In a series of blog posts, statistician Nate Silver has made a compelling case that Strategic Vision, LLC is a public realtions/polling firm which, put bluntly, makes up data.
In this post, Silver looks at a recent poll conducted by SV, which made some news. The upshot of the poll, commissioned by the Oklahoma Council of Public Affairs, is that Oklahoma public high school students are extraordinarily stupid. The poll showed, for example, that only 23% of public high schools students could name America's first president (George Washington).
I saw that poll reported in the news; I almost blogged about it. But where I was left astonished, Silver smelled a rat. And he went on to make a convincing argument that the numbers appear to be, quite simply, fabricated.
In another post, Silver notes that Strategic Vision appears to make up numbers because in its polling, a statistically high number of the Strategic Vision's raw numbers have trailing digits that end in 5,6,7,8 or 9 (as in 148, 326, 49, etc.) Obviously, when you are dealing with polls, trailing digits of 1, 2, and 3 ought to show up just as often as trailing digits of 7, 8, and 9. It's suggestive, albeit not conclusive, that Strategic Vision makes shit up. "Over a sample of more than 5,000 data points, such an outcome occurring by chance alone would be an incredible fluke—millions to one against," writes Silver, who allows that "some intrinsic, mathematical reason that certain trailing digits are more likely to come up than others" may be an alternative, yet unproven explanation.
But Nate Silver isn't alone on this crusade. In fact, he didn't even start it. The governing industry body for pollsters, the American Association for Public Opinion Research (AAPOR) criticized Strategic Vision LLC for refusing to disclose "essential facts" about surveys it conducted prior to the 2008 New Hampshire and Wisconsin primaries (AAPOR was conducting a study because some of the polls in the 2008 election were wildly off). Strategic Vision is not a member of the AAPOR (nearly all major polling firms are), which alone ought to tell you something.
Strategic Vision, which normally gets retained by Republican/conservative clients to conduct polling, has a history of questionable practices:
Details of Strategic Visions polls have long raised flags among pollsters, in part because it refuses — unlike other pollsters — to release "cross-tabs" — the detailed demographic breakdowns of individual polls. A source noted other anomalies to me today. One is that the pollster always reports having called a round number of respondents — unusual in an industry that typically uses large call centers and winds up — as casual poll readers know — with uneven numbers of calls.
Another question is how the firm pays for its polls. Its website lists at least 172 public polls, and at a stated cost of $30,000 a poll, that's an expenditure of more than $5 million — quite a sum for a small firm.
A third question has to do with the firm's offices. Its website, as recently as last month, listed offices in Atlanta, Madison, Seattle, and Tallahassee — all of which match the locations of UPS stores, rather than actual offices. The addresses are now gone from the site entirely, though it now also lists a Dallas presence.
If the allegations are true, this rises above the usual "lies, damned lies, and statistics". This is outright fraud committed by a polling company on the American people.
Did I mention that Strategic Vision routinely conducts polls for Republican and conservative think-tanks and causes?