Personality Research and Assessment in the Era of Machine Learning [post]

Clemens Stachl, Florian Pargent, Sven Hilbert, Gabriella M. Harari, Ramona Schoedel, Sumer Vaid, Samuel D. Gosling, Markus Bühner
2019 unpublished
The increasing availability of high-dimensional and fine-grained data about human behaviorin the form of digital footprints from online repositories and data traces from high-frequencymobile sensing studies, is about to drastically alter the way personality psychologists performresearch and personality assessment. The new opportunities to collect these kinds of dataraise crucial questions about how to best analyze and interpret them. Besides the potentialto model complex relationships in data,
more » ... tionships in data, these methods enable researchers to better judge thegeneralizability and robustness of their results with resampling techniques. However, thecorrect usage of machine learning models needs training and requires the consideration ofissues, specific to this type of modeling. We first provide a brief overview of past studiesusing machine learning in personality psychology. Second, we discuss and illustrate selectedissues and methodological challenges that researchers face: fitting, interpretation, andvalidation of machine learning models. Third, we discuss how latent variables could betreated in the modeling process, in personality psychology. Finally, we conclude with anoutlook on the future role of machine learning models in personality research and assessment.
doi:10.31234/osf.io/efnj8 fatcat:qf42ouyrizc5dhgx6molrhgp4q