1. Using machine learning to identify important predictors of COVID-19 infection prevention behaviors during the early phase of the pandemic.
- Author
-
Van Lissa CJ, Stroebe W, vanDellen MR, Leander NP, Agostini M, Draws T, Grygoryshyn A, Gützgow B, Kreienkamp J, Vetter CS, Abakoumkin G, Abdul Khaiyom JH, Ahmedi V, Akkas H, Almenara CA, Atta M, Bagci SC, Basel S, Kida EB, Bernardo ABI, Buttrick NR, Chobthamkit P, Choi HS, Cristea M, Csaba S, Damnjanović K, Danyliuk I, Dash A, Di Santo D, Douglas KM, Enea V, Faller DG, Fitzsimons GJ, Gheorghiu A, Gómez Á, Hamaidia A, Han Q, Helmy M, Hudiyana J, Jeronimus BF, Jiang DY, Jovanović V, Kamenov Ž, Kende A, Keng SL, Thanh Kieu TT, Koc Y, Kovyazina K, Kozytska I, Krause J, Kruglanksi AW, Kurapov A, Kutlaca M, Lantos NA, Lemay EP Jr, Jaya Lesmana CB, Louis WR, Lueders A, Malik NI, Martinez AP, McCabe KO, Mehulić J, Milla MN, Mohammed I, Molinario E, Moyano M, Muhammad H, Mula S, Muluk H, Myroniuk S, Najafi R, Nisa CF, Nyúl B, O'Keefe PA, Olivas Osuna JJ, Osin EN, Park J, Pica G, Pierro A, Rees JH, Reitsema AM, Resta E, Rullo M, Ryan MK, Samekin A, Santtila P, Sasin EM, Schumpe BM, Selim HA, Stanton MV, Sultana S, Sutton RM, Tseliou E, Utsugi A, Anne van Breen J, Van Veen K, Vázquez A, Wollast R, Wai-Lan Yeung V, Zand S, Žeželj IL, Zheng B, Zick A, Zúñiga C, and Bélanger JJ
- Abstract
Before vaccines for coronavirus disease 2019 (COVID-19) 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. Whereas social and health psychological theories suggest a limited set of predictors, machine-learning analyses can identify correlates from a larger pool of candidate predictors. We used random forests to rank 115 candidate correlates of infection-prevention behavior in 56,072 participants across 28 countries, administered in March to May 2020. 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. Results indicated the two most important predictors related to individual-level injunctive norms. 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 were relatively unimportant., Competing Interests: The authors declare no competing interests., (© 2022 The Author(s).)
- Published
- 2022
- Full Text
- View/download PDF