Nigel Palmer, honorary research fellow with the Centre for the Study of Higher Education at the University of Melbourne is speaking on "Attrition and Completion in Doctoral Education: A Wicked Problem?". He has been researching completion rates for doctoral candidates by international/domestic status and gender (and produced papers on education). He is looking at how to better support candidates, to improve completion times and rates completion of their degree.
Nigel argued that even if universities are full fee paying, students dropping out is still a problem. He then discussed if it was a wicked problem. He suggests it is because of the difficulty of measuring attrition as well as the management of attrition itself.
One reason for worrying about attrition is equity of access to education. Nigel showed a graph showed the increasing number of women commencing a doctoral program over time. However, he then had a second graph showing that women were over represented in educaiton and helath, but many fewer in IT and Engineering. One interesting trend was that in education and health participation was higher in undergraduate degrees than postgraduate ones, while it was revered in IT and engineering.
Nigel then showed a series of graphs showing international versus domestic students for the group of eight universities. Unfortunately all the colored likes look blue to me, so I could not see what the graphs were showing.
Nigel suggested that some measures to lessen attrition, such as research methods courses, will also improve the graduate outcomes (courses for doctoral students is a hot topic at ANU at present).
Nigel described a "crude" research doctoral degree completion rates, stupid statistic. This is calculated from the number of students completing in one year divided by the number there were the year before. This is not a very useful statistic because students normally take longer than a year to complete a degree. There have been attempts to come up with better measures, using longer periods and techniques such as a three year moving average (three years being more typical for degree completion). Unfortunately as Nigel points out such calculations are skewed by recent growth of international enrolments in Australia, which makes the relative competion rate look low.
The US PHD Completion project found completion rates of 49% for humanities to 65% for engineering. Similar studies for ANU showed a completion rate of about 80% for full time students. Niegl's research is showing a similar high completion rate, but it is too early to report definitive results. The figures show an interesting S curve, with the rate of completions increasing rapidly for about the first five years and then tapering off. Nigel pointed out that there were a significant number of completions from five to ten years.
It is surprising how difficult it is to get an agreed measure and then actual statistics for completion rates for degrees. Universities are in the business of producing graduates (and this is a very large export industry for Australia). It would seem reasonable to have agreed measures for university output.
If there were more reliable statistics, it would be possible to consider measures to aid completion rates. But will this really have a useful effect on measures. As an example, if domestic male students complete at a lower rate than international female students, would it be appropriate to set a higher entry standard, or resources provided, on the basis of gender?
Nigel concluded that doctoral attrition was not a wicked problem and this this might engender a sense of helplessness.
I suggest some easy measures to address attrition would be to provide more flexible programs. As an example, Australian postgraduate students could be offered programs which allow for a mix of coursework and research. As is usual in North America, students would start with coursework and then, depending on their interests and aptitudes, focus on coursework or research. Also the option of the student completing a qualification lower than a doctorate, without feeling they have in some way failed, would be useful. In addition it would be useful to have accessible support services, which are available on and off campus, would be useful. Also services for monitoring supervisors in the way teachers are, would be useful.