To verify or not to verify: using partial least squares to predict effect of online news on panic buying during pandemic

Tan, K. L.; Sia, J. K. M.; Tang, D. K. H.

Purpose Coronavirus disease (COVID-19) pandemic has given rise to different dimensions of uncommon human behavior, and panic buying is one of them. Interestingly, panic buying research has not been given much attention. The purpose of this paper is threefold. Firstly, it examines the influences of the theory of planned behavior (TPB) elements (subjective norm, attitude and perceived behavior control (PBC)) on panic buying. Secondly, it investigates online news and the perceived likelihood of being affected (PLA) as antecedents to the TPB constructs. Finally, to examine online news verification as a moderator on the relationship between the TPB constructs and panic buying. Design/methodology/approach Data were collected from 371 respondents and analyzed using the partial least squares method structural equation modeling (PLS-SEM). PLS predict was applied to determine the predictive power of the model further. Findings This study found that subjective norms and attitude influence panic buying. The results further revealed that online news has a direct influence on the PLA and attitude. However, PBC has no such effect on panic buying. Surprisingly, online news verification also has no moderating effects on the relationships between the TPB elements and panic buying. Originality/value This research helps to understand consumer panic buying behavior, especially during shock events such as the COVID-19 pandemic. This study is the first that extends the TPB incorporating both online news and PLA as antecedents to panic buying in the same model. Furthermore, the study serves as an initial attempt to investigate online news verification as a moderator between the link of three constructs of TPB and panic buying, contributing to existing literature. Lastly, it advances the body of knowledge on consumer behavior and contributes methodologically by introducing the PLS approach.