Are you finding it hard to concentrate on work during these sunny, summer days? Do you find it easier to buckle down, concentrate and be productive when it’s rainy out and you have no choice but to stay inside? If you do - and you’re a programmer - then, it turns out, you’re not alone.
Data scientist Dr. Jason Davis recently did an interesting analysis of the effect of rainy days on programmer output. Using data from GitHub, Google Maps and Weather Underground he found that more people checked in code on days when it rained, than on non-rainy days. You statistical modeling geeks will want to dig through his detailed post on the research for all the juicy details and charts.
In a nutshell, Davis used the number of unique people checking code in to GitHub on a given day as the measure of programmer activity. Using data from March 2012 through June 2013, covering the 1,000 most popular locations in GitHub (which gave him just shy of 27,000 data points), he used linear regression to control for a number of factors:
Whether the day fell on a weekend (there are fewer check-ins on weekends)
Whether the day fell during the Christmas/New Year’s Holiday season (there are a lot fewer check-ins during this time)
The number of days since the first day in the dataset, to control for the general increase in GitHub activity over time
The average overall check-in count for several days before and after each day, to control for general fluctuations in activity
Using the data for non-rainy days, he then estimated (“trained”) the model and predicted check-in activity for rainy days. Comparing these predicted results with the actual values, he found that, on average, there was a 10% increase in the number of people checking in code to GitHub on days when it rained over non-rainy days. Davis found this result to be significantly significant.
It makes sense that people are getting more done on rainy days, unless you buy into the theory that rainy days just make people want to stay in bed. But maybe programmers are a more motivated bunch than average people? That’s a topic for another day.
Read more of Phil Johnson's #Tech blog and follow the latest IT news at ITworld. Follow Phil on Twitter at @itwphiljohnson. For the latest IT news, analysis and how-tos, follow ITworld on Twitter and Facebook.