This article is from the archive of our partner . is heralding a report from Gallup that shows a big uptick in the unemployment rate, from 7.7 to 8.9 percent month-over-month. This is Gallup's role in American economics, it seems: to vindicate or refute the government's official employment numbers.

First, the data. Unlike the Bureau of Labor Statistics, which compiles the official data for the Department of Labor, Gallup's figures are a 30-day rolling average of what it finds in its survey. In other words, the most recent figure of 8.9 percent is the average of what it has seen over the past month. Nor is the data seasonally adjusted, as the BLS' numbers are. So here's how the two compare since January 2010. (We used the figure from the 20th of each month to represent Gallup's figure for that month.)

In short, the Gallup data is far more variable, but the trend is the same. The spike at the end of the Gallup chart could certainly signal that the BLS number is about to rise, too — or it could be one of the other numerous times in which the Gallup data has gone much higher or much lower than the government figure.

Which is the point. The variability of that Gallup data is regular fodder for those who want to find fault with the BLS figures. Last February, conservatives made the same argument, pointing to the spike that month as evidence that the economy was doing worse than expected. "Do you believe the Bureau of Lies and Shams and their boss – Barack Hussein Obama," one site asked "or your own eyes and the data from an independent source?" By June, the Gallup figure was back in line with the BLS.

And then in October of last year, the Gallup number dropped way below the BLS number. This coincided with Jack Welch's infamous suggestion that the BLS was massaging its figures to help the president's reelection, so Salon noted the Gallup plunge as a point in the government's favor. That site was wise enough to qualify its point, calling the Gallup number "eyebrow-raising."

That skepticism is still warranted. Each of the last three Augusts, Gallup has shown increases in unemployment, though not to the extent seen this year. Gallup goes up and down far more than the BLS figure does — and is based on a different set of data. Citer beware.

That Salon piece came out about a month before 2012 election. Obama won reelection, in part due to perceptions of how he handled the economy. Gallup's polling was widely criticized for getting the results wrong.

This article is from the archive of our partner The Wire.

We want to hear what you think about this article. Submit a letter to the editor or write to