The Future Is Here
We may earn a commission from links on this page

The nature of online courtship, visualized

We may earn a commission from links on this page.

Facebook's Data Science team has published a thought provoking post that examines the statistical correlation between couples' timeline posts and their burgeoning relationships.

By looking at how the frequency of timeline posts varies in the days leading up to and following the start of a new relationship, the team has identified a sort of digital courtship curve:

During the 100 days before the relationship starts, we observe a slow but steady increase in the number of timeline posts shared between the future couple. When the relationship starts ("day 0"), posts begin to decrease. We observe a peak of 1.67 posts per day 12 days before the relationship begins, and a lowest point of 1.53 posts per day 85 days into the relationship. Presumably, couples decide to spend more time together, courtship is off, and online interactions give way to more interactions in the physical world.


And while the quantity of timeline posts goes down at the start of a new relationship, the quality of those posts (as measured by posts' sentimentality) goes up. Using a word-analysis technique similar to the one used to produce this map of racist, homophobic and ableist tweets, the Facebook team demonstrates that, when a relationship is forged, the emotional content of the posts skews from positive to very positive in the span of about 24 hours.


The observable shift in each graph isn't as huge as the curves might initially suggest (the frequency curve drops from 1.66 posts/day to 1.56/day; the positive emotion level rises from ~.018/post to .022/post), but the correlations that they highlight seem pretty unimpeachable. The dramatic changes observed in each graph, and Facebook's team's explanation for them, are more or less what you'd expect – still, it's always fun to see your suspicions confirmed with data. What we'd really like to see is how these graphs change on a 200 day, 12-month, or 18-month+ timescale.

[Facebook Data Science]