It is interesting to learn that the Ed. department certifies accreditation agencies. That does seem like a way to exercise some influence. But even taking that into consideration, it doesn’t seem like that much influence. To put this in perspective, I would say the federal government has ten times as much influence on curriculum and pedagogy in secondary education as they do in higher education. Yet even in secondary education, the feds influence is far less than the individual states influence. (Note, I think this is for the best)

Perhaps Spellings would have liked “total control of the schools and the curriculum,” leaders, both good and bad, almost always want more power. But framing this as Spellings “one goal” is as bad as when Sean Hannity calls Obama a socialist. The secretary of education will never have that kind of power. The policy proposal under discussion is whether institutions which get a lot of federal money should be required to report some basic data on the students they are educating (info on what data here: http://www.universityofcalifornia.edu/senate/news/source/nces.pdf ) I think that students would benefit from having more detailed and accurate information on the graduation rate, financial aid, and the popularity of different majors. What exactly are the consequences people fear from making this data available?

The type of research on anthro phd education you recommend sounds great and would complement the data collection being discussed.

It is no surprise to me that universities oppose this proposal. The less information that is available, the less competition they face. The less information that is available, the less they have to worry about critics.

I’m open to being convinced that some particular type of data should not be reported, but I think we should embrace the basic idea that more information should be made public.

The first part of this goes to the imposition effort by Secretary Spellings. That debate is well documented and her attempts to impost a unit record database are easily found. She did not have the authority to impose it and the Congress made her back off. However, she discovered that all accrediting agencies (it appears that accreditation is itself voluntary, sometimes involving a choice of which accreditor you want and that agencies have been very loosely watched). In any case, there are a large number but I found out that the one authority the Secretary has is to officialize or not such agencies. There was a huge stir about this and the agencies all rushed to start using her accountancy models as part of their accreditation proces, claiming it was a voluntary action. It is now already taken for granted. Given that, is it voluntary or not?

A good way to get the flavor of this is to go to Inside Higher Education and then search on “accreditation”. You will find 27 pages of articles on the subject in reverse chronological. Skimming through, you can see the contours of this process.

The second issue is whether I believe in large-scale comparisons and databases. My answer is yes and know. Do I think there is a method for deciding if you learn more and better anthropology at Duke or Brown, then I say there is no method that compares apples and oranges. See Robert Hunt´s book on comparison for an anthropologist´s careful view of this.

Do I think that knowing the relationship between what we think we are doing when we teach and what the students find themselves able to do with it throughout their life course? Yes. But even this is a qualified yes.

I am an action researcher and thus I really only accept the merit and value of formative evaluation rather than summative evaluation. Formative evaluation involves evaluating what you are doing in order to improve what you are doing by consulting all parties and by examining the long term effects of what has been done. Maresi Nerad´s work is the closest Iknow to this and even that still needs lots of work. But simpleminded statements about outputs are at least as meaningless as the pathetic ranking systems being used.

So I come back to evaluation for what? Spellings had only one goal: total control of the schools and the curriculum. Many administrators on campuses are happy to adopt this goal and use the justification for slashing subjects they don´t like and building those they do like. If, on the other hand, the goal is to compare the fates of anthropology Ph.D. from 200 schools 5, 10, 15, and 20 years and to talk to their employers too, then let´s get at it. My impression is much of what is taught is utterly useless and much of what is needed is not taught. I would be happy to have both quantitative and qualitative data on this. However, this is not what most people are volunteering for. They are trying to co-opt the evaluation process in order both to keep doing what they have been doing and to put the screws to people they don´t like.

Is this clearer?

Davydd

Can you give any help in looking for more detail about this, at least at the accreditation agency level? And about the local institutional cases, I would love to hear some details or a sketch of a case study, if you feel inclined… Is this happening specifically at the Ivies, for example? At Cornell? Or perhaps the details will just have to await some future research project…

In thinking again about your paper, I gather that you are against the apparent Spellings fantasy of a world where policymakers or student-consumers will judge all institutions according to some fairly simplistic sets of outcome ratings — but I assume you are not actually against large-scale data collection, but rather are more concerned about simplistic or even punitive means of using such data in top-down policymaking? This would suggest a further question: how do your views on national/quantitative data collection relate to your views on the appropriate use of this data? Is higher ed data collection necessarily going to be tightly coupled to the use of that data in governance regimes?

I agree. Currently, the most important variables are whatever U.S. News and World Report decides to measure. Some of what they measure may have some value, but we could do far better if colleges were required to share more information.

I take it the voluntary controls are “statistical, commodity production models” but I still don’t understand what exactly they are measuring, and whether/how this significantly impedes other goals.

My point about voluntary controls is based on a variety of sources. It is relatively easy to document the scramble of at least feighned compliance with this mode of operation in the accrediation agencies. They stumbled all over each other to implement statistical, commodity production models and are still at it. This cascades directly down to institutional officials who are now imposing this kind of accountability on their campuses. Many institutional research offices at big institutions are doing this and many small colleges either have new VP´s for this or have hired consultants. The faculty then are tasked with coming up with outcome models. Is this voluntary? I think the voluntary-involuntary distinction in a deeply ideological universe with lots of money at stake is probably not good enough.

Regarding neo-liberalism, I actually don´t like the term either. However, it certainly is not conservatism or liberalism. What I don´t like about the term is that there is nothing “neo” about pseudo-market corporatism in politics. This is the old time religion that has led us into one cycle of socio-economic disasters after another.

I have nothing against measurement and evaluation. It is true that universities have been shamelessly careless about seeing whether what they do and teach is good for anything and adjusting their behavior in response. However, stupid variables stupidly measured without any thought for the nature of what education actually might be will produce stupid policies and stupid behaviors. Garbage in, garbage out.

Finally, I have found more of value on higher education in the writings of serious conservative economists like William MacMahon and Ronald Ehrenberg than in the diatribes against corporatization that abount.

From a scientific point of view, the ideal would be for the federal government to randomly assign some policies to different states and we could see what effects they had. More realistically, we look at historical data of states/countries (ideally a lot of them) change policies and do our best to evaluate them. Of course, the outcomes we choose to measure are important, every policy is destined to help some groups/goals and hurt other groups/goals. But with sufficient sharing of evidence, I think we could get much closer to agreement on the desirability of various policies.