Predicting Individual Characteristics from Digital Traces on Social Media: A Meta-Analysis

Abstract

The increasing utilization of social media provides a vast and new source of user-generated ecological data (digital traces), which can be automatically collected for research purposes. The availability of these data sets, combined with the convergence between social and computer sciences, has led researchers to develop automated methods to extract digital traces from social media and use them to predict individual psychological characteristics and behaviors. In this article, we reviewed the literature on this topic and conducted a series of meta-analyses to determine the strength of associations between digital traces and specific individual characteristics; personality, psychological well-being, and intelligence. Potential moderator effects were analyzed with respect to type of social media platform, type of digital traces examined, and study quality. Our findings indicate that digital traces from social media can be studied to assess and predict theoretically distant psychosocial characteristics with remarkable accuracy. Analysis of moderators indicated that the collection of specific types of information (i.e., user demographics), and the inclusion of different types of digital traces, could help improve the accuracy of predictions.

Behavioral Patterns in Smartphone Usage Predict Big Five Personality Traits

“…Whereas Mønsted et al (2018) almost exclusively relied on data about communication behavior, we used a wide range of behavioral indicators for the prediction of personality traits. This is also in line with findings of Settanni et al (2018), reporting increased predictability of personal traits through the combination of data-types. Besides the prediction of big five personality trait scores from smartphone-sensed behaviors, our results shed more light on the behavioral underpinnings of personality per se.…”

“…Hence, it is likely that the utilization of behavioral patterns across a range of activities will make it possible to predict other big five personality traits, besides extraversion. This notion is also supported by a recent meta-analysis on personality prediction from social media data, showing that on average, a combination of different data types increased prediction performance in previous studies (Settanni et al, 2018).…”

Personality Research and Assessment in the Era of Machine Learning

“…Such personalization based recommender systems have recently gained popularity as a result of the success of the efforts described earlier to predict personality from digital footprints (Settanni et al, 2018;Youyou et al, 2015), text (Park et al, 2015;Schwartz et al, 2013), and mobile sensing data (Stachl et al, 2019). It is valuable to compute users' personality scores because recommender systems often suffer from a lack of valid constructs on which to base their recommendations.…”

“…However, model validation is extremely relevant for the application of ML in the field of personality psychology. Most studies using ML in personality research have not focused on the analysis of the inner workings of their models (see Settanni et al, 2018). Thus, in the following sections, we discuss two key aspects of model validity, (feedback loops, and biases) and we highlight their importance through examples.…”

Social media-predicted personality traits and values can help match people to their ideal jobs

“…As people engage with social media, they leave behind digital fingerprints-behavioral traces of their personality-which can be detected at a large scale (19)(20)(21)(22). Linguistic analyses of social media information have been used to predict an array of outcomes, including age, gender, political orientation, physical and mental illness, and unemployment (22)(23)(24)(25). However, associations between these factors and career success across a broad range of occupations are unknown.…”