Digital data and personality: A systematic review and meta-analysis of human perception and computer prediction.

Joanne Hinds,Adam N. Joinson

DOI: https://doi.org/10.1037/bul0000430

IF: 22.4

2024-07-13

Psychological Bulletin

Abstract:In recent years, our increasing use of technology has resulted in the production of vast amounts of data. Consequently, many researchers have analyzed digital data in attempt to understand its relationship with individuals' personalities. Such endeavors have inspired efforts from divergent fields, resulting in widely dispersed findings that are seldom synthesized. In this two-part study, we draw from two distinct areas of personality prediction across psychology and computer science to explore the convergent validity of self-reports with human perception and machine learning algorithms, the identifiability of the Big Five traits, and the predictability of different types of data. In Study 1, five meta-analyses of human perception studies integrating findings from 24,124 individuals rated across 30 independent samples demonstrated moderate convergent validity across all traits (ranging from ρ = 0.38 for Neuroticism, to ρ = 0.57 for Openness). In Study 2, a multilevel meta-analysis of computer prediction studies reporting 534 effect sizes across 42 studies also demonstrated moderate convergent validity (ρ = 0.30). Multivariate analyses of the significant moderators highlighted that X, Facebook, Sina Weibo, videos, and smartphones had a negative impact on the variance identified. Finally, in synthesizing the extant literature, we discuss the measures used to assess personality and the analytical approaches adopted. We identify the strengths and limitations across each field and explain how interdisciplinary methodologies could advance the testing and development of psychological theory. (PsycInfo Database Record (c) 2024 APA, all rights reserved)

psychology, multidisciplinary