Development, validation, and comparison of self-report measures for positive and negative affect in intensive longitudinal research

Affect is central to human functioning. Due to its dynamic nature, it is often studied with intensive longitudinal designs, yet the development and validation of measures for this purpose have received little systematic attention. In the present study, we review theoretical and methodological conceptualizations of affect that are relevant for repeated momentary positive and negative affect measurement. We developed a questionnaire including six dimensional affect and 22 discrete emotion items that allowed us to measure alternative momentary affect constructs with single- and multi-item scores. The items were operationalized into two bipolar, six positive, and six negative momentary affect measures. We compared the measures with three quantifiable criteria of construct validity: the amount of within-person variance, within-person sensitivity to emotional events, and between-person relations to depression and neuroticism. The criteria were empirically investigated with a preregistered experience sampling study (N = 153). We identify the measures with the strongest validity evidence across all criteria and evaluate their suitability for specific research questions, by looking at individual criteria. The overall findings provide strong evidence supporting the use of single-item measures of momentary affect. Furthermore, single items provide an efficient low-burden assessment tool that is comparable across studies. For multi-item scales, it is recommended to combine discrete emotion items of similar intensity, while simply selecting and averaging discrete emotion items is problematic concerning our validity criteria. In the future, we encourage the field to conduct systematic research on the use and interpretation of scores that aggregate different emotion items together. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Publication year:2023

Accessibility:Open