Wednesday, April 18, 2012

To get a job in philosophy

Carolyn Dicey Jennings has done amazing work running the numbers on Tenure-Track hiring this year. Here is her report (emphases added):
Overall, what are your chances of getting a job as a philosopher? Well, if we just look at overall numbers, there are 139 jobs (188 including post-docs) and around 800 estimated candidates. That is, over 700 candidates applied to Barnard, so I am conservatively guessing around 800 candidates total. Thus, overall prospects are at around 24% chance of getting any job, 17% chance of getting any tenure-track job, 6% chance of getting a ranked tenure-track job.

What if we add data about your PhD granting institution? Well, for American institutions we can use the NRC numbers on average numbers of PhDs granted from 2002-2006 to estimate the number of candidates from each school on the market and compare that to how many people from that school got jobs. The numbers from this are reported to the right. *NOTE: I only report the top 66 institutions ranked by the NRC, according to the 5% R score. Thus, when I say "ranked by the NRC" I mean "ranked 66 or above for the 5% R rankings of the NRC."

On average, the ranked NRC schools granted 3.9 PhDs each year from 2002-2006. Thus, something like 257 ranked graduate students can be expected to be on the market from the United States in any given year. 132 of the successful candidates were from NRC ranked institutions (101 of successful tenure-track candidates were from NRC ranked institutions, and 27 of successful ranked tenure-track job-getters were from NRC ranked institutions). Thus, one's overall chance of getting any job (post-doc or tenure-track) coming from an NRC ranked institution may be as high as 51%, 39% for any tenure-track job, and 11% for a ranked tenure-track job.

What if we add information about your gender? The top 66 programs I selected from the NRC data had an average of 29% female graduate students in 2005 (and 20% female faculty). Thus, around 75 women from ranked departments are likely to be on the market at any one time. 7 NRC ranked women got NRC ranked jobs, and 33 NRC ranked women got tenure-track jobs in general. Thus, if you are a woman from an NRC ranked department looking for a ranked job, your chances might be around 9%, whereas if you are looking for a tenure-track job in general they at are around 44%. If you are a woman from an NRC ranked school looking for a post-doc, be advised that only 15% of ranked women achieved post-docs this year (5 out of 34 ranked post-doc achievers), whether or not the post-doc was itself ranked. Because of that fact, the chance of a woman from an NRC ranked department getting a tenure-track job or post-doc is about the same as for a man from these departments: 51%.

Means and Medians

1. The mean worldwide Gourmet ranking of the PhD granting institution for those who got a tenure-track job(rank=25) is significantly lower (p=.01) from those who got a ranked tenure-track job (rank=20). Miss Median Job-Getter comes from either Texas or U Chicago, whereas Miss Median Ranked Job-Getter comes from Berkeley.

2. Using Kieran Healy's rankings of areas of specialization as they contribute to the overall Gourmet ranking (ranked from 1-12, M&E to Continental), the mean specialization rank for those who got a tenure-track job (rank=5.2) is significantly lower (p=.01) than for those who got a ranked tenure-track job (rank=4.3). Miss Median Job-Getter does Political Philosophy, whereas Miss Median Ranked Job-Getter does Philosophy of Mind.

3. Those ranked institutions offering post-docs and fellowships (rank=18) have a significantly higher (p<.00001) mean worldwide Gourmet ranking than those ranked institutions offering tenure-track jobs (rank=36). Mr. Median Ranked Post-Doc was hired by UCLA, whereas Mr. Median Ranked Job-Getter was hired by UC Irvine. (Moreover, 2/3 of tenure-track jobs are unranked, but only 1/3 of post-docs are unranked.)

4. Those institutions offering post-docs and fellowships (16% women) hire significantly fewer (p=.01) women than those institutions offering tenure-track jobs (35% women). There is a non-significant difference between the numbers of women hired by ranked and unranked institutions (35% unranked TT, 39% ranked TT, 16% unranked PD, 15% ranked PD).

5. The mean number of peer-reviewed publications for those going to ranked tenure-track jobs was 3 (Median=2), whereas the mean number for all post-docs was 2 (Median=2).

Correlations

6. The most substantial correlation is between the hiring institution rank and PhD granting institution rank, at .3 (a mild to moderate correlation). That is, the higher your PhD granting institution is ranked the higher your hiring institution will be ranked.

7. I also found a negative correlation between PhD granting institution and number of publications (-.17: the lower your PhD granting institution is ranked the more peer-reviewed publications you have) and between gender and number of publications (-.21: if you are a man you likely have more publications than if you are a woman). Some caution is needed here because I only looked up the number of peer-reviewed publications for those getting ranked tenure-track jobs or (ranked or unranked) post-docs. Thus, to get one of these you may have needed more publications if you are from a lower ranked institution and fewer publications if you are a woman.

Comparison with the NRC Rankings

If you do the same thing for the National Research Council’s “R” and “S” Rankings (100 Philosopher Survey), some interesting things happen. For “R,” Miss Median Job Getter comes from UPenn and Miss Median Ranked Job Getter comes from Syracuse. These are Johns Hopkins and UCLA, respectively, for the “S” Rankings. More interestingly, there is no significant difference between ranked and unranked departments preferences in area of specialization, both of which come to a mean rank of around 5.

I then looked at the National Research Council reports for number of PhDs granted between 2002 and 2006, and then compared those averages to the number of successful candidates (post-doc or tenure-track job) to yield a ranking of schools. Interestingly, this ranking correlates equally well (both have a coefficient of .3) with the NRC “R” rankings and the Gourmet rankings. (I also computed the “S” rankings, which do slightly less well, with a coefficient of .25.)

Carolyn provides a link to a protected Excel spreadsheet:
http://neuphi.com/market.xlsx
Everyone thank Carolyn for her work; soooo great.

-- Jaded

86 comments:

gwern said...

> The most substantial correlation is between the hiring institution rank and PhD granting institution rank, at .3 (a mild to moderate correlation). That is, the higher your PhD granting institution is ranked the higher your hiring institution will be ranked.

That's a lower correlation than I would have expected.

Anonymous said...

I'm fairly confused by the graphic. What is placement rate and how is it calculated?

Anonymous said...

Wouldn't a conservative guess be much greater than 800? I think the number should be closer to double the 700 mark. I don't think 7 of every 8 job candidates applied to Barnard.

