March 17, 2012 Leave a comment
This post is intended to disseminate some data that I collected on typical journal publishing rates for recently tenured faculty in Electrical and Computer Engineering. I considered publishing this information in some scientometrics journal, but decided it would be more valuable as a Googlable online resource. This information was useful to me when I prepared my tenure case at Utah State University, and it might be useful to others who need to perform quantitative evaluations of publishing productivity, for themselves or for others.
I believe that this kind of quantitative assessment has been lacking in many academic fields. Administrators commonly rely on statements of opinion from external reviewers, but I am concerned that this review process introduces considerable subjectivity. A reviewer may deliver an evaluation on the basis of what they want to see, but this is not necessarily a well-calibrated judgement. I therefore decided to undertake this study in order to answer a simple question:
What is the actual productivity, as measured by the rate of publications in archival peer-reviewed journals, that is achieved by individuals who receive tenure at research universities in the US?
By addressing the question in this way, we may assess the level of productivity that our peers actually find acceptable, as opposed to what they find desirable. This measure is, of course, only one of the many measures used to evaluate academic productivity. The narrow question of raw publishing is nevertheless an issue that affects many junior faculty who may be accused of low productivity during pre-tenure assessments.
The table below shows percentiles for the 5-year publishing level of recently tenured professors in Electrical and Computer Engineering. For example, the percentile for 5 papers is 34.5%, meaning this percent of faculty published 5 or fewer papers in archival journals during a five-year interval. Among the CVs sampled, the median publishing rate was 1.43 articles per year, and the mean was 1.76 articles per year. In a few cases, individuals were found to receive tenure with as few as 0.33 articles per year (this was confirmed via Web of Science). The highest publishing rate was an average of 5 articles per year.
|Journal Pubs. (5yr)||Percentile|
The data was collected from the CVs of 29 faculty, mostly in positions at Land Grant Universities in the United States. In most cases, samples were taken from faculty who received tenure more recently than 2000, though the exact date of tenure could not always be verified. The majority of samples represent younger faculty who received Ph.D. degrees more recently than 1995. There are at least three samples of faculty who received degrees earlier than 1995, and some faculty members did not report this information on their CV. The data was collected from ECE department web sites at the following universities:
|U. C. Davis|
|N. M. State|
|N. C. State|
The data represent faculty who perform research in the following ECE disciplines:
- Digital Signal Processing
- Materials and device engineering
- Computer Architecture
- Space physics
- Control systems
The data indicate that there is a wide variability in publishing productivity in ECE fields. Some hypotheses for this variability include:
- Some ECE disciplines emphasize conference publication over journal publication.
- ECE disciplines vary in the time, cost, scope of work and/or length per article.
- Some institutions may have more relaxed expectations than others.
- Some individuals may receive tenure based on factors other than raw publishing productivity.
The method used to obtain these results is not ideal. Individual professors do not necessarily disclose all publications on their CVs, however a full listing of journal articles seems to be common practice. It is possible that the results are distorted by some individuals who under-report their publishing. The exact numbers must therefore be considered rough estimates. Although the data is not exact, we can reasonably make some general conclusions, such as:
- It is extremely rare for someone to publish more than 13 articles in a 5-year period.
- The majority of faculty publish fewer than 10 journal articles in a 5-year period.
- Some faculty are receiving tenure with fewer than 5 articles in a 5-year period.
Although the method used to obtain these results is not fool-proof, in my opinion it would be unreasonable to dispute these conclusions. I welcome any criticisms that I may not have considered.