Phase 1 Literature Review |
▪ Extensive review of published questionnaires that assess physical activity and/or sedentary activity supports and frequency of use in the home to identify key components of the home and near environments related to obesity risk |
Phase 2 Item Bank Creation |
▪ Identification of items pertinent to study purpose |
▪ Adaptation, enhancement, and expansion of questionnaire items |
▪ Organization of items by location category (i.e., inside the home, immediately outside the home [i.e., yards], and neighborhoods [i.e., playgrounds]) |
Phase 3 Initial Expert Review of Items |
▪ Expert review (n = 6) of items to identify appropriateness, gaps in assessment, and suitability for use in homes with preschool children |
Phase 4 Item Refinement |
▪ Further item adaptation, expansion, and de novo development |
▪ Revision or elimination of items that were age-inappropriate (e.g., presence of basketball hoop at the home) |
Phase 5 Questionnaire Design |
▪ Creation of questionnaire layout |
▪ Development of scoring procedures |
Phase 6 Content Validity Review |
▪ Second expert (n = 8) review and refinement to establish content validity and confirm scoring procedure |
▪ Refinement of items |
Phase 7 Cognitive Testing |
▪ Cognitive testing conducted with parents of preschoolers (n = 5) to ensure accurate comprehension and establish face validity |
▪ Refinement of items |
Phase 8 Field Testing to Establish Criterion Validity and Test-Retest Reliability |
▪ Field-tested questionnaire with parents age ≥18 and <45 years and having at least one child 2- to 5-years-old |
▪ Part 1: At home visits, parents and researchers simultaneously, but independently, completed the questionnaire (researchers served as the “gold standard” or criterion and followed specific guidelines) |
▪ Part 2: ~2 weeks later, parents completed the questionnaire online again to establish test-retest reliability |
▪ Refinement of the questionnaire based on field testing results. |
Phase 9 Establishing Scale Unidimensionality, Internal Consistency, and Convergent Validity |
▪ Factor analysis of the refined questionnaire with parents (N = 655) having at least one child 2- to 5-years-old completing baseline questionnaires for the HomeStyles intervention. |
▪ Split-half cross-validation with exploratory (n = 327) and confirmatory (n = 330) halves to examine factor structure of the measure using Principal Components Analysis with varimax rotation (orthongonal) and Kaiser normalization to establish unidimensionality of the scales. |
▪ Part 1: Iterative exploratory factor analysis conducted to identify strong factor solutions (minimum loading of 0.4) and eliminate cross-loading items (loadings on >1 scale within <0.2 of each other). |
▪ Part 2: Verify exploratory factor solutions via confirmatory factor analysis. |
▪ Calculate internal consistency of final scales. |
▪ Test convergent validity of the final scales in parents and children with other physical activity measures. |
Phase 10 Final Expert Review |
▪ Expert review (n = 5) of items to confirm content validity of the final questionnaire. |