Beyond Basic Statistics: Tips, Tricks, and Techniques Every Data Analyst Should Know

3CONSERVATIVES, LIBERALS, AND OTHER POLITICAL PAWNS: HOW TO GAIN POWER AND INFLUENCE WITH SAMPLE SIZE CALCULATIONS

The 2012 US presidential election was full of drama. Conservatives beat up on incumbent Democrat Barack Obama, claiming a weak economy and shaky foreign policy were solid reasons to boot the president out of office. Liberals hammered the Republican challenger Mitt Romney, accusing him of being too rich and cold-hearted to understand the concerns of the average American. The polls were up and down, often favoring the president but also showing steadily increasing support for the challenger as the election drew near. This political drama culminated on election night, when conservative commentator Karl Rove, reporting for FOX News, refused to believe swing state Ohio had actually gone for the president. The incredulous Rove stood up and walked to a back room behind the set to consult with the data crunchers. He only relented when they insisted they were “99.9% confident” of their predictions.

Karl Rove wasn’t the only person who was confused. The political polls were confused, too. In the days leading up to the election, two of the nine major polling agencies had Romney winning, three had Obama winning, and four had the candidates in a dead heat. In the end, however, Obama won by almost 4% points, winning 51.1% of the vote to Romney’s 47.2%. Not a huge margin, but in this era of divided politics, it was a decided victory for the president.

Unlike the political pundits ...