I think the percentages in terms of a chance to get a job should all be about half what you have calculated. I think 1600 people on the job market is a better guess at this point. But with independent outcomes and not all the same people applying, I know the math gets hard.

A very nice effort, nonetheless! This is just the kind of thing that should be going on. So, let me say that I appreciate to work. For example, does anyone know how many folders there were at the job placement area at the Eastern. I think that number is a better indicator of total people than Barnard apps.

Oh, and the link the the file doesn't work properly.

Anonymous said...

The rate seems to be the number of placed graduates divided by the average number of annual graduates.

Anonymous said...

I did an analysis of my own (not as sophisticated) to look at the amount of previous positions held by hires. Based completely on Leiter posts, so not at all rigorous, but still a representative survey.

Results:
% of jobs that are tt
2012: 76% (139 of 183)
2011: 75% (84 of 112)

% of hires that reported previous positions
2012: 43% (79 of 183)
2011: 33% (37 of 112)

% of tt hires that reported previous positions
2012: 50% (70 of 139)
2011: 36% (30 of 84)

% of tt hires from previous tt positions
2012: 13% (18 of 139)
2011: 8% (7 of 84)

Conclusion: the backlog of qualified candidates is making it harder to find a tt job directly out of graduate school.

Carolyn Dicey Jennings said...

Anon 1:49, that was my impression, too. Nice to see some numbers!

Anon 12:48-the first set of numbers may be overly conservative, but some people I have spoken with on the hiring side think the numbers are much lower than even 800. I like your suggestion...anyone have that information?

A better link, perhaps, with some updated numbers from a few more postings: http://neuphi.com/market.xls

Anonymous said...

This is based on one year's data, and the data is incomplete (I assume it comes from the Leiter jobs thread). How can this be useful or even very meaningful?

Anonymous said...

Just as a follow-up to Anon 3:22, I know of three hires (1 recent, 2 over a month old) that are not on the Leiter thread; and I'm sure there must be many others as well.

The data we have is helpful, but it is likely skewed away from certain programs / placements (in particular, religious schools and continental departments, both of which aren't always fans of Leiter's blog).

Anonymous said...

to add to 322 and 420:

Also likely to be massively underrepresented are tt community college hires and other sorts of hires at places many R1 placement advisers would find 'not worth mentioning' or that they may not even be aware of (former students may not report their future moves).

Carolyn Dicey Jennings said...
This comment has been removed by the author.
Anonymous said...

This is Anon 4:20 again.

Just to make clear the sort of gaps that exist in Carolyn's data, I see that her spreadsheet reports 1 tenure-track hire and 2 postdocs from Notre Dame. The real stats from ND are at least 3 tenure-track hires and at least 4 postdocs -- and there are still one or two people who are waiting on final decisions.

Carolyn Dicey Jennings said...

That is a good worry...I seem to remember there being something like 180 TT jobs this year? If so, this report probably isn't more than 30 off. That could be a problem if they are all of a certain type. It would be nice if there were a fuller set of data available. Maybe the APA could collect this?

Anonymous said...

Carolyn, I for one appreciate all your hard work and find the data super interesting! So, thanks!!

Anonymous said...

I think this kind of work is definitely what we need in the field. The APA should make this type of work one of its main services.

Not directly related to the data but I have two quick questions:

1. Do you think not getting jobs is humiliation in the field? Do SC's share their low opinions of rejected candidates with others at conferences? How many PFO's is too many and one should just leave?

2. I've been getting quite a few PFO's full of praise in my application and ones from postdocs explicitly encouraging me to apply again next year. I take none seriously. The idea of re-applying the following year to the same postdoc seems to me a little faceless. But I cannot be sure, what do you think?

Anonymous said...

Carolyn,

I don't understand the gourmet rankings used. You claim you used the worldwide rankings, but then lots of schools are given their US ranking e.g. UConn at 51 (mean 2.3) but LSE at 50 (mean 2.7). There are at least 10 schools between in terms of mean score.

Anonymous said...

Of the 700 who applied to Barnard, how many already had jobs but were looking to upgrade? It seems like the more job seekers there are who are also already job holders, the less grim the job market is on the whole, since 1) they are probably not applying to as many jobs as new Ph.D.s and 2) when someone does upgrade their original position will usually be advertised the next year.

Anonymous said...

Nice work, Carolyn. I particularly appreciate the stats that suggest that women aren't more likely to get a job than men - something to use next time I'm told by some disgruntled male philosopher that I only got a job because I am a woman...

(Of course it's possible that women's work on average is worse than men's, so it's possible that roughly equal chances of getting a job still mean some bias towards women in job seeking, but from the sample of job-getters that I know I don't think that is the case).

Anonymous said...

Such a statistic is very important and very useful to have. I very much appreciate Carolyn's effort and I think this analysis is useful even if the data isn't completely accurate.

What makes me incredibly angry and frustrated is that such statistics don't already exist and that we do not have more accurate data to begin with. Collecting such data and doing the analysis is clearly one of the main responsibilities of the APA. In fact, it's hard to see how in the world the APA is supposed to "facilitate the professional work and teaching of philosophers, and to represent philosophy as a discipline" (this is from the APA homepage), if it is completely in the dark about the number of hires, hiring trends, hiring trends broken down by gender/ethnicity, etc. This is truly a disgrace.

Given the lack of better data, the informal data collection and analyses done by individuals are useful and laudable. One can quibble with how accurate some of their estimates are. But the reason why we and they have to rely on such estimates in the first place is because of the incompetence of our professional organization.

Carolyn Dicey Jennings said...

yeah, this is tricky. I was torn between wanting to include overseas institutions and wanting to include as many american ones as the report would allow. I used worldwide Gourmet up to 49 and then estimated the rankings for other non-worldwide ranked institutions based on the mean distribution between rank and mean rating for the worldwide set (I think it was .5 mean rating per rank step) and applying that to the individual means of the (worldwide) unranked institutions. That included 66 institutions, the same number I included from the NRC.

Anonymous said...

"Do SC's share their low opinions of rejected candidates with others at conferences?"

One man's opinion, but honestly, I don't remember who any of the rejected applicants for our jobs are once we reject them. Nothing personal, but with so many names, the rejected applications all blur together. (Occasionally I remember the names of other finalists, but if they were finalists, I would have only positive things to say; there's a reason they were finalists.)

