Academic Analytics matches a huge number of honorific awards (10,000+) to individual scholars in the American academy. AARC researchers recently began digging through this data trove, and some summary statistics by discipline offer a glimpse into the deeper patterns we’re investigating. We started with academic department faculty lists for the 2019/2020 academic year. We then matched national or international awards (no state or local awards) bestowed upon those academics between 2017 and 2019, and created a table showing the number of awards won per faculty member in each discipline (the table can be sorted, and it’s paginated – only showing 10 rows at a time):

Sorted by the number of awards per faculty member (ascending), the fewest awards per person tend to be in humanities fields (theater, languages, etc.). At the other end of the list (sorted descending), the greatest number of awards per person tends to be in engineering disciplines, with some exceptions: Agricultural/Biological Engineering and Bioengineering, Consumer and Human Sciences, Materials Science and Engineering, Applied Physics, and Chemical Engineering.

It’s fascinating to see the distribution of honorific awards, but it calls into question how representative the data are – in other words, its possible that Academic Analytics happens to capture more awards in engineering than in humanities due to a previously unrecognized collection bias. It’s also possible that there are simply more awards available to engineers – maybe there are more scholarly societies who bestow awards in engineering fields? In any case, it’s clear that honorific awards, for which there exists no equivalent of a standard metadata description or widely-accepted unique ID number (such as DOI) should be interpreted in the context of both availability and potential collection biases.

We advocate the use of honorific awards as a post hoc indicator of research excellence (and sometimes precursors to research excellence, in the case of honors bestowed upon early career researchers), but we also caution that they are not as uniformly distributed nor standardized as bibliometric data or data about research grants. Our research is leading us towards creative solutions to these issues, and we welcome your thoughts.

Anthony J. Olejniczak, Ph.D.

Director, Academic Analytics Research Center (AARC)

Preprints have been around for a few decades, but posting preprints to a repository has only become the new normal for scholars in recent years. Preprints allow researchers to stake a claim to their ideas and results by establishing a clear and timestamped record of their work, even if the peer review process drags on for months. Preprints also facilitate rapid communication among scholars, which can be critical during times of crisis; the COVID-19 pandemic, for instance, led to a surge in preprint publications across several fields of study.

The rise of preprints leads to many new research questions, including that asked by Pagliaro (2021) in a recently published paper in the journal Publications: do manuscripts change substantially between preprint posting and the final, peer-reviewed version of the article? Following in the footsteps of scholars who studied this question in Physics and Biological Sciences, Pagliaro studied a small sample of Chemistry articles, finding “the differences between preprints and the corresponding articles published after peer review are small.”

The implications of only “small” changes to preprints as they wind their way through the editorial and peer review process raises questions about the institution of peer review (also discussed by Pagliaro). There are also important implications for bibliometric and scientometric data aggregators (including Academic Analytics), however, who typically include “peer-reviewed” as a criterion for including a scholarly research artifact in their databases (effectively precluding preprints from having a role in the research strategy and planning exercises that are carried out based on these databases).

On one hand, including preprints in bibliometric databases necessitates substantial additional investment in disambiguation and data merging. The preprint (which often has its own DOI) eventually needs to be linked to the final published version of the paper so the same research output is not recorded as two artifacts of research (or at least so the end user can identify that the same manuscript resulted in two artifacts) rather than “double-counting” the manuscript. There are also cases where preprints never result in a peer-reviewed journal article; in these cases, counting preprints among the number of publications for an institution/department/scholar may incentivize the production of preprints for which the author has no intention of ultimately putting the ideas or results through peer-review.

On the other hand, excluding preprints from bibliometric databases signals that preprints are not valuable enough to be considered among the other artifacts produced by scholars (“value” here meaning the purported value conferred through the peer review process). Clearly this is not a fair characterization of preprints, which have tremendous value. With the efficacy of peer review increasingly called into question, it may be time for bibliometric database providers to mobilize resources to solve the problems of “double-counting” and what to do with preprints that never make their way into traditional journals.

We are eager to hear your thoughts on preprints and whether (and how) bibliometric databases can include them to more fully represent the research outputs of scholars.

Reference Cited:

Pagliaro M. Preprints in Chemistry: An Exploratory Analysis of Differences with Journal Articles. Publications. 2021; 9(1):5. https://doi.org/10.3390/publications9010005

AARC scholars work with many datasets describing the publication outputs of research faculty. These datasets are almost always zero-inflated, or at least are skewed toward the lower end of the distribution. This phenomenon is so common we’ve even changed how we perform regression analyses to account for these skewed distributions (e.g., we ran hurdle regressions in our paper on Open Access publication trends). The histogram below shows the average number of journal articles published by scholars in departments classified as “Physics” over the past 10 years:

The data are clearly skewed towards the left side of the plot, between 0-100 articles per person over 10 years. The mean number of articles published over 10 years is 91.7 (green vertical line) and the median number of articles is 32 (red vertical line); however, a few physicists have as many as 1,100 articles over that 10-year span. These scholars are generally associated with massive multi-institution and multi-year projects such as CERN, so we looked at several other disciplines outside the natural sciences to see whether the pattern persists – the image below is for English Language and Literature journal articles over the same 10-year period:

Indeed, the distribution looks similar to that seen for physicists. In English, the mean number of journal articles per person over 10 years (green vertical line) is 3.8, while the median is 2.0 articles (red vertical line). A small number of English faculty members have published upwards of 50 journal articles over the 10-year period.

The skewness of these publication metrics complicates interpretation of discipline norms, with meaningful consequences for the faculty, administrators, and other committee members charged with comparing discipline publishing patterns for strategic planning. In English, the median number of journal articles published is about one half of the mean value, and in Physics the median is about one third of the mean value. Although means are commonly used in bibliometric comparisons, choosing the mean as the unit of comparison biases the data towards the few examples at the extreme right end of the publishing distribution.

For these reasons, we believe the median is often the more appropriate measure for intra- and interdisciplinary analysis and bibliometric comparisons. University-wide planning and evaluation are better served by focusing on discipline (or peer group) norms such as the median, rather than numbers incorporating the most extreme cases and perhaps setting unrealistic publishing expectations.

(the data above are from database version AAD2019-1470)