Publish or perish: Should graduate students publish before graduation?Posted: June 11, 2012
The Graduate Education Initiative (GEI) discussed in Educating scholars: Doctoral education in the humanities was focussed on improving graduate education.
Two important premises of the project were that improvement would involve reducing time-to-degree (TTD) and both reducing and shifting attrition among entering cohorts (so that less people dropped out and those that did dropped out earlier). Driving these premises lay the startling statistics about TTD and attrition in Humanities PhD programmes: about 50% of entering students left their programmes before completion, and about 1/2 of those who stayed failed to complete their degree by the end of their seventh year (the mean TTD across departments was 7 years, 3.5 months).
Speeding up time to degree, however, brings with it potential costs as well: perhaps humanities graduate students are taking so long to write their dissertations because the work is intrinsically time-consuming, or because they are concentrating on quality over speed, or because they are writing articles and otherwise polishing their resumes in preparation for the job market. Does speeding up TTD at the cost of these activities improve graduate education?
That the work is intrinsically time-consuming seems to be borne out by the lack of success the GEI seems to have had on post-candidacy dissertation work (i.e. the actual research and writing of the dissertation). While the programme was able to reduce the mean TTD across departments by about 8.5%, this change came entirely from the pre-candidacy period (i.e. coursework, comprehensive exams, and dissertation prospectus). The mean time spent writing the dissertation, on the other hand, actually went up by 2.6 weeks during the programme on average, with some departments showing much more significant increases (the mean for philosophy and religious studies students was an additional 5 months). Only in Music (and to a lesser extent Classics) did students spend significantly less time writing their dissertations as a result of this programme, booking a mean decrease of 9 months and 2 months respectively.
But is worth foregoing activities like publication that might lengthen TTD? Publishing while a graduate student turns out to have an extremely clear positive effect on post-graduation Tenure Track employment, especially in the longer term. As the following table (p. 195) shows, one publication in graduate school results in almost a 25% better chance of being employed in a Tenure Track position three years after graduation; with three or more publications, your chances of being in a Tenure Track position are double those of a graduate with no publications.
Lengthening time to degree, on the other hand, does adversely affect your chances of getting hired to a Tenure Track academic position: approximately 63.5% of students graduating in less than seven years during the period studied by the GEI were in Tenure Track academic jobs after three years; students with TTD of 7 or more years were less likely to be similarly employed: after three years, about 61% of students who took seven years to complete their PhD were in Tenure Track jobs; 57.5% of students who took 8 years; 48.6% of students who took 9 years; and 39.3% of students who took 11 or more (for some reason students who took 10 years did better than the class before or after them: 51% of them were in Tenure Track Jobs 3 years after graduation).
So what’s a student to do, publish or reduce time-to-degree? According to the authors of Educating Scholars, both: reduce TTD and publish while a graduate student. In fact, as they point out, the two are not necessarily opposed: the students most likely to publish as graduate students are those who finished their PhDs in five through seven years: students who finished in five years have a significantly higher probability of publishing in graduate school than those who graduate in six or seven years; those who graduate in more than eight years have a significantly lower probability of publishing (Table 10.3).
This may suggest that students who are taking a longer to complete their work are doing so because they are having conceptual or research difficulties rather than because their work is necessarily more polished or of higher quality. Perhaps, in contrast, students who do publish are also more able as a result of this process to see their dissertations as discrete (and hence intellectually and temporally bounded) pieces of work that are open to timely completion. Interestingly (and perhaps confirming this hypothesis) students who enter PhD programmes with an M.A. in hand, and hence who already have experience with completing a relatively major piece of scholarship, are both more likely to publish in their PhD programmes and more likely to complete their degree in a timely fashion (p. 212 and table 4.4). This is despite the fact that, on average, they enter their programmes with lower GRE scores.