The Ph.D. degree is a prerequisite for advancement in academia and valued in most research settings. Although there is tremendous diversity in the characteristics of Ph.D. programs, they all require completion of an original piece of work, the doctoral dissertation. Its completion signals that the new Ph.D. is capable of conducting original, scholarly work. Traditionally, the doctoral dissertation has had two major functions: one scholarly and one educational. The former function stems from the commonly accepted notion that the dissertation makes a positive, original, and even significant contribution to knowledge in the discipline. The latter stems from the notion that the dissertation experience provides training in research and scholarly techniques that prepare the doctoral student for a career of productive research. With shrinking prospects for academic jobs in most disciplines (Braddock 1978), the traditional approach to training graduate students is being reexamined (Behnke 1977), and the importance of the dissertation to graduate education and a scientist's career, whether academic or not, is being challenged. Writing with specific reference to biologists, Reid (1978) presented five criticisms of the traditional dissertation: (1) It is not a useful tool in scientific communication; (2) it is a poor educational tool; (3) it places an undue burden on the doctoral student; (4) it is not evaluated by widely accepted standards; and (5) frequently, it generates no publications, even if it is of high quality. These concerns challenge the ability of the dissertation to perform the two traditional functions. In fact, Reid suggested that writing the dissertation might be dysfunctional for the student, since biologists tend to publish short journal articles and not extended monographs. Although he acknowledged that the dissertation frequently serves as an information source for later publication, Reid argued that writing the dissertation may place an undue burden on the graduate student and convey an incorrect impression of normal scholarly work in the field. These are academic concerns; the pertinence of the dissertation to nonacademic research careers is even more problematical. Because true experimental comparisons are not available, our approach to test these assertions was two-fold: We defined a sample of Ph.D.s who have had about a decade of post-Ph.D. experience to establish their career directions. We then gauged the effectiveness of their dissertations by the number of open literature publications generated and the rate of citations that those publications have accrued. Initial listings of the Ph.D.s' publications were compiled from Science Citation Index; these were revised by the respondents and categorized as to their relationship to the dissertation. Automated tabulation of citations to these publications was then performed by the Institute for Scientific Information. As for training effectiveness, we asked the Ph.D.s for their retrospective p rceptions of a number of dissertation feature . We also discussed doctoral training with a small sample of ABDs (persons who have completed all Ph.D. requirements except the dissertation before leaving graduate programs). These issues and others were addressed in a study of the role of the doctoral dissertation in career productivity, which examined a sample of doctorate holders in six scientific disciplines (Porter et al. 1981). The sample consisted of 1969-70 doctorates queried in spring, 1979. This paper focuses on data from the zoologists and biochemists included in the larger study. The other disciplines were physics, electrical engineering, psychology, and sociology. We conceive of the preparation of a doctoral level scientist as a process. The candidate's career aims and such features of a graduate program as research facilities and the rated quality of the department set the stage. The training process then leads to certain critical early career choices.
Read full abstract