by Scott Winship
Today you are rewarded for waiting all weekend to finally – FINALLY – discover the empirical truth about the netroots. You’ve marked it in your Franklin-Covey day planner, you’ve canceled your morning appointments. Some of you probably woke up hours ago to make sure you got this as soon as possible. One or two of you may be on the verge of a breakdown from the anticipation. And tomorrow, more than a few of you will be hung-over from celebrating the arrival of another…Data Day!!
I realize that in truth, this is only going to marginally help the fact that it’s Monday, but I’m here to do my little part. In my seemingly never-ending quest to produce an accurate picture of the netroots (see here and here), I believe I am nearing the end of the journey.
Last year, the Pew Research Center for the People and the Press published a nifty report on Howard Dean’s supporters called, “The Dean Activists: Their Profile and Prospects”. You can view the report here, and you can even download the raw data used to produce it, which I’ve done for this post. The fragrant folks at Pew obtained the Democracy for America database from the Dean folks and drew a random sample from this list of contributors, MeetUppers, and volunteers. The sample is big enough that it’s a lot more useful for our purposes than the Pew survey I messed around with last week. So without further ado, let’s see what we can learn.
First, uh, some further ado. While the survey is large enough that we can make meaningful statements about the characteristics of the “Dean netroots”, the response rate in this survey leaves much to be desired. For the questions I am looking at, just 13 percent of those contacted completed the initial survey. While this is nearly twice the response rate of the BlogPac survey I refused to take seriously last week, it is still awfully low. Without any additional information about who participated and who didn’t, we would have every reason to believe that the two groups might differ in important ways that prevent us from generalizing to all members of the Dean netroots.
But to its credit, Pew took advantage of additional information provided by Democracy for America to consider whether its sample was systematically biased. It turns out that survey participants were more engaged in the Dean campaign than nonparticipants (in terms of contributions, participation in MeetUps and the like) and were particularly more likely to have made campaign contributions. When Pew categorized survey participants by campaign engagement, they found that engagement wasn’t related to political views or to most demographic characteristics. The exceptions were that highly engaged people had higher incomes and more education, were older and more politically experienced, and were more likely to support Dean’s opting out of public financing. Taken together, this evidence implies that the Pew sample is a little older and advantaged than it ought to be and more politically savvy.
To keep consistent with last week’s analyses, I define the “Dean netroots” as those who were liberal or Democrats, who “regularly” relied on blogs for news, and who participated in at least one campaign activity. The difference this time is that they also had to be in the Democracy for America database as of late 2004. My results indicate that 16 percent of “Dean activists” (those in the DFA database) met this definition for inclusion in the Dean netroots. Dean for America included over 600,000 supporters at the peak of his presidential campaign. If we assume that Democracy for America’s database included 650,000 adults in late 2004, then my estimates indicate that around 100,000 adults – or one-twentieth of one percent of them — were members of the Dean netroots in late 2004. The entire group of Dean activists comprised 0.3 percent of adults, which is close to the figure for the “Democratic netroots” from last week. I’ll discuss both groups in what follows, but I’m going to simply refer to “Dean activists” and “the netroots” to make for easier reading.
OK, let’s look at ‘em. I’m going to splice my results in with those reported in Pew’s study, which incorporates other surveys they’ve done to characterize Democrats and Americans as a whole.
While 60 percent of Democrats were women in late 2004, men were just as likely as women to be Dean activists, and they made up 60 percent of the netroots. One in four members of the netroots was under 30 years old, making the group younger than Dean activists, Democrats, and Americans taken as a whole. People under 30 made up twenty percent of each of those groups. Just 11 percent of the netroots was at least 60 years old, compared with 22 percent of Dean activists, 26 percent of Democrats, and 22 percent of Americans. So much for the claim that the elderly are overrepresented in the netroots community. Indeed, one in four members of the netroots was a student.
The netroots was also unrepresentative in terms of race. While 80 percent of Americans and 70 percent of Democrats were non-Hispanic whites in 2004, 90 percent of Dean activists and netroots members were. Blacks and Hispanics were quite underrepresented.
Among Americans as a whole and among Democrats, half of adults had no more than a high-school education in 2004. One in four was a college graduate. Contrast this with the netroots’ seventy-percent college graduation rate, which was possibly lower than the rate for Dean activists as a whole. One-fourth of the netroots had a graduate degree.
These educational differences, not surprisingly, are reflected in income differences as well. While a third of American families and forty percent of Democratic families had less than $30,000 in income in 2004, that was true of only 15 percent of Dean activists and netroots members. In contrast, 30 percent of Dean activists and 20 percent of the netroots had family incomes greater than $100,000. Just 10 percent of Democrats and Americans were that well off. The possible bias in the Dean survey can’t explain such a large disparity.
Protestants made up four in ten white Americans in 2004, split evenly between evangelicals and non-evangelicals. They accounted for three in ten white Democrats but just one in five white members of the Dean activists and the netroots. Barely any were evangelical. In fact, four in ten white members of the Dean activists and the netroots were secular – four times the incidence among white Democrats or Americans as a whole. While four in ten Americans and Democrats attended church at least weekly, just 15 percent of Dean activists and of the netroots did so. One in three Dean activists never attended church, which was also true of one in four members of the netroots but just one in ten Democrats or Americans.
And finally, as a teaser for Part Deux of this profile, a couple of findings on political characteristics. In Pew’s studies, 38 percent of Americans identified as conservative, compared with 20 percent who identified as liberals. Even among Democrats, one in four identified as conservative – nearly as many as said they were liberal (30 percent). On the other hand 80 percent of Dean activists and 90 percent of netroots members called themselves liberal.
Contrast this result with the fact that just 69 percent of Dean activists and 77 percent of the netroots identified as Democrats. The implication is that if these findings really do proxy the netroots community, then the movement is really about ideology rather than partisanship, recent claims notwithstanding.
And that’s about as tantalizing a teaser as I can come up with this early in the morning. More later this week on the political preferences and attitudes of the netroots.
by Scott Winship