[Evaluation of the viability and validity of an instrument to measure community activation]

This study examines psychometric properties of a community activation assessment scale used for the evaluation of the Quebec project within the Federal Provincial Heart Health Initiative. The scale is composed of three sub scales, one for each of the intervention target risk factors.

Data were collected by telephone interviews conducted with 199 key informants from the 24 communities of the project. Reliability was studied using Cronbach's alpha coefficients and intraclass correlations. Validity was explored with factor analysis techniques.

Subscales' internal consistency were very good, ranging from 0.71 to 0.83, but intraclass correlations between informants from the same communities were low. Factor analyses indicated a unidimensional structure for each scale.

These results show that the scale seems to provide reasonable assessments of the opinion of key informants regarding activation in their community. However, ranking communities based on the aggregated scores from local key informants may lead to classification problems.