Roger Simon, the Chief Political Columnist for Politico:

I have never been called by a political pollster and don’t know anybody who has, but I know some pollsters, who assure me they don’t make the numbers up, and I believe them.

From George Gallup’s mock Q-and-A in A Guide to Public Opinion Polls, which he wrote in 1948:

“Why haven’t I been interviewed?  Why have I never heard of anyone who has been interviewed?”

These questions come up frequently in connection with modern public opinion surveys, because people do not understand how it is possible to get an accurate measurement of public opinion when only a small part of the total population is interviewed…In this respect, modern surveys merely apply to public opinion research certain well-established procedures which have been used for years in the fields of engineering, medicine, education, and all the social sciences.  When an engineer wishes to judge the quality of ore in a mine, he examines a few “samples.”  From these samples he makes a highly accurate estimate of the amount and quality of ore in the mine.

Simon again:

We are a nation of nearly 313 million people. So how many people did the pollsters actually speak to? If you have extremely good eyes, you can find the answer in tiny type at the bottom of a chart: The Post-ABC poll was conducted by phone “among a random sample of 1,005 adults.”  That represents 0.0003 percent of the nation at large.


“How many persons have to be included in a poll to obtain reliable results?”

Size and accuracy are inextricably linked in the minds of most laymen.  Invariably the first question that is asked by persons who examine the results of a public opinion survey is: “How many persons were included?”  Actually, the size of the sample (the number of persons interviewed) is far less important as a factor in achieving reliable results in modern polling than several other factors, among the most important of which are the accuracy with which the persons chosen to be interviewed mirror the total group, the wording of the question or questions used to develop the information and the accuracy and lack of bias or influence in the interviewing procedure itself.

Apparently, we haven’t come that far in the 60+ years since Gallup wrote his book.  Again, Simon is the Chief Political Columnist for Politico.

[Cross-posted at The Reality-Based Community]

John Sides

John Sides is an associate professor of political science at George Washington University.