Followup Note on the NRC Report

3/30/2001

In my article which I posted to this e-mail list yesterday, I spent some time on the severe--and obviously deliberate--bias of the makeup of the National Research Council committee: (a) People from industry, including Intel and Microsoft; (b) people from academia, who have huge incentives to toe the industry line, and (c) a couple of pro-labor people. Category (c) was small, and was overwhelmed by the others. I'm sorry to put it this way, but it was a setup from Day One. It was no wonder that the report was phenomenally biased.

(In some senses, the presence of category (b) is the most treacherous. People assume that academics will be fair, and people are unaware of the major incentives academics have to side with industry on these issues. I described some of those incentives in my article posted yesterday, and go into much more detail in Sec. 2.2 of my "Debunking the Myth of a Desperate Software Labor Shortage" paper.)

In a followup to this, I'd like to note that I too have been appointed to an NRC committee, the Panel on the Research on Future Census Methods, 1999. Very early in the process, I saw that I would not have time to "do my homework" on the committee, and resigned.

(In my own presentation to the NRC IT Workforce Committee, I mentioned this point, and chided the committee for what I considered lack of diligence in their work so far. Afterward, one "nonaligned" member of the committee came up to me and agreed that such members didn't have time to devote to the project--in great contrast to categories (a) and (b) above, which could draw upon the immense resources of the industry lobbyists--but wasn't it better for the (c) people to be on the committee and provide some balance than to not be on the committee at all? I said no, and even today I really cannot see reason to answer otherwise.)

At any rate, I still have all the materials sent to me by the NRC for participation on that census committee. In looking at them now, it is striking to see the number of points made about keeping the committee's membership and work impartial. I don't know whether there is a sincere desire in the NRC for impartiality, but if there is, someone dropped the ball in the IT Workforce Project, egregiously so.

One of the documents given to me for the census project was a 4-page form entitled Potential Sources of Bias and Conflicts of Interest, accompanied by a 5-page, small-font policy statement, The National Research Council Policy on Disclosure of Personal Involvements and Other Matters Potentially Affecting Committee Service. Here is a sample excerpt from the policy statement:

...if a report is to be not only sound but also effective as measured by its acceptance in quarters where it should be influential, the report must be, and must be perceived to be, (1) free of any significant conflict of interest, and (2) not compromised by bias, and (3) untainted by allegations of scientific misconduct.

The statement then goes into several pages of exceedingly detailed descriptions of various types of conflicts of interest. The handbook given to me, "Getting to Know the Committee Process," also refers repeatedly to the need to avoid bias. It mentions, for example:

...committee members discuss this information [from the members' conflicts-of-interest forms] in closed session at the beginning of their first meeting and annually thereafter. This information is also reviewed by officials of the institution [the NRC], and if a potential conflict becomes apparent--which is rare--the committee member may be asked to resign. In exceptional circumstances, an individual may continue to server on the comm if the conflict of interest is promptly and publicly disclosed, and the Academy has determined that the conflict is unavoidable. To fulfill our legal requirement for such public disclosure, the Academy posts on its Web site a brief statement describing the unavoidable conflict. When a question of balance arises, the usual procedure is to add members to the committee to acheive the appropriate balance.

Beautiful words, set in a handsomely-printed pamphlet--but bearing no relation to the ugly reality.