The purpose of this study was to test the construct validity and internal consistency of the Social Support Inventory (SSI) for exercise in older adults. Four constructs of the SSI included emotional and instrumental social support delivered by directive or nondirective styles. Support providers try to control or influence recipients' emotions with directive emotional support, but allow recipients to express their emotions without controlling or dominating using nondirective emotional support. Support providers take charge of recipients' resources or responsibilities using directive instrumental support, but allow recipients to assume control and responsibility using nondirective instrumental support. SSI items reflecting these four constructs were generated from qualitative data. In a cross-sectional design, 310 community-dwelling participants (65 years old and older) completed the SSI for exercise behavior. Exploratory factor analysis was used to determine the construct validity of SSI. To represent the theoretical constructs, four factors were extracted by principal axis factoring with Oblimin rotation. Items were eliminated if they did not load on any factor at or above .30, cross-loaded with less than .10 difference, or did not reflect a meaningful interpretation of the content of the factor. Cronbach's alphas were used to estimate the internal consistency for each factor. Four factors accounted for 55.8% of the variance and matched the conceptual framework, including nondirective emotional support (9 items), directive instrumental support (5 items), directive emotional support (6 items), and nondirective instrumental support (2 items). Coefficient alphas ranged from .93 to .78. The SSI for exercise support in older adults demonstrates acceptable construct validity and internal consistency. It is a potentially valuable tool to understand the effect of social support on exercise behavior. Future research should continue exploring the validity of the SSI.
Session #1223 - Graduate Student Poster Session
The 29th Annual MNRS Research Conference (April 1-4, 2005)