Keith Crank has a BS in mathematics education and an MS in mathematics from Michigan State University and a PhD in statistics from Purdue University. Prior to joining the ASA as research and graduate education manager he was a program officer at the National Science Foundation, primarily in the probability program.

Since coming to work for the ASA, I have spent much of my time tracking down data on the statistics (including biostatistics) discipline. Some of this data comes from our own surveys, some from joint surveys we do with the American Mathematical Society, and some from government agencies.

For degrees in statistics, my primary source is the Integrated Postsecondary Education Data System (IPEDS), which is maintained by the National Center for Education Statistics (NCES). I thought this contained exactly the kind of data I wanted until a few months ago.

In August, as I was preparing to send a list of PhD-granting departments to U.S. News & World Report, I found they only wanted those departments that had awarded at least one PhD in each of the past five years. Although I sent an email to each of the department chairs asking for this information, the short time frame on my request meant some did not have time to respond. I did not believe this would be a problem, as I had access to the IPEDS data, which provided information about degrees by discipline and institution.

Graduate Training in Biostatistics
The National Institutes of Health (NIH) recently held an open meeting of the director with representatives from professional societies as a first step in creating a dialogue about how NIH can better address concerns of the members of those societies. I believe many of you would like to see an increase in the number of graduate traineeship awards going to biostatistics. This was the issue I raised during the question-and-answer period.

About a week later, I received an email pointing to a biostatistics traineeship program at the National Institute for General Medical Sciences (NIGMS). This program was initially a separate competition for biostatistics, but has since been folded into the broader Ruth L. Kirschstein National Research Service Award (NRSA) Institutional Research Training Grants (T32). Unfortunately, there are still links to the biostatistics program announcement, which has expired, making it appear there is no longer interest in training awards for biostatistics. I’ve been informed, however, that they do want biostatistics applications.

The program manager for biostatistics is Shawn Drew, who can be reached at DrewL@nigms.nih.gov and (301) 594-3900. The next deadline is January 25, 2010.

However, when I compared the IPEDS data to the responses I received from the biostatistics chairs, I discovered it didn’t match well. (On the statistics side, there didn’t seem to be a problem.) Of the 24 biostatistics departments that graduated at least one PhD in each of the last five years, eight would have been rejected based on the IPEDS data. (By that, I mean the institutions of these eight departments reported no biostatistics PhDs for some or all of those years.) Although I can understand minor discrepancies—because of memory bias and time frame (calendar year versus academic year)—I was surprised to find a problem with one-third of the departments.

Unfortunately, identifying that a problem exists is not the same as solving a problem (or even knowing the cause of it). Let’s begin with the cause of the problem. It could be an issue with the IPEDS survey form. Biostatistics degrees are included under biology, not medicine and public health. (There is a code for health and medical biostatistics, but it may no longer be of use, as there are no data for this code after 2002.) The cause of the problem could be at the institutional level. Degrees from a school of public health should be coded under a public health category, and there is no such category for biostatistics. Or, it could be an issue with the person completing the survey. That person may only be considering public health categories for degrees awarded by the school of public health. (Other explanations also are possible.)

For those of you at academic institutions, do you know who provides the data for the IPEDS Completions survey? Do you know how the data for your department are entered? Please contact me if you can help me resolve this problem.

On a related note, what should I be doing with biometry/biometrics degrees? Should they be counted as statistics degrees?

To contact me, send an email to keith@amstat.org. Questions or comments about this article, as well as suggestions for future articles, are always welcome.