TY - JOUR
T1 - Using machine learning to identify important predictors of COVID-19 infection prevention behaviors during the early phase of the pandemic
AU - Van Lissa, Caspar J.
AU - Stroebe, Wolfgang
AU - vanDellen, Michelle R.
AU - Leander, N. Pontus
AU - Agostini, Maximilian
AU - Draws, Tim
AU - Grygoryshyn, Andrii
AU - Gützgow, Ben
AU - Kreienkamp, Jannis
AU - Vetter, Clara S.
AU - Abakoumkin, Georgios
AU - Abdul Khaiyom, Jamilah Hanum
AU - Ahmedi, Vjolica
AU - Akkas, Handan
AU - Almenara, Carlos A.
AU - Atta, Mohsin
AU - Bagci, Sabahat Cigdem
AU - Basel, Sima
AU - Kida, Edona Berisha
AU - Bernardo, Allan B.I.
AU - Buttrick, Nicholas R.
AU - Chobthamkit, Phatthanakit
AU - Choi, Hoon Seok
AU - Cristea, Mioara
AU - Csaba, Sára
AU - Damnjanović, Kaja
AU - Danyliuk, Ivan
AU - Dash, Arobindu
AU - Di Santo, Daniela
AU - Douglas, Karen M.
AU - Enea, Violeta
AU - Faller, Daiane Gracieli
AU - Fitzsimons, Gavan J.
AU - Gheorghiu, Alexandra
AU - Gómez, Ángel
AU - Hamaidia, Ali
AU - Han, Qing
AU - Helmy, Mai
AU - Hudiyana, Joevarian
AU - Jeronimus, Bertus F.
AU - Jiang, Ding Yu
AU - Jovanović, Veljko
AU - Kamenov, Željka
AU - Kende, Anna
AU - Keng, Shian Ling
AU - Thanh Kieu, Tra Thi
AU - Koc, Yasin
AU - Kovyazina, Kamila
AU - Kozytska, Inna
AU - Ryan, Michelle K.
N1 - Publisher Copyright:
© 2022 The Author(s)
PY - 2022/4/8
Y1 - 2022/4/8
N2 - 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.
AB - 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.
KW - COVID-19
KW - DSML2: Proof-of-concept: Data science output has been formulated, implemented, and tested for one domain/problem
KW - health behaviors
KW - machine learning
KW - public goods dilemma
KW - random forest
KW - social norms
UR - http://www.scopus.com/inward/record.url?scp=85127500709&partnerID=8YFLogxK
U2 - 10.1016/j.patter.2022.100482
DO - 10.1016/j.patter.2022.100482
M3 - Article
AN - SCOPUS:85127500709
SN - 2666-3899
VL - 3
JO - Patterns
JF - Patterns
IS - 4
M1 - 100482
ER -