Abstract

We evaluated the factor structure, reliability, and discriminative capacity of the Positive and Negative Affect Schedule (PANAS) questionnaire in four different samples: two general adult populations (N = 1,548, N = 964), one adolescent population (N = 1,044), and young people with depressive symptomatology (N = 307). Exploratory factor analyses (EFAs) were performed with subsamples from Studies 1 (n = 773) and 2 (n = 527), finding that the two- and three-factor solutions had a good fit. In a confirmatory factor analysis, the two-factor solution resulted in an adequate fit in a second set of subsamples from both studies (n = 775, n = 517). In Study 3, we found good convergent and divergent validity with adequate and significant correlations found for depression (Beck's Depression Inventory), anxiety (State-Trait Anxiety Inventory), and neuroticism and extroversion (Big Five Inventory). In Study 4, the results of an EFA performed in a subsample (n = 154) found that the two- and three-factor solutions were appropriate with the former solution being confirmed in a second subsample (n = 153). Reliability was alpha = .85 for positive affect and alpha = .87 for negative affect. The PANAS questionnaire showed adequate indicators of validity and reliability in adult and adolescent populations as well as in a sample with depressive symptoms.