How researchers used smartphone data to gain insight into a user’s personality
A report released online in the peer-reviewed journal PNAS demonstrates how efficient digital data is to gain insight into a user’s personality.
Conducted at the Ludwig Maximilian University of Munich, a research team there gathered more than 600 participants for their study.
The participants began by completing questionnaires pertaining to their personality traits. Thereafter, a mobile application was installed into their smartphone to gain more insight into their day-to-day digital habits.
As the app ran in the background for 30 days, it collected data on the user’s behavior, communication patterns, music consumption, app selections, and overall phone usage. Researchers then analyzed all the data using machine-learning algorithms, capable of recognizing and extracting patterns of behavioral data, much more efficiently than a human counterpart.
To identify personality traits, the use of the Big Five dimensions: openness, conscientiousness, extraversion, agreeableness, and emotional stability, was instrumental in understanding the communication patterns and social behavior of the data.
According to researchers, the automated analysis revealed that the “algorithm was indeed able to successfully derive most of these personality traits from combinations of the multifarious elements of their smartphone usage.”
In their findings, the data corresponding to communication patterns and social behavior was associated with self-reported extraversion. The study also unveiled other personality traits.
“Information relating to patterns of day and night-time activity was significantly predictive of self-reported degrees of conscientiousness,” the co-authors stated in a news release. “Notably, links with the category “openness” only became apparent when highly disparate types of data (e.g., app usage) were combined.”
“By far the most difficult part of the project was the pre-processing of the huge amount of data collected and the training of the predictive algorithms.”
All in all, the findings open new opportunities for researchers to learn more about how people actually behave in their everyday lives, without the need to rely exclusively on self-assessments or questionnaires.