Where postdoctoral journeys lead

TL;DR

The paper addresses the limited understanding of how postdoctoral training shapes early-career academic success. It constructs a large, cross-disciplinary dataset combining MAG publications and online CVs, defines an eta-index as the early-career citation-impact measure, and uses topical change (via Jensen-Shannon divergence) and mobility data to predict outcomes. Postdoc period strongly influences whether researchers stay in academia; achieving hit papers during either the PhD or postdoc increases likelihood of continued academic work, and moderate topical changes along with international mobility are associated with higher early-career performance. The results argue for treating the postdoc as a central, contributory phase in the academic job market and suggest policy and career guidance should encourage active postdoc engagement, mobility, and topical diversification.

Abstract

Postdoctoral training is a career stage often described as a demanding and anxiety-laden time when many promising PhDs see their academic dreams slip away due to circumstances beyond their control. We use a unique data set of academic publishing and careers to chart the more or less successful postdoctoral paths. We build a measure of academic success on the citation patterns two to five years into a faculty career. Then, we monitor how students' postdoc positions -- in terms of relocation, change of topic, and early well-cited papers -- relate to their early-career success. One key finding is that the postdoc period seems more important than the doctoral training to achieve this form of success. This is especially interesting in light of the many studies of academic faculty hiring that link Ph.D. granting institutions and hires, omitting the postdoc stage. Another group of findings can be summarized as a Goldilocks principle: it seems beneficial to change one's direction, but not too much.

Figures (3)

• Figure 1: Who drops out of academia after a postdoc. Panel A shows the percentage of postdocs proceeding to a faculty position as a function of the relative change in their publication rate between their doctorate studies and their postdoc. The bins divide the number of observations equally, except the $-\infty$ point (representing those without any publications as postdocs), which includes more observations as indicated by the larger area. Panel B shows the effects of having a hit paper during the Ph.D. program or the postdoc. The category "YN" means people with one or more hit papers during their Ph.D. training but none during their postdoc---Y(es) for PhD. student, N(o) for postdoc---and so on. The individuals without a hit paper as a postdoc are significantly less likely to pursue a faculty career than those with at least one (with an effect size given by Cohen's $d>12$). In both panels, the error bars represent standard errors.
• Figure 2: Factors influencing the success of early career scientists. Panel A shows the average values of our success metric $\eta$ for the same categories as Fig. \ref{['fig:bottleneck']}B (those in these categories that stayed in academia after their postdocs). Panel B displays $\eta$ as a function of the topical difference between the publications as a doctoral student and a postdoc measured by the Jensen-Shannon divergence (JSD) of the annotated disciplines of the publications. The smooth line is a fitted fourth-order polynomial. The vertical lines show the (equal sample number) bins (except for the zero divergence observations that form one bin). For reference, in our case (with 19 topical categories), changing from publishing on only one topic to publishing on only another gives a JSD of $0.83$. Panel C shows the average $\eta$ value for physical mobility---whether the Ph.D. moved to another country, another university in the same country, or stayed at the same university. Finally, panel D shows a similar plot for different classes of postdoctoral trajectories---whether the move to or from a top-10 university or research institution, or stays in the top-10 vs. other categories, and how these moves depend on the region of the Ph.D. granting institution---Europe, The US and Canada, the rest of the world.
• Figure 3: Illustration of the $\eta$ measure. We focus on papers published between two and four years after the beginning of the faculty career. We gather the $c_2$ scores of these papers (the number of citations within two years of publication. Then, $\eta$ is the h-index of the set of $c_2$ scores.