« What is so great about social democracy anyway? | Main | Ideas worth pondering »
The productivity performance of universities
This paper presents new evidence on research and teaching productivity in universities using a panel of 102 top U.S. schools during 1981-1999. Faculty employment grows at 0.6 percent per year, compared with growth of 4.9 percent in industrial researchers. Productivity growth per researcher is 1.4-6.7 percent and is higher in private universities. Productivity growth per teacher is 0.8-1.1 percent and is higher in public universities. Growth in research productivity within universities exceeds overall growth, because the research share grows in universities where productivity growth is less. This finding suggests that allocative efficiency of U.S. higher education declined during the late 20th century. R&D stock, endowment, and post-docs increase research productivity in universities, the effect of nonfederal R&D is less, and the returns to research are diminishing. Since the nonfederal R&D share grows and is higher in public schools, this may explain the rising inefficiency. Decreasing returns in research but not teaching suggest that most differences in university size are due to teaching.
Here is the full paper. Here is a non-gated version.
I take "the returns to research are diminishing" to be the fundamental point. The authors also find that private universities are about twice as research productive as public universities, and that private universities have a higher rate of research productivity growth. Public universities have superior teaching productivity.
Posted by Tyler Cowen on November 12, 2006 at 06:52 AM in Education | Permalink
Comments
The point about diminishing returns to research fits with my impression of changes in research over the last 25 and a bit years. So the conclusion does not surprise me. It would be interesting to know if the diminishing returns are concentrated in a few fields. My expectation would be health sciences, one of the more heavily funded areas, would show significant problems.
I do not think the diminishing returns are unavoidable. Based on my impression from health sciences a major factor has been the emergence of a strong feeling that the major questions have been identified and articulated. The resulting belief is that the important research directions have been identified and it is simply a matter of pouring money into these areas to arrive at all the important answers. The concentration of funding into limited areas results in the diminishing returns.
The reasons for the strength of this belief are unclear, but one significant factor may the aging of the research faculty as the baby boom generation ages. Many of the senior positions are dominated by people in their 50's and 60's and this may have engendered a rigidity in outlook. There are other reasons I can think of to do with the enormous amount growth in funding with little oversight and the hype for possible benefits (again largely in health science research) that may have undermined the flexibility and originality of research done.
I suspect this will change with time. Several factors suggest that the first tides of change have started. The growth in quantitative approaches to biology suggests that the older views will marginalized sooner rather than later and the openness that results shou,d allow new fields to emerge. Presumably, this will improve the research in the fields.
Posted by: am at Nov 12, 2006 2:56:42 PM
1.University research work is labor-intensive, perhaps the denominator
of the productivity could be the number of the researchers. This also
would better account for the bias in the inter-university research growth
estimates.
2.Funding bias
a. The endowment could be discounted for the time effects
b. Private funding is discounted; however, in the study of this paper,
to compare the productivity of growth in public and private university,
there is no need to take account for its funding cost.
Posted by: sally at Nov 13, 2006 2:33:48 AM
The paper measures the number of papers and number of citations as
the output. But there is no attempt to show the value of the output.
Thus it is based on the critical assumption that all papers or citations
are equal. This is the same as assuming that the value of a steering wheel
and an engine block are the same in the production of an automobile.
Garbage in garbage out.
Posted by: spencer at Nov 13, 2006 9:58:18 AM
It's cool site please visit our site.
http://www.shopeastwest.com/med
Posted by: tom at May 12, 2008 8:24:55 AM
Last chaos Gold
Buy Last Chaos Gold
Cheap Last Chaos Gold
Posted by: aion kina at Mar 21, 2009 12:56:20 AM
