Abstract

In the early phase of the COVID-19 pandemic, before vaccines became available, a set of infection prevention behaviors constituted the primary means to mitigate the virus spread. Our study aimed to identify important predictors of this set of behaviors–information essential for guiding follow-up research and interventions. Whereas social and health psychological theories suggest a limited set of predictors, machine learning analyses can identify correlates of health behaviors from a larger pool of candidate predictors. We used random forests, a machine learning algorithm, to rank 115 candidate correlates of infection prevention behavior in a study of 56,072 participants across 28 countries, administered in March-May 2020. Results indicated that the two most important predictors related to individual-level injunctive norms—beliefs that people in the community should engage in social distancing and self-isolation, followed by endorsement of restrictive containment measures. The machine-learning model predicted 52% of the variance in infection prevention behavior in a separate test sample—exceeding the performance of psychological models of health behavior. Illustrating how data-driven methods can complement theory, some of the most important predictors were not derived from theories of health behavior—and some theoretically-derived predictors did not turn out to be important.

Cite as

van Lissa, C., Stroebe, W., vanDellen, M., Leander, N., Agostini, M., Draws, T., Grygoryshyn, A., Gützkow, B., Kreienkamp, J., Vetter, C., Abakoumkin, G., Abdul Khaiyom, J., Ahmedi, V., Akkas, H., Almenara, C., Atta, M., Bagci, S., Basel, S., Berisha Kida, E., Bernardo, A., Buttrick, N., Chobthamkit, P., Choi, H., Cristea, M., Csaba, S., Damnjanović, K., Danyliuk, I., Dash, A., Di Santo, D., Douglas, K., Enea, V., Faller, D., Fitzsimons, G., Gheorghi, A., Gómez, Á., Grzymala‑Moszczynska, J., Hamaidia, A., Han, Q., Helmy, M., Hudiyana, J., Jeronimus, B., Jiang, D., Jovanović, V., Kamenov, Ž., Kende, A., Keng, S., Kieu, T., Koc, Y., Kovyazina, K., Kozytska, I., Krause, J., Kruglanski, A., Kurapov, A., Kutlaca, M., Lantos, N., Lemay, E., Lemsmana, C., Louis, W., Lueders, A., Malik, N., Martinez, A., McCabe, K., Mehulić, J., Milla, M., Mohammed, I., Molinario, E., Moyano, M., Muhammad, H., Mula, S., Muluk, H., Myroniuk, S., Najafi, R., Nisa, C., Nyúl, B., O'Keefe, P., Olivas Osuna, J., Osin, E., Park, J., Pica, G., Pierro, A., Rees, J., Reitsema, A., Resta, E., Rullo, M., Ryan, M., Samekin, A., Santtila, P., Sasin, E., Schumpe, B., Selim, H., Stanton, M., Sultana, S., Sutton, R., Tseliou, E., Utsugi, A., van Breen, J., van Veen, K., Vázquez, A., Wollast, R., Yeung, V., Zand, S., Žeželj, I., Zheng, B., Zick, A., Zúñiga, C. & Bélanger, J. 2022, 'Using Machine Learning to Identify Important Predictors of COVID-19 Infection Prevention Behaviors During the Early Phase of the Pandemic', Patterns, 3(4), article no: 100482. https://doi.org/10.1016/j.patter.2022.100482

Downloadable citations

Download HTML citationHTML Download BIB citationBIB Download RIS citationRIS
Last updated: 05 August 2023
Was this page helpful?