deadlines amuse me

exploring how the world works and why it works that way …

Presidential preference polls: Much ado about nothing 17 months early

leave a comment »

Egads! News flash from pollster Gallup Inc.:

PRINCETON, NJ — Mitt Romney (17%) and Sarah Palin (15%) now lead a smaller field of potential Republican presidential candidates in rank-and-file Republicans’ preferences for the party’s 2012 nominee. Ron Paul, Newt Gingrich, and Herman Cain essentially tie for third, with Cain registering 8% support in his initial inclusion in Gallup “trial heat” polling. Notably, 22% of Republicans do not have a preference at this point. [emphasis added]

Yawn. This poll conducted May 20-24 with a random sample of 971 Republicans and Republican-leaning independents tells me nothing I want to know or need to know. I’m not necessarily picking on pollster Gallup; my objections apply to most of these almost weekly presidential preference polls. They mislead and misrepresent more than enlighten. In sum, they represent manufactured noise with little signal.

These “national” preference polls carry the assumption that a national electorate elects the president of the United States. That’s misleading. The vote in each state provides for electors to the Electoral College. That Mitt Romney “leads” a national preference poll means nothing because we’re not told how Mitt fares state by state. But one national poll costs far less than 50 individual state polls. And how many journalism entities are willing to spring for the latter? [One organization, Public Policy Polling, has done several match-up polls, Obama vs. GOP X, in several key states.)

And the “now lead” wording in Gallup’s press release? It’s horse-race language. The press doesn’t like to pay for polls that do not carry “news,” as misleading as such faux news might be.

Gallup’s language is not necessarily true. The margin of error in this poll is plus or minus four percentage points. Gallup even calls my pal Mitt the “clear front-runner”: Romney would be the clear front-runner, but arguably the weakest front-runner in any recent Republican nomination campaign. But he isn’t clearly ahead. In this poll, his support could be as low as 13 percent and that of fiscal conservative Ron Paul could be as high as 14 percent. Even Herman Cain (who’s he?) might be only a percentage point behind Mitt.

Gallup, to its credit (most national polls don’t do this) reported the rate of non-preference: Notably, 22% of Republicans do not have a preference at this point. Frankly, I find it notable that 78 percent of respondents do have a preference 17 months before the election. I wonder: Did Gallup attempt to quantify the depth of conviction with which respondents hold their preferences? Or did respondents throw out a name 1) based solely on “brand recognition” or 2) any name just to get the pollster off the damn phone?

There are sites aplenty that aggregate and proffer preference polls. Some generate a “poll of polls.” One such “average” of many polls is at (whose website and Facebook pages give no indication of who or what funds this site). Such a compilation of the mean of many polls is nonsense. Yet people (and journalists) might use this average of polls to argue political choice and/or public policy. It may also sway the number and size of campaign contributions, particularly of the corporate kind, available to candidates. Such an average is worthless. How does it account for each poll’s differing size of sample, question wording, margin of error, and date conducted? Sheesh.

Take the presidential preference polls or aggregates thereof presented by CNN, The New York Times, The Washington Post, and other journalistic or quasi-journalistic entities with a heaping mound of grains of salt.


Written by Dr. Denny Wilkins

June 2, 2011 at 2:41 pm

Posted in Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: