Previous research has shown that in close relationships, people tend to come from similar demographics. However, there is little known about personalities as measures of closeness. Wu Youyou Ph.D. et al. of Cambridge University aimed to examine personality similarities within romantic couples and friends. The researchers used Facebook to examine “Liked” pages and status updates to analyze language use.
The study received data from psychological tests taken by users on Facebook. Participants were 61.1% female and averaged 24.1 years old. To build the Likes-based personality-assessment model (PAM), Sample One consisted of 295,320 participants who had at least 20 Likes. For the language-based PAM, Sample Two contained 59,547 participants with 500 or more words across statuses. Samples 1 and 2 also completed personality questionnaires. Sample Three had 247,773 participants for correlations.
Three different versions of the International Personality Item Pool were used for personality questionnaires. They measured five traits: openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism. Reliability scores from the data ranged from α = .48 to α = .93.
The Likes-based PAM was built by assigning each participant a point if they liked the given page and zero points if they did not. For each of the five personality traits, a linear regression model was used to predict self-reported personality from the Likes. Likes and self-reported personality scores were correlated to the five traits using a subset of Sample Three, with results ranging from r = 0.25 to r = 0.39.
The language-based PAM was created by extracting words from statuses and transforming them into two predictors: indicators of whether the person used each word and frequency of usage.
The Likes-based scores and language-based results showed positive correlations on all five traits within romantic couples and friends. Self-reports were weak in both romantic couples and friends. All correlations were significant (p < .001).
These findings support that couples and friends have similar personalities. One problem of the study was the bias of having a friend’s liked pages advertised to another’s on Facebook, leading to an overlap of interests and exposure. Future research can look to filter out the bias by removing this advertising feature on Facebook.
- W. Youyou, et al. Birds of a feather do flock together. Psychological Science 28, 276-284 (2017). doi: 10.1177/0956797616678187
- Image retrieved from: https://www.pexels.com/search/couples/