Evaluating the construct validity of sense of belonging as a measure of landscape perception

The construct validity of measures of sense of belonging to a recreational landscape is examined. The evaluation of construct validity emphasizes two concerns: (1) the empirical relationship between measures of two theoretically distinct concepts (belonging and visual preference); (2) the empirical relationship between measures of these concepts and theoretically relevant visitor characteristics. Visitor responses to photo-based measures of belonging and visual preference, as well as a verbal measure of belonging, were obtained during on-site interviews at Cumberland Gap National Historic Park, Tennessee. The results suggested that the photo-based measures of visual preference and belonging were highly correlated, but the verbal measure of belonging was not highly correlated with either photo-based measure. Thus, although the convergent validity of the belonging measures was not supported, the results did support the discriminant validity of the verbal measure of belonging. In addition, concurrent validity was supported by the finding that visitor characteristics that were conceptually linked to belonging were related to measures of belonging. Overall, the majority of results supported the construct validity of a verbal measure of belonging. However, unexpected differences between the verbal and photo-based measures suggest several interpretations related to construct validity that focus on the spatial representation of belonging measures and the social context of the landscape.