Measurement reactivity in ambulatory assessment: Increase in emotional clarity over time independent of sampling frequency

Abstract

Ambulatory assessment (AA) studies are frequently used to study emotions, cognitions, and behavior in daily life. But does the measurement itself produce reactivity, that is, are the constructs that are measured influenced by participation? We investigated individual differences in intraindividual change in momentary emotional clarity and momentary pleasant-unpleasant mood over the course of an AA study. Specifically, we experimentally manipulated sampling frequency and hypothesized that the intraindividual change over time would be stronger when sampling frequency was high (vs. low). Moreover, we assumed that individual differences in dispositional mood regulation would moderate the direction of intraindividual change in momentary pleasant-unpleasant mood over time. Students (n = 313) were prompted either three or nine times a day for 1 week (data collection took place in 2019 and 2020). Multilevel growth curve models showed that momentary emotional clarity increased within participants over the course of the AA phase, but this increase did not differ between the two sampling frequency groups. Pleasant-unpleasant mood did not show a systematic trend over the course of the study, and mood regulation did not predict individual differences in mood change over time. Again, results were not moderated by the sampling frequency group. We discuss limitations of our study (e.g., WEIRD sample) and potential practical implications regarding sampling frequency in AA studies. Future studies should further systematically investigate the circumstances under which measurement reactivity is more likely to occur.