"no one likes a free rider and one of us has already done some work on that front whereas some others of us have just come for the bread after it is baked and complained that it doesn't have enough salt?"

I'm sure this is something we could talk about until the cows come home blue in the face, but loose lips stick out like a sore thumb. :)

Carolyn Dicey Jennings said...

I also have insider information about BU that I did not post, since I am concerned that the data points remain transparent and reliable (Leiter does a nice job of providing a list of hires that is both). Without that, all we have is "anon 4:20 claims ND has 3 TTs," which is hard to base much on.

Carolyn Dicey Jennings said...
This comment has been removed by the author.
Carolyn Dicey Jennings said...
This comment has been removed by the author.
Anonymous said...

"Thus, to get [a ranked tenure-track job or (ranked or unranked) post-doc] you may have needed more publications if you are from a lower ranked institution and fewer publications if you are a woman."

I would be interested in seeing some discussion of this claim.

Continental Pissant said...

Carolyn, thanks for running the numbers!

It would seem that the job market has taken less notice of Brian Leiter's attempt to contruct a version of "Continental Philosophy" that discredits "Party Line Continental" (whatever THAT means) departments than he seems to have hoped.

Seeing all those Leiter-branded "N/A" departments putting up respectable placement numbers ought to raise questions as to the validity of Leiter's dismissal of the education that goes on in such departments, to say nothing of the validity of his idiosyncratic and self-serving construction of "Continental" philosophy.

Anonymous said...

According to the spreadsheet, UC Irvine has 5.2 average grads and this year had three tenure-track hires. But the spreadsheet calculates our placement rate to be .19 instead of .58. Were the two of us from logic and philosophy of science left out of the placement rate calculation?

Thanks for the great spreadsheet. You've done all smokers a huge service by compiling it!

Anonymous said...

Perhaps I'm missing something. There were 188 total jobs, 132 of which went to graduate students from NRC-ranked schools in the U.S. who were on the market for the FIRST TIME? This can't be correct. The figure of 257 new Ph.Ds from NRC-ranked top 66 schools (in the U.S.) comes from multiplying the average 3.9 NEW PH.Ds by 66. So 257 is at best a VERY ROUGH estimate. And we need this figure to generate the 51% success rate figure (as well as others).

Over at Leiterreports, I counted 70(tried hard not to double-count--may have missed one or two) different hires who previously had Ph.Ds (inferred from the fact that they either had VAPs or post-docs). This brings the total number of hires up to 202, which is 14 more than the figure of 188 that the study cites. There is probably a slight discrepancy between the data now and the data that existed when this study was done.

Let's suppose that I've got 14 extra jobs by either double-counting or including postings made after this study was done. That still means that NOT A SINGLE NEW PH.D RECIPIENT FROM A NON-NRC TOP 66 SCHOOL GOT HIRED THIS YEAR. Yet over at Leiterreports, there are a number of new Ph.D hires listed from non-NRC top 66 schools, many of whom are from universities outside of the U.S.

If I have counted and calculated correctly, then there is a contradiction in the data. Have I made a mistake somewhere?

Anonymous said...

11:27,

Here's a stab. SC's must utilize some sort of vetting procedure. Pedigree to many matters. Thus, if you are from a lower ranked institution, your application is 'inferior'. One way to compensate is to publish, since number/quality of publications is itself a vetting metric. Finally, those who acknowledge that women are underrepresented in the profession may consciously or unconsciously pay close attention to a file that is obviously that of a female, yet feel no compunction in tossing the file of a similarly well-qualified male.

Carolyn Dicey Jennings said...

Anon 1:45, I thought more about your worry concerning the ranking of Gourmet schools over 49 and decided that my method is inconsistent with the Gourmet method. That is, the Gourmet method assigns a rank number based on mean value and only one school occupies each rank number, whereas I was trying to fit the distribution of rank number to mean value over 49 to that under 49, resulting in multiple schools of the same rank (based on my own bias that this is more fair to those institutions ranked over 49). I changed the report now to fit the Gourmet model and reported fewer institutions (63) to best fit the number of institutions I selected from the NRC (66), which I may expand to more institutions for both measures in due course. There was no qualitative shift in the findings, but the correlation between the Gourmet ranking and the placement rate is now worse off (it is now .23). See the neuphi.com/market.xls version for this update.

Carolyn Dicey Jennings said...

Ok, another update (mostly for 3:22), posted under "placement" tab of neuphi.com/market.xls. In this update I compare the placement rates from 2009 to those of 2012. The correlation between these placement rates is .32. Thus, there is definitely some variance year to year on this and longer term statistics would be more useful than 1 year statistics. The Gourmet and NRC R rankings, moreover, predicted 2009 placement much better than they predicted this year's placement (.43 and .49 in 2009, respectively, and .23 and .29 in 2012).

Anon 12:23: NCR ranks Irvine and Irvine LPS separately (as it does with Pitt) and reports the number of average grads separately for these, and since LPS is not in the top 66 I left it off of this early analysis.

Anon 12:50: you are right that the 51% statistic would only be true if everyone was "fresh." I haven't thought of a way to calculate it without this assumption (should I assume 3 years on the market for each person? 2?), but I am all ears on that front. This is important also for the first point, since the fact that candidates go on the market multiple years is presumably part of what enables UCLA, for example, to have a 265% placement record this year and a 118% placement record in 2009.

Anonymous said...

I, like 12:39, don't understand what "placement rate" means. I feel that if I understand this, then everything else will be clear.

Carolyn Dicey Jennings said...

Anon 3:11, "Placement Rate" is the number of candidates placed in a tenure-track position or post-doc/fellowship this year divided by the average number of PhDs awarded by that institution per year between the years 2002 and 2006. Admittedly, this latter number may not represent departments' current averages (i.e. a department may have expanded since that time frame), but it is a way of applying the known information about department graduation rates to this year's placement rate. Some schools are over 1 because they have more candidates on the market than usual, either because more graduated this year than usual or because some candidates are from previous graduation years. Under the "placement" tab there are placement rates for 2009, for the sake of comparison.

Anonymous said...

For Pissant and 12:50, from 1:49:

The following table is going to look totally gnarly, but here goes.

For each school, the numbers are 1) total number of hires 2) number of TT hires and 3) number of "direct" TT hires, where the hire did not report a previous position. This addresses the problem of "freshness." However, this does not take into account the size of the graduate program.

