Development and testing the validity and reliability of items and scales to assess physical activity behavior of adults in the expanded food and nutrition education program

Given the rising prevalence of obesity and other chronic diseases in the United States, the Expanded Food and Nutrition Education Program (EFNEP) includes physical activity as a core educational component along with diet quality, food safety, food resource management and food security. According to the Dietary Guidelines for Americans, research supports that participation in regular physical activity helps people maintain a healthy weight and prevent excess weight gain. When combined with reduced calorie intake, physical activity may aid weight loss and the maintenance of weight loss. Two evaluation instruments are used in EFNEP nationwide. Nutritional intake is measured by a 24-hour diet recall at baseline and following completion of the intervention. A ten-item survey, referred to as the EFNEP Behavior Checklist, measures behaviors in the constructs of food safety, food resource management, food security, and nutrition practices. The EFNEP Behavior Checklist currently does not include questions related to physical activity behavior. Therefore, it is important to develop and add valid and reliable items to the EFNEP behavior checklist to document participant change in physical activity behaviors. The community nutrition logic model, the constructs of Theory of Reasoned Action, and the theory of Planned Behavior provided the guiding framework of the development of items/scales. The target population was EFNEP eligible limited-income mothers who had at least one child under the age of 19 living in the household. To accomplish the goals and objectives of this study, a step-by-step procedure was used which included the following phases: a) curriculum review and identification of contents/concepts, b) conceptual frame work & item generation, c) expert review & content validity, d) revision of items & scales, e) cognitive testing, f) psychometric testing & analysis which included construct validity, internal consistency, test-retest reliability and predictive validity. Seventeen items/scales were developed on psychosocial mediating variables based on the constructs of TRA & TPB, which were attitude, subjective norm, perceived behavior control and intention. A total of 12 items/scales were developed in four dimensions of physical activity such as home, yard, walk and work. Content validity was assessed through the expert review (n=8) and cognitive testing (n=14). Data from 302 mothers was collected for factor analysis and internal consistency. Test-retest reliability was also assessed (n=50). Predictive validity of the physical activity items/scales was assessed using a self-report of physical activity and accelerometer data. Content analysis, expert review and cognitive testing were used to develop the items/scales and to establish the content validity. Factor analysis was used to determine the number of underlying factors in the items/scales and as a strategy of item reduction. Internal consistency for most of the final psychosocial items/scales was acceptable with Cronbach\u27s alpha greater than .70. Spearman correlation statistics for test-retest reliability ranged between modest to stronger (rs value between .59-.70 and P value The results of this study indicate the need for further investigation in using the items/scales to evaluate the impact of EFNEP on physical activity behavior of adults. This study provided an important first step in developing and testing items/scales with conceptual foundation and acknowledged essential elements to measure physical activity behavior of low-income population