share of the electorate

Update: CIRCLE has revised this number down to 66 - 32%. Still quite impressive and the largest margin for any age demographic.
---------------------

CIRCLE has released data on the youth share of the electorate and the partisan margin of the youth vote. The numbers are base on the National Exit Polls, but they look fantastic. Young voters are choosing Obama by 38 34 percentage points - towards the upper end of all the polling we saw this year.

Young voters preferred Obama

Young voters diverged sharply from the population as a whole, preferring Obama/Biden over McCain/Palin by 68% to 30% 66 - 32% in the NEP. This is by far the highest share of the youth vote obtained by any candidate since exit polls began reporting results by age categories in 1976. In past elections from 1976 through 2004, young voters diverged by an average of only 1.8 percentage points from the popular vote as a whole. 2004 had set the previous record for an age gap.

CIRCLE also released data on share of the electorate. Young voters are predicted to increase their share to 18%. This may not sound huge, but remember that it's not a measure of youth turnout:

Young people (ages 18-29) represented 18 percent of the voters in today’s election, according to the early released National Exit Polls (NEP) conducted by Edison/Mitofsky. This is the one point higher than in 1996, 2000, and 2004, when young voters represented 17 percent of voters in each presidential election, according to the NEP. However, this number does not indicate how many young people voted or whether there was a rise in youth turnout. In recent elections, the youth share of the vote remained constant, because youth turnout rose at the same rate as the total turnout. See the table below for more information on the difference between turnout and share.

Remember, share of the electorate is not a measure of turnout. We'll have turnout numbers for the youth vote tomorrow morning.

For the data geeks, here's a note on CIRCLE's methodology:

There is no official count of voters by age. Therefore, any statistic on youth share or youth voter turnout is an estimate based on survey data. Like any survey, the National Exit Polls use methods that may introduce sampling bias. However, our estimates of youth turnout from the National Exit Polls (shown above) have produced a trend that closely tracks the trend in the Census Current Population Survey (CPS), which is the other reliable source for estimating youth turnout. Since CPS voting data for 2008 will not be available until spring 2009, our method produces the only current and reliable estimate of youth turnout.

Note - this post got front-paged on Daily Kos.

In 2004, youth turnout was wildly misreported - in the media and in the blogosphere. That reporting was summed up most aptly by this famous quip from the late Hunter S. Thompson:

"Yeah, we rocked the vote all right," quips Hunter S. Thompson, the gonzo journalist himself. "Those little bastards betrayed us again."

Of course Thompson, and the media reports, were wrong. The youth vote did turnout and was the only age demographic to vote for Kerry over Bush.

This year, expectations for the youth vote are higher than ever - perhaps unrealistically so - and the expectations game is already beginning to result in "youth don't vote" stories in local and regional media. For instance, in Florida, the Orlando Sentinel had this to say:

Young people are turning out in disproportionately low numbers. Though major registration efforts this year boosted their totals to nearly 25 percent of the total electorate, voters younger than 35 represent only 15 percent of early voters, making them the worst-performing demographic group in the analysis.

This is incredibly misleading. Here's what the Young Democrats of Florida found when they ran the numbers on early voting in Florida:

According to the Florida voter file, (which should be viewed as relatively but not 100% accurate) in 2004, approximately 392,888 voters between 18-35 voted early or absentee. So far 499,867 voters between 18-35 have voted early or absentee this year. This is a 27 percent increase over 2004.

What happened was a common mistake in which the media used misleading, and not terribly informative, "share of the electorate" data to describe youth turnout instead of more accurate figures like the hard number of votes or % turnout of eligible voters. Unfortunately, such mistakes are all too common in reporting on youth turnout.

The following is a hard nosed look at what we might realistically expect on Tuesday, a list of common mistakes the media makes when reporting on youth, and some tips to help activists, journalists, and bloggers alike accurately assess youth participation on Election Night.

What to Expect When Expecting on Election Day:

Youth Turnout Will Likely Be Higher Than in 2004:

There are three measures of youth participation:

• Total Number of Votes: That's pretty self explanatory.
• The Turnout Rate: This is the percentage of all eligible young voters who cast a ballot.
• The Share of the Electorate: The percentage of the entire voting electorate between the ages of 18 and 29.

