Pollster.com

Articles and Analysis

 

USAToday's "Gallup Guru" Blog by Frank Newport

Topics: Pollsters

One recent development I overlooked in the run-up to the election is the new blog on USAToday.com by Frank Newport, editor and chief of the Gallup Poll.  Dubbed "Gallup Guru," Newport's blog promises to chew over many of the same topics we examine here. 

In an item item last week, for example, Newport reacted to the apparent pre-election belief by White House political advisor Karl Rove that "polls were obsolete because they relied on home telephones in an age of do-not-call lists and cell phones:"

Karl Rove mentions cell phones.  The impact of cell phones on survey research has been a topic of extraordinary analysis by survey researchers over the last several years.  The American Association of Public Opinion Research, as a matter of fact, is devoting a special track of research sessions on the impact of cell phones at its annual conference next May (I am the associate chair of that conference). 

Indeed, it might have been useful if Rove had read a research paper published by Dr. Scott Keeter of the Pew Research Center just two weeks before the election this year entitled:  "Cell-Only Voters Not Very Different".  Keeter concluded his analysis by noting that "… the absence of the "cell-only" population from telephone surveys is not creating a measurable bias in the overall findings."

Frank Newport contributed the debut item to our Guest Pollster corner and responded to comments left by readers.  As such, the comments section in his new blog may offer readers more routine interaction with this Gallup Guru.  One more thing we all need to read regularly. 

 

Comments
David T:

Why do we still go through this nonsense of asking why Rove believed the GOP was going to win? We have absolutely no evidence that he so believed--except that he *told* people he did. This is no evidence at all, unless one can imagine Rove openly expressing doubts--which was unthinkable, because it would have given a psychological boost to Democrats and would have depressed morale among Republicans.

____________________

Chris G:

I read the Pew paper and am suprised by the title, very misleading. If you actually look at the data and read through the analysis, you see that cell-only VAP is, in fact, different. Much younger, of course--but with youth comes the other typical covariates, such as more likely to be independent, not nearly as registered, more liberal, more unmarried. Also, the May report from Pew indicates that even controlling for age there may be substantive demographics differences.

The only reason why these differences do not numerically influence polls, according to all of these data, is because *currently* the cell-only population is a small portion of the VAP, about 10%. But in a close election and an area that may have an unusually high proportion of cell-only users (and registration drives), this can have an unpredictable influence.

There are many other systematic sources of bias, of course, but this is just another example of how MOE is understated. The typical computation of MOE assumes "random" (normal) error, an assumption which clearly does not hold.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR