Longitudinal analysis versus cross-sectional analysis in assessing the factors influencing shoppers’ impulse purchase behavior – Do the store ambience and salesperson interactions really matter?

Abstract

Numerous studies in the marketing literature focused on consumer behavior in general, but relatively few studies have examined Impulse purchase behavior (IPB). Although few studies examined IPB, the vast majority of the studies were conducted using the cross-sectional design. These studies suffer from certain limitations like random measurement error, common method bias, causality & validity-related issues that are inherently associated with the cross-sectional design. Despite these limitations, very few studies have examined the IPB using the longitudinal design. Multilevel structural equation modeling (ML-SEM) is conducted in the study to analyze the longitudinal data for examining the changes in the causal effects of the factors influencing the shoppers' IPB over a period of time. Additionally, structural equation modeling (SEM) is conducted to examine changes in the causal effects of the factors influencing IPB at each time point of data collection. Drawing upon the stern's model and stimulus-organism-response model, the study examines the causal effects of the factors influencing the IPB. The results of ML-SEM indicate significant fluctuations in the factors influencing IPB over time. Similarly, the results of SEM indicates that few factors (like store ambience and salesperson interactions) have shown a significant influence on IPB in the initial time points (i.e., during the initial store visits of shoppers), but became insignificant over a period of time in their subsequent store visits. The findings suggest that the store crowd, secondary customers influence, and in-store promotions show a significant influence on the IPB, compared to the store ambience and salesperson interactions.