This year, the hard number of ballots cast by young voters and the turnout rate are both highly likely increase. Let's keep that in perspective, though. Youth turnout is not likely to climb into the 60 or 70% range. The highest youth turnout ever was 55%, recorded in 1972. I would be extremely happy to see us match that number this year. Who knows, maybe we'll be surprised and it will be higher, but we shouldn't go into Tuesday expecting that it will be higher.

Even if youth turnout rises significantly, there is no guarantee that the youth share of the electorate will show a comparable increase.

This was the big problem in 2004: youth turnout rose significantly, but, because older portions of the electorate also increased their turnout rate, the youth share of the electorate held steady at 17%. It is highly possible that increased turnout among African Americans and other groups, or even decreased participation among depressed (young) McCain supporters, could prevent young voters from increasing their share of the electorate on Tuesday.

Again, this isn't to say that youth won't increase their share of the electorate, but don't be surprised if it holds steady at 17%. More importantly, don't use that "share of the electorate" figure as an accurate measure of youth participation. More on that below.

Don't Compare Apples to Oranges:

There are two measures of youth turnout from 2004 - those taken from national exit polling, and a more accurate measure taken from the Current Population Survey. While the CPS data is more accurate (and it is what you will find on most fact sheets from CIRCLE), it also does not come out until months after the election and uses a different methodology than exit polling. To ensure that we are not comparing apples to oranges on Election Night, it is best that, when measuring youth turnout, we compare the 2008 exit polls to the 2004 exit polls. Here are the exit poll numbers from 2004. Use these as your baseline when reporting on Tuesday's youth turnout:

18 - 29 year olds:

• Vote Count = 19.4 million
• Turnout = 48%
• Share = 17%

Common Mistakes (and Basic Facts) About the Youth Vote:

Some of these might be repetitive from above, but they bare repeating. Use these as a guide when reporting on young voter turnout on Tuesday night:

1. When reporting on youth participation, do not confuse "share of the electorate" with "turnout." Share of the electorate is a measure of the proportion of young voters who cast a ballot in relation to all other voters. Turnout is the percentage of all eligible young voters who cast a ballot. Share measures the influence of young voters within the electorate as a whole. Turnout tells us whether or not more young people showed up at the polls. Please do not confuse them.
2. It is possible for turnout to rise, while share of the electorate remains steady. Indeed, this is exactly what happened in 2004. Young voter turnout (18 - 29) increased by 9 percentage points from 40 to 49% (an increase of about 4.3 million votes). However, young voter's share of the electorate remained steady at 17%.
3. Young voters can only be held accountable for their own actions, not those of the entire electorate. If the youth vote's share of the electorate holds steady from 2004 to 2008, that will mean that older voters also went to the polls in higher numbers. Young voters cannot be held accountable for that. As such, turnout and the hard number of votes are the only accurate measure to gauge the success of efforts to get out young voters.
4. Rising youth turnout is a trend, not a fad tied to the popularity of Senator Obama. Contrary to conventional wisdom, or media reports from 2004, Obama's campaign is not solely responsible for higher youth turnout, though it has played a crucial role during this election cycle. Youth turnout began to rise in 2004, when youth it jumped by 9 percentage points, from 40 to 49%, and 4.3 million more young voters cast a ballot than in 2000. This trend continued in 2006, which saw the first increase in young voter turnout during a midterm election since the 1980s. It reached a new height in early 2008 when youth turnout in the primaries was double that from 2000, the last comparable year. In some states, youth turnout in the primaries was triple or quadruple that of previous years.
5. The margin of victory among young voters may be just as important as the overall increase in youth turnout. In 2004, 20 million young voters cast a ballot, with 54% selecting John Kerry. That gave Kerry an advantage of 1.6 million votes over President Bush among young voters. This year, if 22 million young voters cast ballots and 62% choosing Obama vs. 38% for McCain (numbers roughly found in most polling), that would give Senator Obama an advantage of 5.28 million votes.
6. Youth turnout is about access, not apathy. When young people are registered to vote - they turn out. According to the US Census, 81.6% of all registered young voters actually cast a ballot in 2004. That is on par with other portions of electorate. The more campaigns and independent organizations work to register young voters, and the easier we make the registration process, the higher youth turnout will be.
7. Regardless of youth turnout on Tuesday, young voters have already played a crucial and decisive role in this contest. In the Iowa Democratic caucuses, young voter turnout tripled and their share of caucus-goers was equal to that of the "reliable" 65+ demographic. Obama won the support of 60% of Iowa's youth, catapulting him to the front of the Democratic pack. Similar levels of support from youth in the following primaries and caucuses were the foundation of Obama's primary success.

