Wednesday, July 26, 2006

Turnout: the registered youth of this great nation are temperamental, not lazy

The issue of youth voting got me curious. Everything I've read is so depressing... 'youth voting down,' 'where are them dang kids?', 'waste of time looking for American Youth,' etc. No wonder they don't want to vote, everyone crabbing at them about it -youth are temperamental.

The previous post in this series took a look at research that indicated youth were just as responsive to GOTV efforts as adults, IF these youth could be contacted at the same rate as adults. The problem is that these dang youth are elusive and move around a lot. So, let us look at the turnout of youth using a measure that, in a crude way, might isolate youth who have been contacted at high rates recently: turnout as a percentage of registered voters.

We will look at the following groups youth (18-24 years), young adults (25-44 years), and adults (45-64 years), using Census Bureau data from the Current Population Survey (CPS), namely:

Table A-1. Reported Voting and Registration by Race, Hispanic Origin, Sex and Age Groups: November 1964 to 2004
http://www.census.gov/population/www/socdemo/voting.html

We won't look at elderly voters in this post. The terminology used here may be idiosyncratic. I don't know if 44 year old people consider themselves to be young adults. But I've heard 60 is the new 40, so it might be OK.

The CPS data have some problems. It is survey data, rather than a compilation of actual voting and registration records. So, people tend to say they are registered or have voted when they did not and the turnout rates tend to be higher using CPS data instead of FEC data. But the FEC does not keep track of much demographic data, and CPS data have been validated for looking at patterns over time and between groups at the same time.

Graph 1.--Turnout as a proportion of registered voters: youth, young adults and adults, 1972-2004.



Graph 1 above shows turnout as proportion of registered voters in each age group. The turnout of registered youth for recent presidential elections seems to be not so much low as very volatile. It took one presidential election cycle to go from the highest it's been in twelve years to the lowest since 1972. Then it took four years to go from the second worst turnout ever to almost the level of 1992. There does seem to be a long term decline in turnout for off-year congressional elections, though. These are the same patters we saw for the 18-20 year old turnout as a function of Voting Eligible Population in the previous turnout post. What is interesting here is that the young adults, who one would like to think are more mature and regular about things, show the same pattern as youth. In fact, relative to past performance, the drop in young adult turnout after 1992 is even more pronounced. The variation in percent turnout is less than for youth, but youth adults are a larger group, so it may be even more significant in terms of votes. The adults are the most stable of all. Let's use the adult turnout as the reference gold standard, and see if there are any patterns in the ratio of youth and young adult to adult turnout.


Graph 2.--Ratio of youth and young adult turnout to adult, registered voters, 1972-2004.

Graph 2 shows the ratios. I think there is a pattern that confirms the impressions from Graph 1. If we use the adult turnout as a gold standard, then for presidential elections, we see a long slow decline after 1972, followed by a volatile period starting in 1992. The good news is that youth turnout can be large, and, as a percentage of adult turnout, come close to 5 percentage points of turnout in 1972. What I think is interesting is that young adult turnout follows almost the same pattern. So, for presidential elections, if the registered adults are interested enough to turn in large numbers, so are the youth and young adults.

Also, the congressional elections turnouts of both groups don't look so gloomy. There may be a slight recovery in turnout of registered voters in 2002, however that is only one off-year election, and may be due as much to a decrease in adult turnout following 1992 as increased youth and young adult enthusiasm.

So, at least for the registered population the recent history suggests that youth are volatile more than apathetic or dilatory. At least that is how I see it. So, for elections, the bottom line is that registered youth may turnout in surprising large numbers, at least if the election is compelling enough to attract a large proportion of adults. So the question is, who will they vote for? Interesting also that young adults look like a more stable version of the youth.

previous post in this series