How much can personality predict prosocial behavior?

Abstract

Explaining prosocial behavior is a central goal in classic and contemporary behavioral science. Here, for the first time, we apply modern machine learning techniques to uncover the full predictive potential that personality traits have for prosocial behavior. We utilize a large-scale dataset (N = 2707; 81 personality traits) and state-of-the-art statistical models to predict an incentivized measure of prosocial behavior, Social Value Orientation (SVO). We conclude: (1) traits explain 13.9% of the variance in SVO; (2) linear models are sufficient to obtain good prediction; (3) trait–trait interactions do not improve prediction; (4) narrow traits improve prediction beyond basic personality (i.e., the HEXACO); (5) there is a moderate association between the univariate predictive power of a trait and its multivariate predictive power, suggesting that univariate estimates (e.g., Pearson’s correlation) can serve as a useful proxy for multivariate variable importance. We propose that the limited usefulness of nonlinear models may stem from current measurement practices in personality science, which tend to favor linearly related constructs. Overall, our study provides a benchmark for how well personality predicts SVO and charts a course toward better prediction of prosocial behavior.

Plain language summary

The importance of being kind, helpful, and cooperative—of behaving prosocially—is undeniable. And yet, not everyone is equally prosocial. In fact, research in the social and behavioral sciences consistently shows that prosociality differs between individuals and is predicted by individuals’ personalities. This research, however, typically only manages to explain a small portion of the variance in prosocial behavioral tendencies, rarely going beyond 9% (corresponding to a correlation of r = .30). One possible explanation for the lack of better prediction is the widespread use of constrained statistical models including only few personality traits as predictors at a time. Using a large set of predictors and state-of-the-art statistical models, we estimate that personality traits account for 13.9% of the variance in Social Value Orientation, an established indicator of prosocial behavior. By modeling multiple personality traits in tandem, we thereby break the 9% ceiling—a feat that single-predictor models rarely manage. However, we also find that more complex (i.e., nonlinear) statistical models only slightly increase the amount of variance explained compared to more restrictive (i.e., linear) models. This suggests that the pursuit of better personality models of prosociality does not hinge on increasing the complexity of statistical models; instead, we suggest it starts by embracing a wider range and variety of personality traits.