How many of you remember the Weekly Reader from your childhood school days?
Well........Weekly Reader has been doing something really cool for 56 (counting the upcoming election cycle) years. Holding a student Presidential Election Poll.
Over the past 52 years, students voting in the Weekly Reader Presidential
Election poll have predicted the winner of the election 12 out of 13 times.
And a very important statistical sidenote - the only time the Weekly Reader Presidential Poll was incorrect was 1992 when the kids picked Bush Sr. instead of Clinton. But.......keep in mind Perot was strongly in the mix to confuse the issue and make it a very atypical election.
Here's more good stuff from the Weekly Reader:
The poll's been right more than 90 percent of the time. The
eligible candidate that kids named as their choice for president has won
12 out of 13 times -- and in eight out of the last nine elections.
Weekly Reader voters often have been on target about the Electoral College. In 1972, the poll predicted that Richard Nixon would win 49 states, and that George McGovern would take only Massachusetts and Washington, D.C -- mirroring the actual results precisely.
In 1980, kids predicted the huge sweep by Ronald Reagan.
The Weekly Reader survey has, on occasion, been more precise than
major pollsters. For example, most polling organizations said the race
between President Jimmy Carter and Ronald Reagan was too close to call. But Weekly Reader kids voted for Reagan by a large margin -- exactly what happened in the general election. In fact, the kids' results exactly matched the fraction of votes (41 percent) that Carter received.
--
This student survey has been widely acknowledged as a key indicator of who will win. In 1984, U.S. News and World Report asked, "Where do experts differ on the contest for president? Which will come closest?" It put together a chart that listed eight top pollsters, including Gallup and Harris, and listed Weekly Reader prominently.
Here's the link:
Weekly Reader Kids Vote
Here's a link to a USA Today story about the Weekly Reader Poll prior to the 2004 election:
USA Today Story
So - why do I bring this up? It just so happens I'm "in charge" of a fairly large group of kids. I've held mock elections with my group in 2000 and 2004. Before I reveal the results - here are some general (and purposely vague to protect the innocent!) demographics:
Location - Western North Carolina, suburban.
Age Group - 14-17.
Economics - Middle-class to upper middle-class, lilly white.
Poll Time - Late September, early October, approximately 1 month before General Election.
2000 Results - Our 2000 mock election was an admittedly seat-of-the-pants operation. A simple show of hands. Approximately 25 votes. Much to my dismay, Bush won. Don't have (or remember) the exact numbers.
2004 Results -By 2004, I had improved my methods somewhat. We used real ballots, and I tracked, and published (for the kids) the results. 32 votes total. 19 for Bush (59%). 13 for Kerry (41%). The actual North Carolina numbers would be Bush (56%) / Kerry (44%). Our sample size was obviously small, but given our demographics, a Repulican lean above the actual result is not surprising.
2008 Results - Methods continue to get better. And my group is steadily growing with a sample size of 54. The results are as follows:
Sample Size - 54 (25 boys, 29 girls)
Obama - 29 votes (54%)
McCain - 25 votes (46%)
The boys favored Obama 15-12, the girls favored Obama 14-13
I also asked for each kid to list the single most important issue to them. The results are as follows:
Economy - 20
Foreign Relations - 13
Oil Dependency - 7
Environment - 5
Health Care - 2
Other (Abortion, Religion, World Population) - 3
No response - 4
So does our mock election agree with the Weekly Reader to date? Straight from the Weekly Reader web site:
Who will win this year? The choice between Republican John McCain and Democrat Barack Obama is a historic one. Polls so far have shown the contest to be very tight. Want to know the verdict of Weekly Reader's voters? Check back here on October 29, when the 2008 survey results will be posted! Then compare those results with the tally on November 4. Did the Weekly Reader magic work again?
This year's results from the Weekly Reader Poll will be released October 29th.