People are happier when the sun in shining. Not shocking, right?
But what might be surprising is just how extensively good, bad or ugly weather conditions can either boost or drag down people’s sentiments.
New research led by Stanford University environmental economist Patrick Baylis relied on billions of social media posts as its data source, and proves for the first time on a large scale that there’s a strong connection between meteorology and our moods.
“We find substantial evidence that less ideal weather conditions relate to worsened sentiment,” Baylis and his coauthors said, also noting that there’s “some observational evidence that the weather may functionally alter human emotional states.”
And the connection is a remarkably powerful one — rotten weather can sour our outlook as severely as a terrorist attack or an earthquake would, researchers found.
The study, written up in MIT Technology Review, looked at 3.5 billion social media posts from tens of millions of Facebook and Twitter users. It found that there’s a significant increase in negative sentiments when the weather is bad — whether that be too hot, too cold, too wet, too humid or too cloudy.
MIT Technology Review called it “the first large-scale evidence that our moods depend significantly on the weather.”
Baylis’ team began its research by assuming that they could assess people’s moods by using the way they express themselves on Facebook and Twitter as a reasonable yardstick. That means they assumed happier posts indicated better moods, while sad or negative posts suggested foul moods.
Critically, researchers didn’t just analyze posts that were about the weather itself, according to Quartz. They looked at everything users posted — whether they were griping about a downpour or not.
By looking at positive or negative keywords in Twitter and Facebook users’ posts, researchers could then link those positive or negative words to where each post came from and when — and then look at the weather conditions in their location when the post happened. That allowed Baylis and his colleagues to study how social media users’ outlook fluctuated as their weather did.
Then they compared weather’s impact to other events, including an August 2014 Bay Area earthquake and the San Bernardino terrorist attack in 2015. And researchers found something intriguing: The change in sentiment those events caused was comparable to the change spurred by bad weather, according to MIT Technology Review.
How could the research be helpful?
It could help companies target their marketing with a better idea of how the weather might make audiences more or less receptive to it, according to the MIT Technology Review. The research could also make it easier for “intelligent personal assistants” like Siri and Alexa to adjust their demeanor based on how the weather is influencing people.