Election Day Weather Reports

The folks at Harvard's Kennedy School have apparently put up a website that monitors voting conditions in every state. This brings up a questions I've always wondered about the media's fascination with weather conditions on election day.

Every two years I hear reports about what the sky is going to look like tomorrow and how this is going to have an effect on turnout. I have to say, I'm skeptical about the relevancy of it in the first place. Don't get me wrong, it many cases, I'm sure weather plays a role in final turnout. If a bad snow storm hits, I could understand how that could keep some people home, and I do believe that primary caucus turnout would be subject to particularly bad weather. Local races of low interest also might be somewhat vulnerable, but I really don't see how rainfall would have any effect on a high-profile Presidential race. After all, most people drive to the polls, not walk, and if you live in an area where it's raining on election day, it probably means you live in an area where you're used to rain and drive in it all the time. I know there are a lot of less-than-passionate voters out there, but does anybody really stay home for that reason?

I would like to think that somebody has done some research on this to try to see if there is any quantitative evidence to back it up. Of course I could see a lot of problems with doing a completely accurate study. Simply comparing areas with heavy rainfall vs. those that were sunny wouldn't take into account a whole host of other important factors such as historical participation habits and local ground operations. (Urban environments in competitive areas would be subject to more investment from respective campaigns, and therefore, see a more intense turnout drive).

I imagine however, you could compare a number of areas, maybe by county,(to be safe, maybe divide urban vs urban and rural vs rural)measure their turnout vs. national turnout, then compare that number to the historical average of that county in previous elections. Even then you would only be able to compare Presidential year elections as off-years would be subject to fluctuations in competitiveness of downballot races. This is also making an assumption that the importance of a competitive downballot race plays no effect on overall turnout when there is a Presidential race at the top of the ticket. That's a somewhat risky jump, but seeing as significant statistical evidence of a drop off in voting for downballots is always present, I think it might be an acceptable and necessary omission. Plus for the sake of this study, we would be concerned mostly with dispassionate voters, who would be the most likely to only vote at the top of the ticket. This study might be made more difficult with the emergence of early voting, which allows voters to avoid election day variables. It could be argued however, that early voters are far more likely to have a high interest level, and show up regardless of weather conditions.

If this is a reasonable thing to look at, I wonder what the numbers would look like? My guess is that you might find some small discrepancy, but anything under 5 percent could easily just be random fluctuations that could exist for a number of reasons. But I suppose that's my point, this is all just speculation and yet we treat it as something else.

Update: I think there is a group that could be potentially influenced by rainfall - elderly voters who don't have a ride to the polls. I still see this as somewhat suspect as both parties will ensure their seniors have a ride to the polls as one of the most crucial steps of any ground operation. However I will concede that there could be a number of seniors who would feel comfortable driving in normal weather conditions, but not in the rain. I don't know if this would be offset by the fact that seniors are significantly more likely to vote than other age groups however. Worth noting.