The last column (4) is the average of the ranks (among those listed) of 1-3. E.g., MIT was 1st in total hires and TT hires, but 2nd in "direct" TT hires, so average rank is 1.33. The schools are listed in order of average rank (4), then by number of placements (1).

MIT 9 8 4 1.33
Berkeley 7 6 5 1.67
UCLA 9 6 3 2.00
Pennsylvania 6 6 5 2.00
Chicago 9 6 2 2.33
Oxford 7 5 2 3.00
Fordham 5 5 2 3.67
UNC 4 4 3 4.00
Boston Univ. 5 4 1 4.33
Michigan 5 3 2 4.33
USC 4 3 2 4.67
Cornell 3 3 3 4.67
WUSTL 3 3 3 4.67
Riverside 5 3 0 5.00
Rutgers 5 3 0 5.00
Wisconsin 4 4 0 5.00
Toronto 4 3 1 5.00
Western 4 2 2 5.00
Irvine 3 3 2 5.00
Stanford 3 3 2 5.00
Connecticut 3 3 1 5.33
Oregon 3 3 1 5.33
Indiana 3 2 2 5.33
NYU 3 2 2 5.33
Melbourne 3 3 0 5.67
Binghamton 2 2 2 5.67
Columbia 2 2 2 5.67
Vanderbilt 2 2 2 5.67
Princeton 4 1 0 6.00
UCSD 3 2 0 6.00
Cambridge 3 1 1 6.00
Notre Dame 3 1 1 6.00
Northwestern 2 2 1 6.00
Texas 2 2 1 6.00
Arizona 4 0 0 6.33
Pitt, HPS 3 1 0 6.33
ANU 2 2 0 6.33
Syracuse 2 2 0 6.33
King's College London 2 1 1 6.33
York 2 1 1 6.33
UBC 2 1 0 6.67
Auckland 1 1 1 6.67
Buffalo 1 1 1 6.67
Davis 1 1 1 6.67
HU 1 1 1 6.67
Johns Hopkins 1 1 1 6.67
Missouri 1 1 1 6.67
New School 1 1 1 6.67
Pitt 1 1 1 6.67
South Carolina 1 1 1 6.67
CUNY 1 1 0 7.00
Erfurt 1 1 0 7.00
Florida State 1 1 0 7.00
Harvard 1 1 0 7.00
Iowa 1 1 0 7.00
LSE 1 1 0 7.00
Memphis 1 1 0 7.00
Minnesota 1 1 0 7.00
Purdue 1 1 0 7.00
Rice 1 1 0 7.00
Sheffield 1 1 0 7.00
South Florida 1 1 0 7.00
St. Andrews 1 1 0 7.00
Leeds 1 0 0 7.33
Maryland 1 0 0 7.33
UCL 1 0 0 7.33
Virginia 1 0 0 7.33

Anonymous said...

Here's the same data for 2011:

Rutgers 8 7 3 1.33
Wisconsin 7 5 3 2.00
UNC 7 5 2 2.33
Arizona 5 4 3 2.67
Oxford 5 4 2 3.00
Berkeley 4 3 4 3.00
Toronto 4 3 2 3.67
Michigan 4 1 3 4.00
NYU 4 2 2 4.00
Princeton 3 3 2 4.00
Yale 3 3 2 4.00
MIT 3 3 1 4.33
CUNY 3 2 2 4.33
Maryland 3 2 2 4.33
UCLA 3 2 1 4.67
Pitt 2 2 2 4.67
Columbia 2 2 1 5.00
Syracuse 2 2 0 5.33
Harvard 2 2 0 5.33
Stanford 2 2 0 5.33
UCSD 2 1 0 5.67
Texas 2 1 0 5.67
Boston Univ. 2 1 0 5.67
Rice 2 0 1 5.67
Nebraska 1 1 1 5.67
Cornell 1 1 1 5.67
Illinois 1 1 1 5.67
USC 1 1 1 5.67
Pitt, HPS 1 1 1 5.67
UCSB 1 1 1 5.67
St. Louis 1 1 1 5.67
Virginia 1 1 1 5.67
Stony Brook 1 1 1 5.67
New Mexico 1 1 1 5.67
Carnegie Mellon 1 1 0 6.00
Villanova 1 1 0 6.00
Duke 1 1 0 6.00
Albany 1 1 0 6.00
Minnesota 1 0 1 6.00
Western 1 1 0 6.00
Indiana 1 1 0 6.00
Beijing 1 1 0 6.00
UBC 1 0 1 6.00
Rochester 1 1 0 6.00
Vanderbilt 1 1 0 6.00
Connecticut 1 1 0 6.00
Colorado 1 1 0 6.00
Massachusetts 1 1 0 6.00
Lorenz Institute 1 0 0 6.33
Cambridge 1 0 0 6.33
Chicago 1 0 0 6.33
St. Andrews 1 0 0 6.33

Carolyn Dicey Jennings said...

Anon 1:05,

"Finally, those who acknowledge that women are underrepresented in the profession may consciously or unconsciously pay close attention to a file that is obviously that of a female, yet feel no compunction in tossing the file of a similarly well-qualified male."

That may be right, but it is good to remember that the proportion of women in NRC ranked PhD granting institutions is the same proportion of females that are ultimately hired. Thus, the fact that these women have fewer pubs may reflect the added difficulty of publishing as a woman, a fact that hiring committees may be sensitive to. In any case, the fact that they are compensating for that added difficulty does not seem to disadvantage male candidates, who are hired in the very same proportion that they make up PhD graduates.

Carolyn Dicey Jennings said...

Anon 4:43,

awesome! If I have it right, that is 67 "fresh" tenure-track hires, which is about half of the total number (at 142). So we should probably cut the above probabilities that I mentioned in half, to something like 27% for any position, 20% for a tenure-track position, and 6% for a ranked tenure-track position (if you are coming from an NRC ranked institution). I will put this into the next update!

Anonymous said...

I would just like to chime in that I applied for the Barnard job and I am not a job-seeker; I have a secure position and I am not otherwise on the market, but the desirability of the Barnard job prompted me to put my hat in the ring. Therefore I would not rely upon the number of applicants for that job as any kind of indicator of number of applicants.

p.s. I do agree that this is something that the APA should be doing, and it's shameful that they aren't taking responsibility for doing it.

Anonymous said...

