physical and Health categories, followed by Personal and Spiritual, Occupation and Education, and Body Image were most commonly listed for hoped-for selves, while categories of Body Image, Health, and Personal and Spiritual, were most commonly cited for most important feared possible selves. Participants rated the importance, self-efficacy and outcome expectancy of their most important hoped-for and feared selves highly. Participants also felt highly efficacious in overcoming scheduling and barrier obstacles with regard to exercise. Comparisons were made across exercise levels, differentiating between those meeting or not meeting the recommended level of physical activity (Godin, 2011). Multiple logistic regression analyses, controlling for age, were used to test for significant relationships between motivational variables and exercise. Steps taken to achieve the most important possible self (odds ratio [OR] = 1.88, 95% CI = 1.21 – 2.92), steps taken to avoid the most important feared self (OR = 1.58, 95% CI = 1.04 – 2.40), scheduling self-efficacy (OR = 1.58, 95% CI = 1.18 – 2.10), and barrier self-efficacy (OR = 1.58, 95% CI = 1.04 – 2.40) were related to meeting the recommended level of physical activity, compared to not meeting the recommended level. These cross-sectional results suggest that the incorporation of college women’s possible selves and other motivational factors into interventions to increase women’s physical activity may be a promising area for future research. / text

abstract: Purpose: This study explored the potential correlates of exercise self-efficacy among older adults with a self-reported diagnosis of arthritis. Methods: This study was a secondary data analysis and used a cross-sectional design. Data was collected from a convenience sample of Non-Hispanic White and Non-Hispanic Black individuals between 2006-2008 (N=208). Descriptive statistics were run to assess means and frequencies within the sample. Bivariate statistics (Pearson and Spearman correlations, T-tests and one-way analysis of variance) were run to examine relationships between the independent and dependent variables. Multiple linear regression analyses were conducted to examine independent predictors of self-efficacy for exercise (SEE) and barriers self-efficacy for exercise (BSE). Results: Participants were predominantly female (85.6%), white (62.9%), retired (58.1%) and had a mean age of 66.6 [10.7] years. For education level, 23.4% reported a Master’s degree or higher and 18.6% reported they had at most a high school degree or GED. Nearly 47% of the sample were classified as obese based on self-reported body mass index (BMI) and 68.3% of the sample were not meeting the American College of Sports Medicine physical activity (PA) recommendations. Participants reported a relatively high BSE (22.6) and an average SEE (22.7). Significant positive associations were seen with outcome expectation for exercise (EOE), social support, and total minutes of PA and negative associations with BMI, physical function, pain, and negative affect with SEE and BSE. Meeting the PA guidelines (t134.5=4.60, 95%CI= 4.7(6.71-2.68), p<0.001) and being white (t164=2.82, 95%CI=2.82(0.57-5.08), p=0.014) were associated with SEE and BSE (t165=3.42, 95%CI= 4.37(6.89-1.85), p=0.001) and (t164=2.34, 95%CI= 2.95(0.46-5.43), p=0.021), respectively. In regression analyses, significant predictors of SEE were education (p=.006), physical function (p=.006) and EOE (p<.001). Significant predictors of BSE were physical function (p=.020), social support (p=.031), EOE (p=<.001), education level (p=.037), and total minutes of PA (p=.022). The variables in the SEE model accounted for 50.5% (R=.737, R2=.505) of the total variance and the variables in BSE model accounted for 41.1% (R=.672, R2=.411) of the total variance of the model. Discussion: EOE appears to be an important predictor of SEE and BSE. Examining the temporal relationship between EOE and SEE is warranted. / Dissertation/Thesis / Masters Thesis Exercise Science 2016