Measuring leisure motivation: A meta-analysis of the recreation experience preference scales

One approach to studying the motivations for leisure is to focus on the desired goal states that are attained through participation in leisure. The Recreation Experience Preference (REP) scales were developed for measuring these goal states. In an attempt to provide a summary integrative analysis of the structure of the REP scales, the present study conducted a meta-analysis of 36 studies that have used REP items. The studies were used to obtain population estimates of correlations between scale item pairs. Correlations were then used as input to confirmatory factor analysis that tested the structure of domains (item group ings that represent a broad goal construct) and the structure of scales (within-domain item groupings that represent dimensions of the broader goal construct) established in previous research. Results provided support for the a priori domain and scale structures. Inter-item correlations were computed for do- mains and scales and compared ''within'' and ''between'' clusters. The results show high average inter-item correlations within scales and domains and relatively low average correlations between domains and scales. Variability of interitem domain correlations due to response scale, instruction set, geographic location of study, and type of recreation area visited was tested. Overall consistency in domains and stales was shown. Recommendations are made to help assure appropriate applications and advance refinement of the REP scales.