"Thus, the fact that these women have fewer pubs may reflect the added difficulty of publishing as a woman, a fact that hiring committees may be sensitive to."

What is the evidence for the claim that it is more difficult to publish as a woman? Don't most journals use blind review?

Carolyn Dicey Jennings said...

http://lemmingsblog.blogspot.com/2007/04/apa-report-status-of-women-in.html?m=1

http://feministphilosophers.wordpress.com/2008/01/25/anonymous-refereeing-some-evidence/

http://feministphilosophers.wordpress.com/2009/10/07/women-in-philosophy-whats-getting-left-out/

Anonymous said...

Anon 7:30pm

If there are sub-fields in which most of the practitioners are women, and if journals know this, it would provide a method for effective discrimination that beats blind review.

IF there are sub-fields in which women tend to publish more, and if journals know this, that could also be used.

The idea that some people have (I don't know if it is true, but it sounds plausible) is that if by virtue of the content of a sub-field, or even by accident, it comes to be the case that women are prominent in some sub-field, that sub-field will come to be perceived as soft or not serious by men in the profession. They won't be conscious of evaluating the sub-field that way because of the connection to women, but the idea is they will do it nonetheless. Since it seems like some fields are dominated by women (feminism and various feminist approaches to traditional philosophical topics), and some fields have higher concentrations of women than others (ethics more than M&E), this tendency on the part of men in philosophy, if it exists, would have an effect on publication rates.

Anonymous said...

It may be harder to publish as a woman. But there isn't evidence about this on the table. I guess it may be *easier* to publish as a woman. So, when CDJ says

In any case, the fact that they are compensating for that added difficulty does not seem to disadvantage male candidates, who are hired in the very same proportion that they make up PhD graduates.

I think that's quite wrong. If men publish more but still get hired at the same rate, then that's evidence that they are disadvantaged.

Anonymous said...

Here's another hypothesis, to add to the stack:

Of women and men with equal philosophical acumen, men send out more papers, and thus get more published. Search committees for assistant professor jobs are judging, and more or less tracking, philosophical acumen, not total number of publications. (As was mentioned in previous discussions, beyond 1 or 2, publications don't add much to your applications).

This would explain the fact that women and men get hired in proportion to their representation in grad school, as well as the fact that men get hired with more publications.

Anonymous said...

to 9:04
Are you seriously suggesting that men are disadvantaged in the field of professional Philosophy?

Anonymous said...

There are a lot of things that can affect publication rates that aren't just straightforward discrimination by editors (though 8:24 does target an important problem for women - and, by association, men - working in certain areas). Feeling encouraged and like one's ideas are worth publishing can contribute greatly to publishing rates. It is often very hard to know oneself whether one's ideas are worthwhile, or just "obvious". I can really only speak from my own point of view on this, but this means I end up publishing only things that seem really clearly worthwhile to me (although I'm not a perfect judge of such things). Which means I pass up on publishing things that are probably publishable somewhere, which would up my publication rate...but that I don't think would make me a better candidate.

Anonymous said...

9:04 here again.

CDJ, your links at 8:17 don't reveal evidence that women have a harder time publishing, do they? One of them shows that women do publish less -- but precisely the point here is that women publish less and do just as well on the job market anyway.

8:24, I entirely agree. Now we just need to see whether this sort of (very plausible) effect is at work, and if so what it's strength is.

B said...

This is great work. But I am having a hard time taking it seriously because of the limited information available. For example, my graduate program is listed as having 3 TT hires this year, when the actual number is 6. (They haven't been reported on Leiter yet.) I can't help but think that this is the case with other departments. And when the data's wrong, the analysis of it is bound to be wrong. If there were a method of data collection such that we could guarantee we knew about every TT hire, then what Jennings has done here would deserve to be published by the APA every year. Hopefully just having this analysis of limited data will prompt some sort of official data collection method.

Anonymous said...

Just to remind everyone...

"Blind review" is a bit of a misnomer. It's true that the readers are generally blind, in that they don't know who has submitted the article they are reviewing. But the editorial staff and very often the editor knows the identity of the author's article before it's sent out to readers (unless when you submit your article, you don't include any identifying information...which would be idiotic).

The readers make their suggestions, but the editor has final say, and the editor very likely knows (or can easily find out) the identity of the author.

Carolyn Dicey Jennings said...

Anon 9:04,

The first of those links reports the distribution of female authors across different journals in philosophy. The second shows that when a non-philosophy journal implemented blind review it accepted more female authors. The third tells a story linking these that explains that:

" Some top philosophy journals do not make use of anonymous refereeing. Very very few make articles anonymous to the editor, even though the editor often rejects a huge proportion of papers without sending them to be refereed."

Moreover, the editor selects referees, which could have a big imact. I don't think it is epistemically responsible to say that all of that is just to say women publish less.

Anonymous said...

What in the hell is "philosophical acumen"?

I find the idea that some people are "better" at doing philosophy than others bizarre. Yes--some people are more prestigious, well-respected, influential, popular, etc., etc., than others, but that just means they have more academic capital than others--and all of this is relative to specific professional networks anyway. Someone who is considered a great philosopher in one professional network might be considered a complete dunce in another professional network, and vice versa.

The concept becomes even more unintelligible as we move backward in time. Did Quine have more or less "philosophical acumen" than Heidegger? Did Nietzsche have more or less "philosophical acumen" than Kant? Did Descartes have more or less "philosophical acumen" than Aquinas? One could go on like that all day...

If you find Philosopher X's ideas compelling and Philosopher Y's ideas wrongheaded, vague, ambiguous, mistaken, whatever, you may come to the conclusion that Philosopher X is a "better philosopher" than Philosopher Y. Fine. But let's not pretend it's necessarily because Philosopher X is objectively superior to Philosopher Y in virtue of some mysterious faculty called "philosophical acumen." I'm more inclined to think one prefers Philosopher X's work to Philosopher Y's work in the same way one prefers coffee to tea, or death metal to reggaeton, or whatever. Non disputandum de gustibus.

Anonymous said...

6:26 says

"I find the idea that some people are "better" at doing philosophy than others bizarre.
...
The concept becomes even more unintelligible as we move backward in time. Did Quine have more or less "philosophical acumen" than Heidegger? Did Nietzsche have more or less "philosophical acumen" than Kant? Did Descartes have more or less "philosophical acumen" than Aquinas? One could go on like that all day..."

