Convergent Validity, Internal Consistency, and Respondent Experience of Existing Instruments
by Berre Deltomme, Karen Gorissen, Bert Weijters
Abstract
The influence of human behavior on climate change and environmental decline is receiving increased attention; and therefore, it has led to an increase in studies that measure pro-environmental behavior (PEB) as a predictor, a covariate, or an outcome variable. To this end, (validated) self-report scales have traditionally been the main measurement tool, but lately, several experimental instruments have also been developed to measure PEB. Measurement instruments that are considered to measure the same construct should provide consistent results, i.e., they should show high convergent validity. However, it is not clear whether substitute measures for PEB show this necessary convergent validity and how they compare to each other in terms of internal consistency and respondent experience. To address this, we investigated thirteen validated self-report scales and three experimental tasks on their psychometric qualities (i.e., validity and internal consistency) and respondent experience. Therefore, we assigned 340 participants and randomly administered half of the instruments to each of them. The results show that, in general, convergent validity is lacking, which shows that the measurement instruments cannot be considered equivalent. As for respondent experience, the experimental tasks are most time-consuming, are perceived as most fatiguing, and are most sensitive to multitasking. The self-report scales are most sensitive to socially desirable responding and acquiescence bias. Our insights cater to researchers in environmental psychology and sustainability seeking to employ validated self-report scales or behavioral tasks in measuring PEB.