Thread by @hardsci

Is who you follow on Twitter correlated with your mental health? Using machine learning, we find that the high-degree accounts that people follow - celebrities, public figures, etc. - can predict anxiety, depression, post-traumatic stress, & anger (out-of-sample R = .2)

https://twitter.com/CostelloCK/status/1354502697689481223

Substantive upshot: A user's Twitter experience - what they see in their feed - is heavily affected by their decisions of who to follow. Those curation decisions, in the aggregate, are associated with mental health

Some notes and points of interest:

1. "Predict" is used in the machine-learning sense, think of this as a correlation - causality wasn't tested

2. The size of the correlation is neither huge nor neglible. We don't think it can be dismissed but it shouldn't be exaggerated either

3. Mental health prediction was mostly common variance, with a little bit of specificity for anger

4. Methods nerds: this is an example of how preregistration, k-fold cross-validation, and holdout validation can be combined to control bias at various stages (model fitting, selection, etc.)

5. Some suggestive signs of algorithmic bias

See @CostelloCK's thread & paper for more

Oh crap, forgot to mention under #4 - this study with a complicated beast of an analysis was a registered report! Even *more* protection against bias. The RR process was enormously helpful, big thanks to our editor @chrisdc77