Channel One & the election
10.21.2004Polls — while usually pretty accurate — don't always mean that much. At least, that's what many people think. But a fairly good proxy for the upcoming election is a recent Channel One presidential poll taking by high school & middle school youths across the country. If you don't know, Channel One is a nation-wide cable network available in a growing number of public schools across the country.
What makes this particular poll credible, is that votes very closely followed what we know about individual state polls. The only surprises were a few close (or "battleground") states that might still go to either candidate. And the 1.4 million students who voted crushed the margin-of-error question, since most polls usually target about 1,200 people.
Why do kids' votes instruct us about the "real" election, two weeks away? Because political opinions are still mostly influenced by parents. Actually, the two variables political scientists have discovered that most accurately predict someone's vote in any election are their mother's & father's party ID. Lending strength to the theory that we are "socialized" into our political opinions primarily through our parents' influence. Thus, the way school adolescents vote in a nation-wide poll likely reflect the way their parents plan on voting.
BTW, Channel One's student voters chose Bush over Gore in 2000. I can't find an online reference to the 1992 or 1996 Channel One poll, but they both went to Clinton (I voted in the 1992 one as a high school junior).
-----
UPDATE 1: Here's a 1999 APSA (American Political Science Association) conference paper by Kent Jennings, Laura Stoker, and Jake Bowers on the issue of parental transmission of political attitudes to their kids. They find that the reports from the original Jennings & Niemi paper (1968, APSR) — that parents & their children hold very similar political views — still holds, even for Gen Xers.
UPDATE 2: Adding drama to the Channel One election, Nickelodeon's poll of 400,000 children gave the election to Kerry (57/43), but didn't break down results by state (so no electoral college prediction). Like Channel One, Nickelodeon has correctly predicted the presidential winner every election since they've started polling kids, in 1988.
NOTE: It's interesting to note the difference in media coverage between the Nickelodeon & Channel One polls.
Posted by Miguel at 02:44 PM
Comments
What about teenage rebels?
Couldn't a case be made for the elections having the opposite result of student polls because at this age students are likely to revolt against their parents' authority?
Posted by: Marco at October 21, 2004 06:27 PM
Interesting point about media coverage.
You know what I 'm wondering, though?
I recently read about a problem in the 2000 election that when first newsstations called Florida for Gore in the early evening that may have discouraged potential Bush voters in the part of Florida that is in Central Time. The stations called the state after voting had ended in most parts of the state but not everywhere.
The point being made was that Bush might have received more votes in Florida if the stations had not prematurely called Florida for Gore.
Along the same line of reasoning one could argue that if Nickelodeon's poll receives great media coverage this could discourage Bush voters.
However, the opposite might be the case.
People registered to vote can be lazy enough not to vote, especially if they think that their candidate is undisputed. Calling an election early for one candidate could result in people celebrating his victory rather than going to the ballot box.
Now that I mention it I am reminded of a scene in the 2003 movie Head of State.
****Spoiler****
In it conservative voters rush to the polls when the election is prematurely called for Chris Rock.
****End of Spoiler****
So my point is, it may not be wise to cheer for my candidate as if he had won if he still hasn't.
Posted by: Marco at October 22, 2004 04:46 AM