Read on Twitter
This absolutely blew my mind last night. HT to @magi_jay for putting me onto this. By looking at the exit polls (they are flawed, but still helpful), it seems the common belief of there being a large age gap in voting behavior is exaggerated, and it's really a racial gap. Thread:
At first look, it appears that there is a large age gap in voting, especially among 18-29 year olds. As you can see, they supported Biden by 60-36, while seniors supported Trump 52-47. But CNN has a very helpful feature which allows you to break out exit polls by race.
When you do that, it changes a lot of things. If you look at the age breakdown for whites, the margin is significantly less. Trump won all age group of whites, including young whites by 53-44. His victory among seniors was only slightly higher, 58-41.
Among black and Hispanic voters, this gap is non-existent. In fact, younger black voters were more likely to support Trump than older black voters. So then with limited white white voter age gap, and no hispanic/black voter age gap, how is the age gap so large?
The answer to that is the racial makeup of age groups are much different. Looking at the exit poll breakdowns, among white voters, 16% were 18-29 while 27% were 65+. Among black voters 21% 18-29, 16% 65+ Among Latinos, a whopping 31% 18-29, only 9% were 65+.
Simply put, 18-29 year old voters in this country are a much more racially diverse group of voters than older voters, which are much more heavily white. This is the changing demographics that you always hear about. And it's more responsible for age group differences than age is.
That is not to say there's not some age variance between young and old (mainly among whites), but it is not nearly as much as people think. So what does this mean? I think it affects a lot how we look at campaigns. You hear a ton about Democrats needing to turn out the youth vote
Which is certainly true to an extent. But Democrats need to focus that message much more to turning out non-white youth, rather than white youth focused messages, since it is not that big an advantage over older whites. It also calls into question a lot of assumptions
We always assume that younger people are much more liberal than their parents. And there may be some truth in that (ideology does not always track vote preferences), but often people are actually voting close to their actual parents. Younger whites voting like older whites, etc.
Figuring this out changes a lot how I'll look at analysis too. For example, in the GA runoffs, the electorate is still looking on track to be close to as diverse as November, possibly more, but it's looking to be much older. This isn't as bad for Dems as thought.
Since race, rather than age, is more correlated to voting patterns. Again, it's clear that at least among whites, age does play a modest impact. Just not as big as thought. I didn't know this, and it blew my mind. Maybe it will interest you all, or not. /fin.
