Date of Award

June 2017

Degree Type


Degree Name

Doctor of Philosophy



First Advisor

Noelle Chesley

Committee Members

Marcus Britton, Celeste Campos-Castillo, Dana Garbarski


Experiment, Gender, Health Disparities, Physical Activity, Socioeconomic Status, Survey Methods


Survey research on the overall health and physical activity of the United States has relied on self-reports from questions that ask about leisure-only activity. Leisure activity patterns are known to be plagued by social forces that inhibit access and opportunity for women, compared to men, and for lower-socioeconomic individuals, compared to higher-socioeconomic individuals, making the further unpacking of leisure and other time use patterns imperative. To address this, the objective of this dissertation is to assess the different pathways individuals take to engage in health-benefiting physical activity and investigate the reliability and validity of physical activity survey questions as they relate to gender and socioeconomic disparities in physical activity.

The intention of this project is to ultimately inform best practices for survey question wording and, more generally, public health policy on physical activity. The research accomplishes the overall objective by pursuing the following specific aims: (1) To assess whether gender and SES physical activity ‘gaps’ are artificially produced through inaccurate measurement of physical activity, which will be completed by analyzing time use patterns of a nationally representative pooled sample from the American Time Use Survey (ATUS); (2) To offer new evidence and discovery on the potential impacts of priming language on physical activity questionnaires that highlight different opportunities for moderate physical activity. Building on the knowledge gained from the statistical analysis described above, the next step will be to collect new survey data using an experimental design. The experiment consists of eight conditions that include priming language for physical activity questions that previous national survey have overlooked or have not taken into account.

The project contributes an in-depth understanding of the theoretical relationship among gender, SES, social patterns of physical activity and time use in general, and understandings of physical activity as they relate to individuals’ survey questionnaire responses. This contribution is significant, because it can potentially transform future survey research as a means of studying the physical activity habits among diverse social groups.