Pollster.com

Articles and Analysis

 

Living in "Pollsterland"

Topics: Measurement

In her new "Poll Positions" column, CBS News polling directory Kathy Frankovic discusses the way pollsters love to ask questions that ask, "what if?" Although, as she points out, such questions were a favorite of polling pioneer George Gallup, they come with a potential danger:

[S]ometimes trying out those ideas may put us just outside of the real world and into what might be called Pollsterland.

It's not just that we can ask questions about Iraq, or health care, or immigration, or baseball's Barry Bonds, all of which are aspects of current reality. But we also can ask questions about an invented reality, such as: What would happen if . . .?

She goes on to list several examples, especially the "most famous" what-if of all, the one that begins, "if the election were held today..."

Frankovic also notes that Gallup started asking "as early as 1958...whether Americans would support an African-American candidate for president." My colleagues in the American Association for Public Opinion Research (AAPOR) picked up the discussion on our members-only "listserv," where someone pointed out that Gallup started asking 21 years earlier about whether Americans would support a woman for president.

This observation prompted one AAPOR member to recall the remarkably leading language used the very first time George Gallup asked a "what if" question about a potential woman presidential candidate. I checked the Gallup archives (sub. req.), and here it is along with the results (from a survey conducted in January 1937 among 2,888 adults nationwide):

Would you vote for a woman for President if she qualified in every other respect? [emphasis added]

33% Yes
64% No
3% No opinion

"Pollsterland," indeed.

PS: Thanks to AAPOR member SL for remember the 1937 Gallup question.

 

Comments

A couple observations on the Gallup question:

1. This is not the only case where Gallup gave the short shrift to women. Gallup regularly under-sampled women in his polls -- typically by a ratio of 2 men to 1 woman. Why did he do this? In part, because he though that he was wasting his time talking to women. As he wrote in 1940 "How will [women] vote on election day? Just as exactly as they were told the night before" (an aside for those who are interested in old polls. I talk about Gallup's sampling procedures in an article in the Winter 2006 issue of Public Opinion Quarterly and Sarah Igo has a very interesting new book on polling in the 1930s called "The Averaged American").

2. In February 1937, Gallup asked about support for non-protestant candidates, but in a somewhat gentler way. Specifically, he asked, "Would you vote for a (Catholic/Jew) for President who was well qualified for this position?" I'm not sure how much the question wording made a difference, but 60% said they would vote for a Catholic and 45% said they would vote for a Jew.

____________________



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