Psychometric Testing of a Policy, Systems, and Environmental (PSE)-Focused SNAP-Ed Evaluation Tool.

OBJECTIVES: To validate the Healthy Youth Environment Survey, a pre-then-post survey completed by teachers to capture nutrition and physical activity-related changes in the classroom and school environment. METHODS: Analysis included 679 pretest surveys. A factor analysis was conducted to establish construct validity and Cronbach α was established to assess internal consistency reliability. RESULTS: The factor analysis extracted 10 main factors, with 97% of variables loading (≥ +0.30 factor loading) onto appropriate factors. After deleting one question, 6 of the 10 factors demonstrated acceptable internal reliability (Cronbach α ≥ 0.70). CONCLUSIONS AND IMPLICATIONS: After revision and subsequent retesting of construct validity, internal reliability, along with cognitive testing, The Healthy Youth Environment Survey may be suitable for use by other SNAP-Ed agencies implementing multi-level policy, systems, and environmental (PSE) interventions within youth education settings.

Validity and Reliability of the Healthy Families Survey: A Key Component of the Maryland Supplemental Nutrition Assistance Program-Education (SNAP-Ed) Evaluation.

OBJECTIVE: This study established the validity and reliability of the Healthy Families Survey, a 45-item survey measuring nutrition and physical activity behaviors among children and parents enrolled in the Maryland Supplemental Nutrition Assistance Program-Education. METHODS: Analysis included 1,376 pretest surveys. A factor analysis was conducted to establish construct validity, item analyses were conducted to determine item relevance for the target population, and Cronbach α was established to assess internal reliability. RESULTS: The factor analysis extracted 9 factors, with 91% of variables loading appropriately (>+0.30 loading) onto factors. Item analyses identified 6 variables with low (<0.20) or high (>0.80) mean values at baseline. Seven of the 9 subscales had Cronbach α ≥ .60. CONCLUSIONS AND IMPLICATIONS: The Healthy Families Survey demonstrated both strong construct validity and internal reliability and can be used by Supplemental Nutrition Assistance Program-Education agencies to assess child and parent nutrition and physical activity behaviors.

Feasibility of bidirectional text messages in evaluating a text-based nutrition education program for low-income parents: Results from the Text2BHealthy program.

Text messages are increasingly used in the delivery of health education programs. One appealing aspect of this approach is the possibility of remotely collecting participant data to use in program tailoring or evaluation. The purpose of the present study is to test the feasibility of using text messages to collect participant data. Using data from 33 texted evaluation questions sent through the Text2BHealthy nutrition education program for low-income parents (n=108-1521) response rates under different incentive and prompting strategies were examined. Response rates are generally low across a pilot year and three program years, ranging from 10-55%. While incentives seemed to be ineffective at improving response rates, results indicate that prompting participants to respond may increase response rates. Individuals who respond to an initial question are highly likely to respond to a follow-up question (88-99%) and to report positive behaviors (68-100%). Responses received through text may be unrepresentative and positively biased. Text messages may be a supplemental data collection strategy in nutrition education programs, but low response rates and response bias undermine data quality.