Title

Parental perceptions of influence on children's physical activity behavior: Instrument design and validation

Date of Completion

January 2009

Keywords

Education, Multilingual|Health Sciences, Public Health|Health Sciences, Recreation|Hispanic American Studies

Degree

Ph.D.

Abstract

An epidemic of childhood obesity is sweeping the nation—with nearly 32 percent of white children classified as overweight or obese and almost 37and 43 percent of African-American and Mexican-American children respectively falling into this category. Physical inactivity has been identified as a significant contributor to overweight. Parents have been identified as the "gatekeepers" of children's activity levels by either inhibiting or promoting physical activity depending on parental personal behaviors. Objective. The objective of this research project was to test the construct validity of an instrument designed to better understand how parents reinforce, give feedback, or support healthy lifestyles among their children, especially regarding physical activity, among both English- and Spanish-speaking participants. Methods. The survey was written in English and translated into Spanish. Surveys were distributed in the appropriate language at two sites, with a total of 549 completed surveys returned. Survey data were analyzed using exploratory and confirmatory factor analysis. Results. A number of statistical analyses were utilized to attempt to answer the research questions that were proposed for this study. Differences in sample sizes between site one and site two made t-test comparisons difficult, and problems with the demographic data inhibited the interpretation of the regression analyses. However, both the exploratory and confirmatory factor analyses provided critical insight regarding the construct validity of the English and Spanish versions of the survey. Conclusion. The instrument was composed of three subscales, which represented "Perceptions", "Instrumental Support", and "Emotional Support". While similar models have been tested extensively among English-speaking populations, equivalent research has not attempted to apply this theory to non-English speaking participants. The English version of the instrument displayed construct validity for assessing parental perceptions of their influence on their child's physical activity behavior. The results of the CFA for the Spanish instrument however, indicated that the model was different than the model that supported the English data, so the proposed model is not the same in both English and Spanish—an important finding that points to the need for broader engagement and understanding of the cultural underpinnings that may be at play in this situation. ^

COinS