Crosstab conjecture - Part 2 - Ahhh-loozah

Updated
Crosstab conjecture - Part 2 - Ahhh-loozah
Crosstab conjecture - Part 2 - Ahhh-loozah

At the end of 2008, as President Bush was on his way out, bloggers on the left made some hay of an incidental poll finding that fewer people would admit to voting for Bush than the vote total shows actually did.

Since then I’ve seen at least one poll make a similar case about President Obama during a patch of unpopularity. (Ok, that was just a Newsmax poll, but still.) And actually, as far as demonstrating the phenomenon, people also lied (misreported?) to take credit for having voted for Obama (or to deny voting for McCain/Palin [and also about having voted at all]).

The Slate piece linked in the previous sentence offers a lot of possible explanations for what’s going on, but since this is a blog post with “conjecture” in the headline and we’ve already acknowledged that amateur beard stroking over a single piece of a single poll won’t be putting any rockets into orbit*, I’m going to go out on a limb and postulate that Americans prefer to be associated with winners and distance themselves from losers (with allowance for cheering for underdogs).

And so while I was poking through the PPP poll about Ron Paul’s Iowa surge* I was interested to note how Iowans reported their 2008 voting.

Just for the record, here are the actual 2008 Iowa caucus results:

Mike Huckabee       40,841   34.4%
Mitt Romney           29,949   25.2
Fred D. Thompson 15,904   13.4
John McCain          15,559   13.1    
Ron Paul                11,817  10.0  
Rudolph Giuliani      4,097   3.5  
Duncan Hunter          524    0.4 
Tom Tancredo              5     0.0 

Predictably, a greater percentage claim to have voted for the eventual winner, John McCain, than actually voted for him. In the “memory election,” John McCain came in second instead of a distant fourth. But what’s odd is that McCain’s memory votes appear to have come from Mitt Romney, not from Mike Huckabee.

Again we can only speculate about why that seven percent went into denial about voting for Romney last time. And I’m not aware of any evidence about predicting votes based on self-reported past voting. But in a general sense it seems obvious that losing an election would carry some stigma for a candidate.

In a race that seems more like a personality contest than a battle of ideas, with all past scandals and gaffes erased of consequence, it seems at least as significant as any other influencing factor that some Iowa voters look at Romney and think, as my Long Island friends say with right-angled thumb and index finger pressed to their foreheads, “Ahhh-loozah.”
__________________________

*See Part 1

Crosstab conjecture - Part 2 - Ahhh-loozah

Updated