Measuring residents' perceptions of social-ecological healthy communities: development and initial psychometric evaluation of a multidimensional scale.

INTRODUCTION

Effective community health promotion requires information about residents' health-related characteristics and the interpersonal, environmental, organizational, and policy supports they perceive. The Multidimensional Assessment Scale for Social-Ecological Healthy Communities MASE-HC was designed as a multidimensional resident-reported instrument, not as a unidimensional measure or an objective community-level rating. We report its development and initial psychometric evaluation.

METHODS

Items were generated through literature review, a two-round Delphi survey, content-validity assessment, and cognitive interviews. A purposive sample of 412 adult community residents was recruited from one community health service center in Anhui Province, China. Subsample 1 (n = 206) was used for item analysis and exploratory factor analysis, and Subsample 2 (n = 206) was used for confirmatory factor analysis, convergent and discriminant validity assessment, and hypothesis-based associations with self-rated health. Internal consistency was evaluated in the total sample (n = 412).

RESULTS

Data from 412 participants were analyzed. EFA identified a six-factor solution for the final 39-item MASE-HC, accounting for 61.411% of the variance. The final CFA specified six correlated first-order factors and did not evaluate a higher-order general factor. After four post-hoc within-factor residual covariances were added, the refined model showed generally acceptable fit (χ2/df = 1.741, RMSEA = 0.060, SRMR = 0.0649, CFI = 0.906, TLI = 0.898, and IFI = 0.907); TLI was marginally below 0.90. The Fornell-Larcker criterion was satisfied for all six factors, whereas convergent validity and internal consistency were comparatively weaker for Diet and Nutrition. Self-rated health was positively associated with five of the six subscales.

CONCLUSION

The MASE-HC showed preliminary support for a six-factor multidimensional structure in the present sample. The Diet and Nutrition subscale requires further refinement, and independent psychometric evaluation and replication in larger, more representative samples are needed before routine use or any community-level comparison.