On Twitter, everyone wants to project the best version of themselves that they can – after all, the whole world is watching. But can your activity on Twitter really give clues about your personality? And can this data single you out as a psychopath?

The folks at the Online Privacy Foundation (OPF) recently teamed up with the statistical and analytics crowdsourcing platform Kaggle to attempt to answer these questions.

OPF began its ‘Twitter Big 5 Experiment’ by asking volunteers to answer a questionnaire that helped to identify their personalities. Though the experiment was based on the five factors of personality model, the team actually examined eight personality elements: openness, conscientiousness, extraversion, agreeableness, emotional stability, narcissism, Machiavellianism, and psychopathy. Following the survey, OPF monitored the volunteers’ Twitter activity, examining things like frequency of tweets, number of retweets, and number of followers.

After it collected the data, OPF posted two open competitions on Kaggle, one aimed at identifying the best model for predicting personality traits based on Twitter activity, and the other focused specifically on models that can identify individuals showing high degrees of psychopathy. The competitions, which ended on June 29th, received over 200 submissions between them; no winners have been announced yet.

With more individuals becoming immersed in social media sites, it is important to understand how our actions online appear to those outside our social circles, and what that means for users’ privacy. Part of OPF’s mission is to tackle these issues.

“As an organization, one of our research goals is to understand just how well personality can be predicted by activity on social network sites,” OPF co-founder Chris Sumner wrote in a forum post linked to the competition.

Last year, Sumner’s group conducted similar research on Facebook users’ activity. The takeaway from that study? If you attempted to figure out a person’s personality by looking at their Facebook, your conclusion would often be incorrect. Stay tuned for the results of the Twitter survey.