This is just silly. Is Babe Ruth better than Willy Mays? Hard to say. But they are both in the hall of fame!! And it isn't just because of taste. Just because it is hard to come up with a set of objective measures that issue in clear and uncontroversial rankings doesn't mean that there is no such thing as doing well or poorly with respect to philosophizing.

Anonymous said...

6:26

Did Quine have more or less "philosophical acumen" than Heidegger?

less

Did Nietzsche have more or less "philosophical acumen" than Kant?

less

Did Descartes have more or less "philosophical acumen" than Aquinas?

less

One could go on like that all day...

ok!

Anonymous said...

Is there any actual evidence of discrimination against female authors on the part of editorial staff at philosophy journals? It's one thing to say that a review process isn't completely blind. It's quite another to present actual evidence of discrimination in the review process.

Mr. Zero said...

Is there any actual evidence of discrimination against female authors on the part of editorial staff at philosophy journals? It's one thing to say that a review process isn't completely blind. It's quite another to present actual evidence of discrimination in the review process.

Comments like this demonstrate ignorance about how unconscious, latent biases work. Non-blind review procedures could easily result in its being more difficult for women to publish, even in the absence of intentional discrimination. That's the point of blind review. We're not going to relitigate this issue.

Carolyn Dicey Jennings said...

6:26,

Putting aside the term, we could say that general philosophical ability is shared equally by the sexes, and thus take the fact that proportions of men and women are held up from phd to job as evidence that hiring committees are sensitive to whatever general ability we take to be equally distributed. Differential distribution of publications, then, would be taken as a sign that these do not perfectly match up to the general ability, rather than as evidence that one sex is inferior in general ability after all. I think this is the more reasonable move, given the fact that the system is not fully anonymized and we have ample evidence that women are discriminated against when recognized as such.

Carolyn Dicey Jennings said...

Anon 7:31,
there may not be evidence so direct as that, but this webpage gathers the objective evidence of bias in academia in general: http://homepages.inf.ed.ac.uk/perdita/GenderBias/

Continental Pissant said...

B wrote,

"If there were a method of data collection such that we could guarantee we knew about every TT hire, then what Jennings has done here would deserve to be published by the APA every year. Hopefully just having this analysis of limited data will prompt some sort of official data collection method."

Hear hear. Given that a vast majority of these positions are advertised in APA, it seems that APA ought to institute some sort of report-back method for their listings. This could be as simple as a departmental login to an online form, clicking on a 'yes' 'no' option for whether one's search was successful, and then entering the name of the department from which the candidate graduated. We're talking about a couple of dozen keystrokes, maximum.

Anonymous said...

7:31 here.

Thank you, Dicey Jennings, for answering my honest question--unlike Mr. Zero, who dismissed it without answering it, and then tried to silence any further discussion (how philosophical!!).

Since when did it become unphilosophical and improper (as you imply) to ask for evidence to back up a claim, Zero? If I'm ignorant, then that's all the more reason to put the evidence in front of me rather than dismiss my question.

Carolyn Dicey Jennings said...

B, Continental, etc:

I share your worry, but

1) It may not be that the unreported hires are structurally distinct from the reported hires (that is, so long as the sample is representative of the set, we can move from findings on the sample to conclusions on the set);

2) The first time I collected data I had information on only around 129 jobs and later added 60 more. I noticed that the major trends stayed the same, with only minor differences (but I only ranked up to worldwide Gourmet 49 in those early iterations). The comparative values are below, where "1" represents the earlier values and "2" represents the later values (neither set are from the current sheet, since they are from two earlier iterations):

overall gender 1) 32% f 2) 30% f
overall mean hiring rank 1) 22.10 2) 22.33
overall mean phd rank 1) 18.02 2) 17.64
overall mean specialization contribution rank 1) 5.13 2) 4.99

Carolyn Dicey Jennings said...

addendum:

Correlation between (Gourmet) hiring rank and phd rank 1) .47 2) .43

(and for the worldwide top 50 Gourmet it stays at .43 in the current iteration, but it is .10 for the top 50 NCR R and .08 for the top 50 NCR S)

Anonymous said...

I'm not all that surprised that a relatively mild correlation leads people to rush to worry that the men are being discriminated against, but the data are compatible with a number of things:

1) Women have fewer publications of higher quality. (Perhaps the tendency for women to be self-effacing means what they do send out is higher quality.)
2) Women who get jobs are disproportionately represented in non-prestigious teaching schools where publication is not as much of a priority (and may be perceived as a flight risk).
3) There are other reasons to hire someone besides research output and promise.
4) Hey, we're mostly guessing at the number of people on the market and working with a limited dataset so who knows yet?

None of this suggests that we need to go as far as committees discounting men's publication records to favor women.

Mr. Zero said...

Mr. Zero, who dismissed it without answering it, and then tried to silence any further discussion (how philosophical!!).

I didn't attempt to silence further discussion. What I said was, further discussion of these issues would have to reflect a basic understanding of the issues involved. I don't think it's unphilosophical to expect a person who advances a position in a debate, as you have (e.g. when you say, "It's one thing to say that a review process isn't completely blind. It's quite another to present actual evidence of discrimination in the review process."), to have a basic familiarity with the topic (e.g. to know what the purpose of blind review is and why it's important).

If I'm ignorant, then that's all the more reason to put the evidence in front of me rather than dismiss my question.

Not necessarily. It depends on whether you did your homework. If you're ignorant because you didn't do your homework, you shouldn't expect other people to do it for you. If you don't understand why unblind review procedures would make it harder for women to publish, even if there was no "actual" overt discrimination going on, you haven't done your homework.

Anonymous said...

even if there was no "actual" overt discrimination going on

S/he didn't refer to '"actual" overt discrimination', but used the word 'actual' in the context of 'actual evidence'. As we can all see.

Mr. Zero said...

S/he didn't refer to '"actual" overt discrimination', but used the word 'actual' in the context of 'actual evidence'. As we can all see.

Ok. I guess I goofed the quotes. But how does that affect the overall point? It's not as though there's no actual evidence: the evidence of non-blind review procedures is it.

Anonymous said...

