It happens every election cycle. Political consultants, campaign managers, and even candidates think that they know best when it comes to running a campaign, creating a message, and targeting voters. However, most of this so-called “wisdom” comes from years of doing it the same way without having anyone challenge their methods. We always hear that “this is always the way it has been done” or “we have been successful with doing [insert method here] in past campaigns”. But have they? As far as I know, nobody in party politics have tested to see if their particular campaign methods either work or fail. They do not exclude rival explanations, thus they do not get an accurate assessment of what does and what does not work in political campaigns. Basically, people simply guess as to what method works or not.
In political science, as well as in any science, we try to determine the difference between correlation and causation. Unfortunately, many political consultants and campaign managers do not understand the difference between the two. For example, a political consulting firm that does direct mailing might say that they are successful because eight of the ten campaigns they provided direct mail pieces for actually won. But was it the direct mailers? What about other factors? Were the eight winners incumbents? Were they in more partisan districts that favored their candidacy? Did a good debate performance help? As for the losers, were they horrible candidates? Did they have criminal records? Were they outspent 10-1? Were they unknown to the voters? Were they against the incumbent?
The above example shows the difference between political scientists and people working in politics in general. Political scientists seek to understand why things happen. Political consultants do now ask this question, but simply say that the direct mailers were successful because there is correlation between their clients win-loss records and their services. For political scientist, this is not an acceptable answer. Scientists ask questions, test hypotheses, and try to figure out why things happen. They do not simple observe.
If we look at science and direct mailers, Green and Gerber (2004) conducted experiments to determine what campaign methods work. As for direct mailers, Green and Gerber found that it took 200 direct mailers to persuade one voters in a partisan election. As far as robocalls, they found that they are absolutely useless. These two didn’t formulate an opinion to determine what works, but tested hypotheses and found out what really did work.
This way of thinking is what the Florida Democratic Party needs. Why is this the case? Well, let me explain.
First, a political scientist has spent years reading scholarly literature. This literature focuses on a number of subjects, such as messaging (with Lynn Vavreck providing some great studies), voting behavior (with Michael Lewis-Beck being the top of his field), and campaign tactics (again, Green and Gerber). In order to be published these scientists need to have their works peer reviewed. As someone who as gone through the peer review process, it is extremely brutal. Every aspect of your methodology is questioned. Only after much scrutiny is your article published. Therefore, it is important to have people working in party politics who actually understand what works and what does not work because they know the scientific literature.
Second, having a political scientist would save a political party money. Let’s go back to the robocalls. Political scientists have found that they are basically the most useless thing that you can do in a campaign. Yet, thousands of dollars are spent in Florida by Democratic candidates, and probably the Florida Democratic Party, for something that is utterly useless. A political scientist who is immersed into the political science literature would be able to advice on the most cost effective methods of campaigning.
Third, political scientists seek to understand voters. The disconnect between academia and the practical world is great on this point. And, again, the practical word goes off of assumptions, while the academic world tests hypotheses. And even when those in media and political campaigns think they understand the position of academia, they are incorrect. The perfect example is Chris Cillizza of the Washington Post. He stated in an article last year (that I am currently trying to find) that if you ask academics about issues, they would say that they are important. Well, actually no. Most political scientists think that left-right issues are not really important when it comes to vote choice. Yet, how often do we hear political consultants talk about issue position? Yes, in many cases they are just wasting oxygen. Political scientists, again, spend time asking question and trying to figure out why voters vote the way they do.
Fourth, political scientist can benefit a political party by constantly conducting research. Most, if not all, political scientists are not satisfied with the status quo. For some reason, there is an attitude in political science to try to prove other political scientists wrong, unless they work closely together. It is this drive that seeks to improve on models and help expand the current literature in the field. In the case of the Florida Democratic Party, a political scientist would be extremely beneficial. For years, Florida Democrats have complained that the party continues to do the same thing over and over again without success. Political scientists are looking for the next best thing out there.
Knowledge of what works, saving money, understanding vote choice, and testing methods to increase total votes are just some of the benefits of having a political scientist. Just having one scientist in the Florida Democratic Party would drastically change the direct of the party, as well as the effectiveness. However, having two political scientists, constantly competing against one another (since that is what we do) would provide the party with more options.
Are there some drawbacks to having a political scientist? Yes, but mainly just one. The main criticism that many, including me, have had about pure political scientists is that they have no practical experience. These individuals sit in their ivory towers and are unaware of what is really going on in the political world. As a result, we have a lot of political science literature out there that is basically useless. For example, I was at a presentation of a paper to be published by one of the leading political scientists in Canada, where he and his co-author discussed whether having a “sense of duty” leads to higher turnout rates. But to those of us in the practical world, we would basically find that information useless. How would campaigns or political parties be able to measure “duty”? Are we going to go up to people and ask them their “duty level”? Yes, even the best political scientists provide us with literature that is useless.
In addition, many in political science nowadays really don’t know politics in general. Many are trying to get academic jobs and are good at research methods. Instead of using methodology to show how well they understand political science, they use political science to show how well they know methodology. Therefore, it is important to have political scientists that have been on the ground and have worked in campaigns when hiring someone for a political party. Simply picking a political scientist is not sufficient.
Yes, this is a long article, but it is something I have been passionate about. Whenever I hear a political consultant talk about spending money on a campaign tactic that has been empirically tested as being impractical, I want to put my head through the wall. Yes, campaigns are being run really, really, really poorly because political consultants have no idea what works and what doesn’t work.
If Democrats can believe in science when it comes to climate change, then it is about time we rely on scientists when it comes to politics.