In all likelihood, we are standing on the brink of an historic election, and we may well witness youth turnout unlike any we've seen in decades. Let's make sure that, whatever the final numbers, we have an accurate reporting of that turnout and don't make the same mistakes that so many reporters and bloggers made after our disappointing loss in 2004.

In 2007, just before the primaries, I created a "cheat sheet" for journalists to correct common misperceptions about the youth vote, and help journalist accurately report on young voter turnout. It was pretty well received and even got picked up for an anthology about new media.

Judging from some of the reporting we've seen this year, I thought an updated version geared towards the general election would be useful:

1. When reporting on youth participation, do not confuse "share of the electorate" with "turnout." Share of the electorate is a measure of the proportion of young voters who cast a ballot in relation to all other voters. Turnout is the percentage of all eligible young voters who cast a ballot. Share measures the influence of young voters within the electorate as a whole. Turnout tells us whether or not more young people showed up at the polls. Please do not confuse them.

2. It is possible for turnout to rise, while share of the electorate remains steady. Indeed, this is exactly what happened in 2004. Young voter turnout (18 - 29) increased by 9 percentage points from 40 to 49% (an increase of about 4.3 million votes). However, young voter's share of the electorate remained steady at 17%.

3. Young voters can only be held accountable for their own actions, not those of the entire electorate. If the youth vote's share of the electorate holds steady from 2004 to 2008, that will mean that older voters also went to the polls in higher numbers. Young voters cannot be held accountable for that. As such, turnout and the hard number of votes are the only accurate measure to gauge the success of efforts to get out young voters.

4. Rising youth turnout is a trend, not a fad tied to the popularity of Senator Obama. Contrary to conventional wisdom, or media reports from 2004, Obama's campaign is not solely responsible for higher youth turnout, though it has played a crucial role during this election cycle. Youth turnout began to rise in 2004, when youth it jumped by 9 percentage points, from 40 to 49%, and 4.3 million more young voters cast a ballot than in 2000. This trend continued in 2006, which saw the first increase in young voter turnout during a midterm election since the 1980s. It reached a new height in early 2008 when youth turnout in the primaries was double that from 2000, the last comparable year. In some states, youth turnout in the primaries was triple or quadruple that of previous years.

5. The margin of victory among young voters may be just as important as the overall increase in youth turnout. In 2004, 20 million young voters cast a ballot, with 54% selecting John Kerry. That gave Kerry an advantage of 1.6 million votes over President Bush among young voters. This year, if 22 million young voters cast ballots and 62% choosing Obama vs. 38% for McCain (numbers roughly found in most polling), that would give Senator Obama an advantage of 5.28 million votes.

6. Youth turnout is about access, not apathy. When young people are registered to vote - they turn out. According to the US Census, 81.6% of all registered young voters actually cast a ballot in 2004. That is on par with other portions of electorate. The more campaigns and independent organizations work to register young voters, and the easier we make the registration process, the higher youth turnout will be.

7. Regardless of youth turnout on Tuesday, young voters have already played a crucial and decisive role in this contest. In the Iowa Democratic caucuses, young voter turnout tripled and their share of caucus-goers was equal to that of the "reliable" 65+ demographic. Obama won the support of 60% of Iowa's youth, catapulting him to the front of the Democratic pack. Similar levels of support from youth in the following primaries and caucuses were the foundation of Obama's primary success.

In thinking over my last post about Gallup's estimates, I had a horrific thought. What if youth turnout skyrockets but, because turnout is again up among all demographics, the youth share of the electorate decreases?

That would be a communications nightmare. Let's pray it doesn't happen.

In addition to their excellent fact sheet on voter registration and election laws earlier today, CIRCLE also released a fact sheet providing the definitive data on youth turnout in 2000 and 2004, and youth demographic data for 2008. The page also links to an interactive flash map that breaks the youth vote turnout data down on a state by state basis.

Every reporter, blogger, and youth advocate should have this page bookmarked.

And here are the demographic breakdowns of the youth vote for 2000, 2004, and 2008: