Week 9: Special Interview and 2016 Presidential Election

Hello! On Monday, I had the incredible opportunity of interviewing our Head of School about his opinions on various voting methods and several examples of how different elections play out on Phillips Academy's campus. Questions that I asked him included:

1. In what locations have you voted?

2. When you've voted in those areas, have you used any other voting methods/electoral systems besides plurality?

3. What do you think about the plurality (simple majority) method?

4. Are you familiar with other voting methods besides plurality?

5. If so, what are your thoughts on the other methods? Which method is the most fair/equitable/accurate (these can all be different, however you interpret the definitions of these words), in your opinion?

6. How would you define "fair,""equitable," and "accurate" in the context of voting methods.

7. Could you tell me about different voting methods that are used on campus?

Our Head of School acknowledged that he didn't know too much about the ins-and-outs of plurality and other systems, but his comments were still quite insightful. He also mentioned the use of a Borda count, approval voting hybrid in elections in which faculty vote on a school schedule. I found this fascinating and will follow up with various faculty and staff he mentioned were part of the scheduling committee.

Another great activity I did this week was something I've been looking forward to since the beginning of the term: applying presidential election results to different voting methods and seeing what the outcomes are! I decided to start with the 2016 presidential election and it ended up being way tougher than I had expected.

First of all, I had to predict how voters would have ranked all of the candidates, even the smaller party candidates.

The preference profile in the above picture was my first try at deciding how voters would vote. You can probably see, but there was a lot of erasing and rewriting involved! I then used this data and applied it to instant runoff, Borda count, and the Condorcet method. I found that Hillary Clinton beat Donald Trump in the instant runoff election by a margin of 3.07%. Then came the interesting parts: I somehow found that Jill Stein would have won the Borda Count and Gary Johnson using Condorcet's method. I knew that I must have done something wrong, because I was comparing my answers and data with the ones professionals at Civis Analytics calculated (https://www.vox.com/policy-and-politics/2016/11/25/13733322/instant-runoff-ranked-voting-2016). At a meeting with my mentor, I showed her the preference profile I had created and the results I got. She noticed that I had blocked a lot of voters' preferences, i.e. not all Clinton voters would have ranked all of the candidates in the way that I decided above. So my next step was making the preference profile more nuanced.

The table above is the second one I created. However, as you can, I found that now Gary Johnson would have won the Borda count election. Thus, I'm still continuing to revise the preference profile, and will eventually post the final result!

As this course is wrapping up, I am beginning to work on my final presentation, which will encompass everything I've learned so far, as well as some reflection as to how the course went!