Strengthening Constitutional Self-Government

No Left Turns

A Note on Polls

Attempts to measure intense participation by blacks and college youth for Obama drive the polls’ disparity. Some issues: Does traditional polling overlook the fact that many younger folk have only cellphones, which are ordinarily not called? Doesn’t the definition of a "likely voter" undercount the youth vote? But what about the fact that younger voter turnout is never as high as predicted? On top of all this is a political correctness before pollsters, which is then itself corrected in the voting booth.


More important, keep in mind some political history, especially when you see a poll showing Obama above 50% nationally: no Democrat has won a majority of the popular vote since LBJ in 1964 and FDR in his triumphs. Before that there is quite a long stretch of history--back to Franklin Pierce. Dems have typically been a sectional party, not a national majority party, neither by popular vote nor by electoral vote. BHO is no LBJ or FDR.

Discussions - 6 Comments

Interesting points. That is why I actually prefer Gallup. With Gallup as a baseline you can adjust in the direction you think is most likely. When you look at a Rassmusen poll doing so is dangerous because in essence all the brightest economist/actuary number crunchers have already implemented a bias filter to derive the likely voters from the registered voters. So Rassmusen is the best current guess all things considered. The question is can Rassmusen really answer all the questions you ask, in short the existance of the category "likely voters" is an answer in the affirmative.

To answer your question I would compare Gallup to Rasmussen. Gallup using only registered voters has McCain and Obama tied at 46%, while Rasmussen has it at 49% to 46%. So Rasmussen is essentially saying that among those most likely to vote there is a three percentage boost for Obama. Rasmussen may be saying that black and youth turnout will be higher, but we don't really know this unless we know exactly what weights they give to what in differentiating between registered and likely voters.

Well I actually like Rasmussen Markets the most.

John, thanks. In the past, "likely voters" has been more accurate than "registered voters". The polls generally have a slight bias against Republicans, who tend to be more private about expressing their political views to strangers on a phone. I assume the pollsters try to correct these things, but of course they may wind up overcorrecting....

Glad to see that others have noticed how hard it is for Dems to win presidential majorities. Only two of them have done it since the GOP has existed, as Ken points out. On the other side, Republicans have won majority votes many times. So the issue tends to be whether there is a significant 3rd party alternative, because they mostly siphon votes from Republicans (maybe Nader in 2000 was a small but important exception).

The rule is best stated this way: Americans normally vote to keep Democrats away from the White House. When they have a 3rd party alternative to the GOP, the majority often divides, allowing the Dem to win a plurality.

The question is whether this year we are facing a repeat of 1932, the way the Dems are trying to play it. So much for Obama's "postmodern" politics, in which he keeps dredging up variations on PWA and other New Deal oldies.

Excuse me, but didn't Jimmy Carter win a (slim) majority of the popular vote in 1976?

Also, bear in mind the ancient wisdom: "Past results are not a guarantee of future performance." If the American past was a reliable guide to what's happening in our country today, a person like Barack Obama would not be one of the two final contenders for the presidency, and the one more likely to win.

On re-reading my previous comment (#5), I think I should clarify something. In pointing out that the possibility (and, unfortunately, likelihood) of an Obama presidency shows how much the country has changed, I was not referring to Obama's racial identity. Of course, that a person not considered "white" can be a serious presidential candidate is a change, too (and a very good one), but that was not the change I was referring to.

Leave a Comment

* denotes a required field
 

No TrackBacks
TrackBack URL: http://nlt.ashbrook.org/movabletype/mt-tb.cgi/8142