Beer and politics: How well do they really mix?
Recently a local news outlet reported about a study that claims that the type of beer that a person drinks directly correlates with which political party they vote for and how likely they are to vote. The outlet claiming to be “skeptical” took a version of a similar poll on Mill Ave in Tempe. The video, which can be found here, shows the newscaster asking a whole four people about their beer and political preferences. Despite the extremely small sample size and a whole host of other biases that were not accounted for, the newscaster was still able to find one outlier, who was a democrat that likes a “republican” beer.
I’m pretty sure by their tone the newscasters on this story weren’t taking it seriously and saw it as a fluff piece, but I think this is a great opportunity to demonstrate how skepticism really works. Let’s start with the original story, which shows a graph that shows how a particular brand of beer is preferred by democrats or republicans and how likely they are to vote.
The first question I wanted to ask was what about people who don’t drink beer? How are they counted? The second question is what about people who vote for candidates in other parties? Unfortunately the data that was used to make the graph doesn’t seem to be publicly available. All we do know is that 200,000 Americans were interviewed about their beer and political preferences. Without knowing the methodology of this study a lot more questions come into play. For example: How were these interviewees chosen? If they were chosen based on primary election voters, then the data can be skewed toward people who identify as either republican or democrat. Also, how was beer availability controlled? Not all the beers listed would be available in all regions, for example Yuengling isn’t available within 100 miles of Phoenix.