I have recently discovered that not all journals use "double-blind" review; some allow the referee to know the author's name, while the author does not know the identity of the referee. I only discovered this by a odd series of circumstances that led me to inquire of a journal I was refereeing for. It made me wonder about others. Many do not post online their policies regarding the process, and some who do post, do not specify whether the review is single-blind or double-blind. For this and other reasons, I think it is quite possible that women authors are subjected to bias, implicit or overt, in the publication process. Cheerio.

Anonymous said...

Anon 7:31 here.

Mr. Zero, my question was not whether it is possible to provide some sort of explanation as to how nonblind review could end up leading to a sexist review process. It's quite obvious that there are all sorts of ways in which it could lead to this.

My question was whether we had any actual evidence that it IS making the review process sexist. These are very different questions, and my comment was absolutely clear as to which question I was asking.

Everything you said addresses the first question rather than the second. Dicey Jennings provided some data that helped to answer the second question. This data cannot be expected to be common knowledge among graduate students in philosophy. It's not like we learn this stuff in our comps or distribution requirements.

Anonymous said...

I agree with several others. 800 is way too low of a number. I never applied for the Barnard job, nor did anyone on the market I know.

If the actual candidate pool is two times larger, that makes all your calculations twice as dismal...


Sigh.

Anonymous said...

"I had thought, for example, that if you had one publication you were pretty much ok, whereas someone else with much more experience said three was necessary. It seems that I was wrong and he was right, and I find that to be of use."

Christ. I have four publications and I didn't even get an interview.

Anonymous said...

"I find the idea that some people are "better" at doing philosophy than others bizarre."

Totally bizarre. Clearly everyone is perfectly equal and no one is better than anyone at anything.

To suggest otherwise is just offensive!

Anonymous said...

"This data cannot be expected to be common knowledge among graduate students in philosophy. It's not like we learn this stuff in our comps or distribution requirements."

You also don't learn how to tie your shoes there. Does Mr. Zero not get to presume such knowledge?

Zero said that your comment expressed a lack of familiarity with the issue. He didn't say that you were guaranteed to acquire such familiarity in the course of taking graduate philosophy courses. There are things you learn being a reflective, serious adult living in the world as it is. If you aren't that then your comments deserve to be silenced. Stop whining.

kg said...

I really appreciate the fact that this blog exists, as I have learned so much about the state of my profession from both the posters and the commenters. However, I am often blown away by how rude and hostile the comments are, and 4/21/4:18pm is an example of the kind of unnecessary ad hominem sniping that makes me not want to read this blog anymore.
--Karen Gover

Anonymous said...

4:18 here. Ad hominem gets such a bad rap and I don't know why. I get why substituting an ad hominem for an argument is bad, at least when an argument is called for. If a position meets some basic minimum criteria, then it deserves to be opposed with reasons.

Some positions though clearly don't. So I take it that if someone came on to this blog and said 'Women are less able to think abstractly than men because their brains are hardwired only for the concrete tasks of child-rearing', no one would have a problem with responding to that with something other than an argument. If I were to call the person out with much stronger language than I used above this would not only be unproblematic, it might be called for. After all I only implied that the person was not a serious, reflective adult. Bigotry calls for much stronger invective.

I know some people think differently. They think that whenever reasons and evidence exist to respond to a position those reasons and evidence should be brought up in a clear, calm respectful way. That is silly. That would be appropriate if mistakes like the bigotry I described just now were the result of simple ignorance or cognitive oversights on the part of a well meaning person attempting to find the truth.

Some positions are such that well meaning people attempting to find the truth are extremely unlikely to hold them. You won't change the minds of people who hold them by adverting to evidence and logic, because they don't care about their beliefs being true in the first place. They probably don't consciously represent their own goals as being different than the truth, but that is mostly because they do not reflect on their own cognitive goals and strategies in the first place. That is the kind of thing you do if you care about the truth. And there are people who don't. Truth is the norm of assertion and thought for reasonable people. Not all people are reasonable in this sense.

So sometimes the thing to do is to respond to a person not in a way that will give them sufficient evidence to change their mind, but that will give them sufficient motivation to start caring about being correct. You do things to make the person take the question more seriously, or to actually care what the right answer is. I think shaming is a good way to do that when the mistakes involved are minor.

The person I was so implicitly rude to struck me as a person who was not thinking about sexism in a serious way. But he didn't seem so bad. He didn't seem like a bigot or a generally unreasonable person. People like that sometimes respond well to having other people in their community respond with frustration and anger at their opinions. It provides motivation to do better.

And I figured that what this person needed was motivation. Paying attention to the news, to history, and sometimes just to what is going on around you will be sufficient to make it clear what a pervasive social problem sexism is. Paying attention and reflecting would make it clear that when you have a group dominated by men (especially older men), and there is a way for that group to systematically discriminate, than the reasonable default position is that such discrimination is happening, at least sometimes.

Now as a matter of fact I also thought the complaint he was making against Zero was ridiculous. The implicit standard he was invoking was, 'if I wasn't taught X in grad school it is unfair to expect me to know X'. That is really silly. So silly that I first tried to find a different way to read his post. But further reading did not reveal, to me, any interpretation that looked better.

Finally, to call someone unserious or unreflective, and to accuse him of whining is not really that harsh. Certainly you could find better examples of ad hominems on this site. I would suggest Dr. Killjoy's from the thread just previous. It was really mean. It was also entirely appropriate.

Anonymous said...

4:18 here again. While I agree with what I said earlier (I normally do), in hindsight I see that I am guilty of major thread derail/hijack. I apologize. Not every thread is a place to have a conversation about the norms for internet discourse.

I would like to thank Dicey Jennings for her work and echo the sentiment that the APA ought to do it every year.

Carolyn Dicey Jennings said...

4:18 (and others),

Thanks! I collected the data for myself, but I did try to make it clearer so that other people could understand it and make use of it, and am thus glad to hear that it was helpful.

Also, for what it's worth, I agree with the general sentiment that we should be responsible knowers and should sometimes call out those who are irresponsible knowers, but I also think that this should be kept in tension with humility and recognition for how difficult it is to know and how much we rely on others to help us come to knowledge. I think it would be fair to say that the number of intelligent and thoughtful people who do not find implicit bias to be obvious is an indication that the line should be moved toward humility in this case.

Anonymous said...

This is 7:31.

"After all I only implied that the person was not a serious, reflective adult."

