Is a photoshop exercise trying to tell me something?
I thought it would be fun to look through the main hashtags for the two presidential political campaigns and do a little photoshop exercise.
The methodology: I went to #MakeAmericaGreatAgain and #ImWithHer on Twitter looking for supporters of each candidate. I pulled the first 30 or so avatar images from each and made sure to pull an image of each candidate. The only criteria I used was that it appeared to be a picture of the user and that they were generally facing forwards. I imported each layer into photoshop, reduced transparency to anywhere from 15–25% depending on the image. With an image of a person with sunglasses, for example, I kept a lower transparency level so they wouldn’t overpower the image. I then aligned everyone by the eyes and mouth. After each, I placed the candidate’s picture on the top and lowered to about 10% transparency.
I only did Trump at first. To be honest, I was just curious if I would end up with an old white guy. One interesting thing is that my images were not all old white guys. There were a few minorities, several woman and many young men. When I posted the initial results to my FB page, I was asked if I could also do one for Hillary. Reactions ranged, but overall people seemed amused. One of my friends then asked if I could display the images without the final overlay of the candidate. Unfortunately, while I had the Hillary file open still, I hadn’t saved the Trump file. But this just meant that I was going to get to do the Trump exercise again with a whole new batch of pictures!
I lost a bit of youth on the second round. Though, I will note that this face that emerged on the Trump side was not driven by any particular group of pictures. I kept trying to turn on and off certain layers or people that I thought it resembled and, while the hair color and outfit would adjust prominence, the face didn’t change.
As for the Hillary supporters, I really liked the face that had emerged. It seemed friendly and inviting and peaked my curiosity. I toggled on and off layers until I was able to isolate the four avatar images that really drove the face that had emerged. I then broke them out:
As this was a very tiny sample-size of people, I would hasten to draw any conclusions here. But it was a fun exercise and very interesting to see the final results.