Update: The Average Donald Trump Supporter
From Facebook’s Insights, Comparing May 2016 to Dec 2015
Back in December, I wrote this article (which got over 44,000 views), which showed some insights about the Donald Trump Facebook audience that can be extracted from Facebook’s Audience Insights tool by anyone with a Facebook ad account. I decided to quickly compare what has changed in the data. There are plenty of other things you can extract, and I didn’t compare everything that I got the first time around from the tool, since that would take a lot more time (hopefully some might have interest in doing that themselves).
Age and Gender
It still skews very male (57% vs. 46% of US Facebook users), but slightly less so than before when it was 60% male. The most dramatic shift is in the age profile, which is now more 25–44, in line with the overall averages.
This was the profile back in December:
Activities and Devices
A more credulous group? Decide for yourself, as there is a massive increase in the number of ads clicked on Facebook amongst this crowd: up to 44 from 29 previously. Post likes and comment figures are similar (22/32 vs. 25/33 in December) but page likes are way down from 52 to 37.
Old data here:
Far more desktop and mobile, and desktop-only internet users now, and Android users have increased dramatically as well.
Income and Household Size
An interesting shift here, which is to be expected given the change in nominee status — more lower-income ($30–50k) but also a spike in upper-income, $250k+ annual income versus really under-indexing at the end of 2015.
Old data:
Job Titles
Veterans and Military (which were 1 and 2 in December) are no longer at the top of the list among Trump Facebook fans.
The old data is here.
Top Categories of Interest to Trump Fans
You can see some changes, this is what it looks like now:
And this is what it looked like before:
Page Likes
Fewer personalities, more publications, seems to be the trend…
Vehicles
Still not the hybrid crowd, and actually even more so now…
Previous data, from December 2015: