Let’s admit it. The news media are poll-obsessed, especially with politics. But don’t blame pollsters if journalists over-work surveys or neglect necessary caveats.
Take Michigan’s presidential vote. A November 3 poll for Detroit’s Fox2 put Hillary Clinton over Donald Trump by 46 to 41 percent.
Did news reports mention the “margin of error” of plus or minus 3.1 percentage points? With that factored in, Trump could actually have been slightly ahead, and in fact he won Michigan by three-tenths of a percent.
In “Inventing American Religion: Polls, Surveys, and the Tenuous Quest for a Nation’s Faith,” sociologist Robert Wuthnow doubted any polls are representative nowadays since “response rates” among randomly selected respondents are so low. GOP pollster John McLaughlin pursues another complaint, that sampling techniques consistently undercount Republicans.
So religion scribes must be wary. Nonetheless, blest be the Gallup Poll, especially for trend-tracking because it has posed the same questions across years and decades, e.g. the famous “did you, yourself, happen to attend” worship this week?
Sure, fibs and faulty memories may inflate the results, but the downward trend line is noteworthy.
Gallup’s annual “Values and Beliefs” poll in May finds the most permissive U.S. views to date on 10 of 19 moral issues -- though adultery still gets mere 9 percent acceptance. That got more coverage than Gallup’s subsequent report on the poll’s responses about the Bible that presumably shape moral opinions. (Note the sampling error of plus or minus 4 points, with no “response rate” in the fine print.)