Friday, November 2, 2012

I don't like your research results so let's burn your report

Typically, empirical academic research proceeds like a conversation.  Someone finds a result.  Someone else questions the methodology, data, or theory and then redoes the analysis taking into account the issues that he or she raised.  Then someone else does it again.  Eventually you have a body of work that can be examined on a "meta" level and some conclusion can be drawn. 

The Congressional Research Service does private research for people in Congress.  According to the NYT, "The Congressional Research Service has withdrawn an economic report that found no correlation between top tax rates and economic growth, a central tenet of conservative economic theory, after Senate Republicans raised concerns about the paper’s findings and wording."  (see link here)

Republicans did not say whether they had asked the research service, a nonpartisan arm of the Library of Congress, to take the report out of circulation, but they were clear that they protested its tone and findings.Don Stewart, a spokesman for the Senate Republican leader, Mitch McConnell of Kentucky, said Mr. McConnell and other senators “raised concerns about the methodology and other flaws.” Mr. Stewart added that people outside of Congress had also criticized the study and that officials at the research service “decided, on their own, to pull the study pending further review.”

You can look the report by clicking the link.  It is not partisan and is very data driven.  The Times article ends with a weird paragraph about the author of the report.  Here it is:


Mr. Hungerford, a specialist in public finance who earned his economics doctorate from the University of Michigan, has contributed at least $5,000 this election cycle to a combination of Mr. Obama’s campaign, the Democratic National Committee, the Democratic Senatorial Campaign Committee and the Democratic Congressional Campaign Committee.

This paragraph seems to imply (at least to me) that an empirical researcher can be overly biased by his or her political leanings and that somehow giving to political campaigns is a subversive activity.  Again, normal research methodology would be for this work to be countered by other work.  The work would not be removed from circulation. 

2 comments:

  1. With the lack of clarity in the Romney tax plan and this particular report coming out and saying why it is impossible for his plan to reduce the deficit, it is easy to see why the Republican party was fighting to remove these findings. To me, however, it seems as if like any economic report there are biases. As noted in the final paragraph, it seems as if the creator of this report possibly had his own desires and feelings involved in the research. Based on this article and many others alike I simply find that it is nearly impossible to trust any economic report that is being released whether they claim nonpartisan or not.

    ReplyDelete
  2. In general, I agree with the author, Republicans are definitely being shady with the way they are handling this report. I think Dr. McKinney makes a good point in bringing up the last paragraph of this article because this bias is a real concern. You can only find the facts you choose to look for, so by choosing to examine whether the Bush tax cuts helped the economy, the author of the report allowed himself the opportunity to find contrasting evidence. On the other hand, this is a totally valid question to ask, and essentially is just seeking to validate the claims of Republicans. If Mr. Hungerford had found the opposite in his study he would still presumably would have released his findings, even though they contrasted with his political views. To me it seems more likely that Mr. Hungerford is a Democrat because of his research, rather than a researcher because he is a Democrat. As a person who hopes to one day do a job similar to Mr. Hungerford's, where policy analysis and recommendations ultimately end up in the realm of politics, I would hope that the field is based more on empirical research than on confirmation bias.

    ReplyDelete