Using supervised machine learning methods in psychological research

Abstract

Supervised machine learning can be used in many areas of psychological research, enabling the analysis of more complex data. Our aim is to describe the types, operation and use of supervised machine learning in psychological research. We review the benefits of machine learning, as well as the concepts of overfitting, bias, and variance that help in model selection and ensure robustness of the results. We also briefly describe the most important supervised machine learning algorithms and describe the key steps in the preparation of variables and data. An example analysis is presented to illustrate how the choice between stairs and elevator of university students can be modelled using supervised machine learning. At the end of the paper, we discuss the limitations of machine learning and its place in the education of psychologists. We hope that the knowledge presented will help psychologists to use machine learning more effectively and creatively.