Intelligent Systems Division

Leveraging big data to examine the causal effects of weather on emotions

Tuesday, April 24, 2018, 11:00am - 12:00pm PDTiCal
6th floor large conference room
This event is open to the public.
AI Seminar
Nick Obradovich, MIT

Does the weather actually alter human emotional states? On first glance, it sure seems to. But existing empirical evidence on this question is inconclusive due to factors -- like selection, confounding, and previously limited data -- that make causal inference quite difficult. To address these issues, we leverage tools of causal inference drawn from climate econometrics and employ over three and a half billion social media posts from tens of millions of individuals from both Facebook and Twitter between 2009 and 2016. We find that bad weather worsens the emotional content of human expressions. In this talk, I'll walk through the factors that bedevil estimation of the causal relationship between weather variables and mood, outline our causal identification strategy, and discuss the magnitude and social significance of our results.

Nick Obradovich is a computational social scientist at the MIT Media Lab. He studies the social impacts of climate change and has investigated the potential for warming to impact human sleep patterns, emotional states, mental health quality, physical activity rates, and governance capacity. He is currently examining the effects of exposure to hotter temperatures on human communication patterns and on the structure of social networks.

« Return to Events