Predictive Profiling of Crisis-Proof Individuals: Current Behaviours

Abstract

The world has recently experienced the COVID-19 pandemic, a crisis with ongoing impacts. However, this is neither the first nor the last one of its kind. Every individual reacts to these situations differently with a variety of different effects. Hence, verifying the possibility of profiling individuals based on their personality and psychological predictors could be a useful step and the tool developed using such a method can have a great potential to implement important precautionary measures for the future. In our previous study (Yilmaz & Karanth Ramadeva, 2023), a large cross-cultural and multidimensional data set was collected (N=209). In our analyses, we used the current behaviour scale from the dataset to compute the factor loadings using Confirmatory Multidimensional Item Response Theory (IRT) models. The current behaviours of people during the pandemic were loaded into three higher dimensions which are - social distancing, taking health precautions, and personal hygiene. Further, we explored the individual distributions of current behaviours' dimensions and observed a normal distribution for each dimension. Therefore, a simple approach as such a median split is applicable to characterize the individuals as crisis-proof or not. As a future direction to this project, we aim to apply supervised machine learning with predictors such as personality traits, risk perception, and emotional regulation to understand the significant features which are necessary to predict distinct profiles of individuals, in the context of crisis-proofing.

Yilmaz, F., & Ramadeva, T.K. (2025). Predictive Profiling of Crisis-Proof Individuals: Current Behaviours, "forsch!" - Studentisches Online-Journal der Universität Oldenburg, 1.

https://nbn-resolving.org/urn:nbn:de:101:1-2507161811416.607008976919