There tends to be a lot of hearsay and rumors in the academic job market and much of what gets spread around is based on N=1, as in, a friend of mine was on a search committee and told me....
For this reason a paper that just came out in the Israel Journal of Ecology & Evolution might be of interest.
Marshall et al. used a survey-based approach via the EvolDir and ECOLOG-L listserves to obtain data from individuals recently hired into TT positions. Their disclaimers are as follows:
We recognize several limitations of data gathered from online, anonymous, voluntary surveys. We were explicit in our instructions that participants only take the survey if they had been offered their first tenure-track job or equivalent position within the last four years and that they answer the questions as they applied at time of hire, but we are fully aware that confusion with regard to either of these instructions could inflate the numbers. Additionally, we cannot account for the bias of surveying principally from subscribers to EvolDir and ECOLOG-L email directories, as subscribers as these directories tend to be geared towards research rather than teaching issues. Persons employed at institutions requiring heavy teaching loads and lighter research requirements may have been less likely to participate. Therefore, the surveyed faculty may not represent a true cross section of successful first-time academics.
What they ended up with was a data set of 181 participants from countries all over the world. They asked pretty basic questions, including number of years as a postdoc, number of pubs (total & 1st author) at hire, etc. On the publication front, they found that:
...first-time hires had, on average, two first-author publications in journals with impact factors between 2 and 10... and one first or co-authorship paper in a higher impact journal like Science, Nature, PLoS Biology, or Trends in Ecology and Evolution...
For many of the categories, the variance was high. For instance the average age of hire for people at PhD-granting institutions was 33.1 +/- 4.1 and the average time spent as a postdoc was 3.08 +/- 2.15. The average number of pubs for that same group was 12.75, but with a whopping SD of +/- 7.63.
Most of the data are pretty much summed up in the first paragraph of the discussion, where the authors state:
Although significant variation exists in all categories and within all categorical groups, the qualitative message of this study is that prospective ecologists and evolutionary biologists are required to dedicate significant resources to publishing high quality papers,
applying for grants, and teaching courses if they want a reasonable chance of eventually landing a permanent position at a college or university. This will not come as a surprise to most, but what is striking are the qualification of the average successful candidate regardless of level of institution, region of the world, or gender. The successful candidate will most likely be in their early 30s, will have spent several years as a postdoc, taught multiple courses, received several grants, and will have published more than ten articles, with the majority of these articles appearing in high impact journals (Table 1, Table 2). These statistics suggest that all students considering careers in ecology or evolutionary biology should expect a highly competitive market that most likely will require substantial time investment.
I found it interesting that they found some different trends, based on gender and between the US and Europe / UK.
On average, successful applicants from the UK and Europe were younger at age of hire, spent more time as postdocs, had more publications, and received more large grants than individuals from the US (Table 1, Table 2). This could possibly be accounted for in part by the fact that many European Ph.D.s take only 3 years to complete rather than the typical 4–6 years in the US. Female applicants from doctoral institutions in the US generally had lower averages than males in these same categories, but this pattern did not exist in comparisons between genders within the UK category. However, it should be strongly noted that these differences between genders for doctoral institutions in the US are qualitative and not statistically significant.
The Marshal et al. paper was also followed by two responses, one by Roy Turkington, a faculty member in the Botnay Department at the University of British Columbia, and the second by Douglas W. Morris, who is a faculty member at Lakehead University. The Turkington article focused on a resent faculty search at UBC, where two positions were available. Dr. Turkington breaks down the candidate pool for those searches and reports much the same story as the Marshall et al. article. The Morris paper, entitled "Life History and Multi-level Selection in Academe" is at least worth a read for such gems as:
Euphemisms called “labs” coexist in structured universal aggregations where they compete with one another for scarce resources. Labs cooperate to produce copious numbers of zygotes, most of which disperse synchronously each year. The strongest find their way into the protective brood pouches of crusty adults who shed soft-shelled offspring at regular intervals (slowly developing zygotes die by the incompletely understood process of academic apoptosis). Juveniles develop a hard external carapace by intermittently joining and extracting themselves from other labs. The hardened but vulnerable sub-adults then join a common pool where they compete for space and position on rapidly eroding substrate in the universal aggregation. Many become dormant and fail to contribute to the gene (meme) pool. Some return to the lab as brood-rearing helpers. Few survive the rampant competition and frenzied cannibalism in the pool. Not all of the survivors are safe on the fragile substrate. A second apoptosis-like event eliminates the weak and meek. Only the most persistent or aggressive remain.
All in all, it's worth a read, even if you are not in the fields included. Data from early TT faculty don't often surface easily, and despite the inherent biases of the study, there is something tangible here to look at. It may not be what many are looking to hear right now, but it's data.