Privacy Note: SameGrain is a privacy-focused social media platform where anonymity is supported and promoted. The data presented in this blog is anonymized, having any attribution to individual users removed.
We've all heard that you shouldn't talk politics at work. But with a 2016 election cycle that has already seen its share of headlines, it may be hard to avoid. For this blog post, I investigated the intersection between our political and professional lives, and as it turns out, what you do for work can be a good indicator of who you vote for.
Teachers Like Clinton, Police Like Trump
SameGrain (available in the Apple App Store) introduces users to people who are like them in millions of potential ways, including their political affiliation and profession. We asked SameGrain users who they were considering voting for in the 2016 US presidential election, and compared users' choice for president to their professions. The results are found in the interactive plot below. Results are limited to the current five leading presidential candidates: Clinton, Sanders, Trump, Cruz, and Rubio.
For each profession, the length of the bar indicates how likely a supporter of the candidate is to have that profession, relative to the average user. For example, if you set the plot to Bernie Sanders, you will see that his voters are 1.7 times more likely to work for non-profit, volunteer, or social service organizations compared to the average user. You can compare two candidates by using the drop-downs above each panel.
Unsurprisingly, Sanders and Clinton (both Democrats), have similar demographics. They both receive higher than average support from the visual art, education, entertainment, fashion, technology, publishing, non-profit, and recreation industries. Clinton enjoys significantly more support from those in government, finance, and law than Sanders, while Sanders captures slightly more students. With a combined 63% of profession categories having average or better support, the Democratic candidates can claim relatively wide support across many professions.
The Republican candidates' (Trump, Cruz, and Rubio) voters are dominated by executives, law enforcement, public safety, the military, and transportation professionals. Perhaps due to his experience as a building developer, Trump uniquely enjoys above average support from construction industry professionals. The largest differences in support between the political parties is with law enforcement and members of the military. The average voter for the Republican candidates is approximately 4 times more likely than a Democratic voter to be in law enforcement, and approximately 3 times more likely to be in the military. We found Republican candidate support was not as wide across the professions as Democratic support; 51% of professions had average or better support for the Republican candidates (recall Democrats had 63%).
While Trump has the highest support of any single profession (Trump's relative support from construction and law enforcement lead all other candidate-profession pairs), his support from educators and teachers ranks as the lowest support of a candidate from any single profession. Indeed, Trump's controversial and divisive reputation is born out in the data. One measure of a candidate's divisiveness is the standard deviation of relative support amongst all professions, which essentially determines how well all professions agree. The grey band in the plot shows the standard deviation of each candidate; narrow bands indicate agreement among professions, while wider bands indicate more disagreement. Trump and Cruz have the largest standard deviation, and are thus considered more divisive by this measure.