Using Machine Learning to Identify Important Predictors of COVID-19 Infection Prevention Behaviors During the Early Phase of the Pandemic van Lissa, Caspar J van Lissa Caspar J Stroebe, Wolfgang Stroebe Wolfgang vanDellen, Michelle R vanDellen Michelle R Leander, N Pontus Leander N Pontus Agostini, Maximilian Agostini Maximilian Draws, Tim Draws Tim Grygoryshyn, Andrii Grygoryshyn Andrii Gutzgow, Ben Gutzgow Ben Kreienkamp, Jannis Kreienkamp Jannis Vetter, Clara S Vetter Clara S Abakoumkin, Georgios Abakoumkin Georgios Abdul Khaiyom, Jamilah Hanum Abdul Khaiyom Jamilah Hanum Ahmedi, Vjolica Ahmedi Vjolica Akkas, Handan Akkas Handan Almenara, Carlos A Almenara Carlos A Atta, Mohsin Atta Mohsin Bagci, Sabahat Cigdem Bagci Sabahat Cigdem Basel, Sima Basel Sima Kida, Edona Berisha Kida Edona Berisha Bernardo, Allan B I Bernardo Allan B I Buttrick, Nicholas R Buttrick Nicholas R Chobthamkit, Phatthanakit Chobthamkit Phatthanakit Choi, Hoon-Seok Choi Hoon-Seok Cristea, Mioara Cristea Mioara Csaba, Sara Csaba Sara Damnjanovic, Kaja Damnjanovic Kaja Danyliuk, Ivan Danyliuk Ivan Dash, Arobindu Dash Arobindu Di Santo, Daniela Di Santo Daniela Douglas, Karen M Douglas Karen M Enea, Violeta Enea Violeta Faller, Daiane Gracieli Faller Daiane Gracieli Fitzsimons, Gavan J Fitzsimons Gavan J Gheorghiu, Alexandra Gheorghiu Alexandra Gomez, Angel Gomez Angel Hamaidia, Ali Hamaidia Ali Han, Qing Han Qing Helmy, Mai Helmy Mai Hudiyana, Joevarian Hudiyana Joevarian Jeronimus, Bertus F Jeronimus Bertus F Jiang, Ding-Yu Jiang Ding-Yu Jovanovic, Veljko Jovanovic Veljko Kamenov, Zeljka Kamenov Zeljka Kende, Anna Kende Anna Keng, Shian-Ling Keng Shian-Ling Thanh Kieu, Tra Thi Thanh Kieu Tra Thi Koc, Yasin Koc Yasin Kovyazina, Kamila Kovyazina Kamila Kozytska, Inna Kozytska Inna Krause, Joshua Krause Joshua Kruglanksi, Arie W Kruglanksi Arie W Kurapov, Anton Kurapov Anton Kutlaca, Maja Kutlaca Maja Lantos, Nora Anna Lantos Nora Anna Lemay, Edward P Jr Lemay Edward P Jr Jaya Lesmana, Cokorda Bagus Jaya Lesmana Cokorda Bagus Louis, Winnifred R Louis Winnifred R Lueders, Adrian Lueders Adrian Malik, Najma Iqbal Malik Najma Iqbal Martinez, Anton P Martinez Anton P McCabe, Kira O McCabe Kira O Mehulic, Jasmina Mehulic Jasmina Milla, Mirra Noor Milla Mirra Noor Mohammed, Idris Mohammed Idris Molinario, Erica Molinario Erica Moyano, Manuel Moyano Manuel Muhammad, Hayat Muhammad Hayat Mula, Silvana Mula Silvana Muluk, Hamdi Muluk Hamdi Myroniuk, Solomiia Myroniuk Solomiia Najafi, Reza Najafi Reza Nisa, Claudia F Nisa Claudia F Nyul, Boglarka Nyul Boglarka O'Keefe, Paul A O'Keefe Paul A Olivas Osuna, Jose Javier Olivas Osuna Jose Javier Osin, Evgeny N Osin Evgeny N Park, Joonha Park Joonha Pica, Gennaro Pica Gennaro Pierro, Antonio Pierro Antonio Rees, Jonas Rees Jonas Reitsema, Anne Margit Reitsema Anne Margit Resta, Elena Resta Elena Rullo, Marika Rullo Marika Ryan, Michelle Ryan Michelle Samekin, Adil Samekin Adil Santtila, Pekka Santtila Pekka Sasin, Edyta M Sasin Edyta M Schumpe, Birga M Schumpe Birga M Selim, Heyla A Selim Heyla A Stanton, Michael Vicente Stanton Michael Vicente Sultana, Samiah Sultana Samiah Sutton, Robbie M Sutton Robbie M Tseliou, Eleftheria Tseliou Eleftheria Utsugi, Akira Utsugi Akira Anne van Breen, Jolien Anne van Breen Jolien Van Veen, Kees Van Veen Kees Vazquez, Alexandra Vazquez Alexandra Wollast, Robin Wollast Robin Wai-Lan Yeung, Victoria Wai-Lan Yeung Victoria Zand, Somayeh Zand Somayeh Zezelj, Iris Lav Zezelj Iris Lav Zheng, Bang Zheng Bang Zick, Andreas Zick Andreas Zuniga, Claudia Zuniga Claudia Belanger, Jocelyn J Belanger Jocelyn J Before vaccines for 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-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. © 2022 The Author(s). Cell Press 2022 1