It is important to question sources of information. Who is the author of this? What is the methodology behind this study? But how can we distinguish between legitimate criticism of, say, data collection, and biased favoritism?
Take the case of the recent release of new unemployment numbers, which place the most recent preliminary estimate of the unemployment rate at 7.8%. The data is compiled by the Bureau of Labor Statistics (BLS), using surveys of both households and businesses (here is the official release, and it includes links to FAQs detailing the methods of data collection from these surveys).
From a political standpoint, BLS figures are rarely, if ever, questioned on a basis of partisan integrity. The data are public enough that any misrepresentations can be spotted by anyone with an internet connection. Which leads to the reaction to the release of the unemployment numbers, as described in the first link above.
As it turns out, many conservatives question the legitimacy of the data as a suspiciously-timed spark of positive news for the President. In questioning the integrity of the data, they would like to suggest that these BLS estimates are unreliable.
So, is it a matter of serious methodological criticism that they bring up what they see are flaws in the estimates? Or is it a political jab in an attempt to discredit data which help the opposing candidate?
Well, if the individuals questioning the data are so certain of BLS biases, why only question this report and these numbers? If they had a serious concern about the accuracy of BLS reporting, or the ability of the BLS to perform its duty, I would think they would be willing to voice this concern immediately. A closer look suggests a different story. In fact, as employment numbers were poor and damaging to the President, these individuals stayed silent – and in many cases, even supported the same BLS figures as concrete evidence of the President’s poor performance.
Then, why claim the figures couldn’t possible be accurate now? It isn’t a question of methodology, but of ideology. Unfortunately, you can’t selectively agree with the data when its conclusion supports your own, then bash that same data when its conclusion is working against you. This is disingenuous.