So, anyone who is unfamiliar with the actual data on implicit bias is not a serious, reflective adult? Or perhaps you meant to say that any philosopher who is unfamiliar with implicit bias is not a serious, reflective adult. Either one is a remarkably strong claim.

"That [those reasons and evidence should be brought up in a clear, calm respectful way] would be appropriate if mistakes like the bigotry I described just now were the result of simple ignorance or cognitive oversights on the part of a well meaning person attempting to find the truth."

Yeah...in other words, reasons, evidence, clarity, and respect are appropriate in this particular case. I have no idea where you got the idea that I was not a well meaning person attempting to find the truth.

"So sometimes the thing to do is to respond to a person not in a way that will give them sufficient evidence to change their mind, but that will give them sufficient motivation to start caring about being correct. You do things to make the person take the question more seriously, or to actually care what the right answer is. I think shaming is a good way to do that when the mistakes involved are minor."

I am wondering what evidence you could POSSIBLY have for thinking that I do not care about reaching the truth in this particular case, or that I am not taking the issue seriously.

"The person I was so implicitly rude to struck me as a person who was not thinking about sexism in a serious way."

On what basis? That I asked for evidence to substantiate a claim? Also, your rudeness was quite explicit, not implicit.

"But he didn't seem so bad. He didn't seem like a bigot or a generally unreasonable person."

In other words, he is precisely the sort of person who would respond to reason and evidence, and for whom ad hominem and straw man are inappropriate tactics.

"And I figured that what this person needed was motivation. Paying attention to the news, to history, and sometimes just to what is going on around you will be sufficient to make it clear what a pervasive social problem sexism is."

I don't know where/how you made the inference that I don't think sexism is a pervasive social problem, or that I needed motivation to take the issue seriously.

"Paying attention and reflecting would make it clear that when you have a group dominated by men (especially older men), and there is a way for that group to systematically discriminate, than the reasonable default position is that such discrimination is happening, at least sometimes."

Of course discrimination happens sometimes. This by itself is not enough to generate the conclusion that it happens in the particular case of blind review. All I asked for was evidence to substantiate this claim.

"Now as a matter of fact I also thought the complaint he was making against Zero was ridiculous. The implicit standard he was invoking was, 'if I wasn't taught X in grad school it is unfair to expect me to know X'."

If you willfully read my view REALLY, REALLY, uncharitably, you might get that idea. Alternatively, you could interpret my claim in the following way: the fact that we aren't taught x in grad school is one piece of evidence in favor of thinking that x should not be expected to be common knowledge among graduate students.

"Finally, to call someone unserious or unreflective, and to accuse him of whining is not really that harsh."

Um...yeah, it REALLY is...unless you're such an asshole that you routinely call people much worse things.

Bye the way, this all started because I politely asked that someone provide evidence to substantiate a claim.

Anonymous said...

Since nothing else seems to be happening on this thread, it can't really be derailed can it?

"Alternatively, you could interpret my claim in the following way: the fact that we aren't taught x in grad school is one piece of evidence in favor of thinking that x should not be expected to be common knowledge among graduate students."

I am not sure what makes you think that this means something different from:

"if I wasn't taught X in grad school it is unfair to expect me to know X"

when everyone knows that you are, or have been, a graduate student. Notice that its 'me' and not a variable. Now you are right that I did not read you to be making a claim about what is common knowledge, but rather to be making a claim about what it is fair to expect you to know. Obviously it could be fair to expect someone to know something that most people do not, if most people are culpably ignorant of some significant fact. But I don't think that this makes a difference to how charitable my reading was. If you think the reading you suggested makes a reasonable point, then I disagree

You didn't seem to understand the structure of my argument. That is understandable because I didn't revise or even re-read it. Mistakes happen that way. So let me lay it out.

(A)Some people think ad hominems are always a bad thing to use.
(B)But that is clearly wrong. Consider the case of the bigot.
(C)In such cases ad hominems are appropriate because it can potentially provide a motivation to improve (though I wasn't clear on this point. the odds are the bigot is not going to change much at all. The motivational benefit is supposed to come from people seeing the bigot shamed. I bring this up because there is an important dis-analogy between the bigot and you on just this point)
(D)Granted that ad hominems can be ok if they are used to provide motivation for someone to improve their approach to an issue, it could be legitimate to use them in other cases where the problem is motivation to take a good cognitive approach to an issue.
(E)You are such a case.

So at no point was I claiming you were a bigot. I would understand your confusion on this point if only I had not said explicitly that you didn't seem like a bigot to me. The discussion of the bigot was meant to introduce a principle that would then be extended to cover your case. Sometimes to discover something interesting about X it helps to think about Y, which is similar in some, but not all ways.

"I don't know where/how you made the inference that I don't think sexism is a pervasive social problem, or that I needed motivation to take the issue seriously."

Good question. Here is the answer:

"Of course discrimination happens sometimes. This by itself is not enough to generate the conclusion that it happens in the particular case of blind review. All I asked for was evidence to substantiate this claim."

You might object that you only posted this after I was mean to you. That is true, but you posted much the same thing earlier and I am too lazy to scroll up to find it. So lets review: 'X happens sometimes' and 'X is pervasive' don't mean the same thing. In addition, the implicature of the first cuts against the second. 'Sexism happens sometimes' is often followed up with 'but it is a significant over-reaction to....' where the ellipses are filled in with some proposal to help ameliorate the effects of sexism. Whether you know it or not you are repeatedly talking about this issue in a way that suggests that you don't think it is a problem. Now since I said ‘pervasive’, lets ask whether it follows that sexual discrimination occurs in blind review? Of course it doesn’t. Should all our inferences be deductive? Of course not. Would pervasive sexism give you reason to think it likely that sexual discrimination occurs in blind review? Some. I think good enough to shift the burden of proof.

Anonymous said...

Also, one of these words is not like the others: “unserious”, “unreflective”, “asshole”. Guess which one!

Clue: 'Middle School'

uturn said...

"Without that, all we have is "anon 4:20 claims ND has 3 TTs," which is hard to base much on."

Not anymore! (ND has posted results to the Leiter hiring thread)

Carolyn Dicey Jennings said...

uturn:

cool, thanks. I will update the spreadsheet at some point.

Anonymous said...

7:31, one last time.

I lack the patience to respond yet again point-by-point to the continued and baseless attacks on me. I will merely respond to one point:

"(D)Granted that ad hominems can be ok if they are used to provide motivation for someone to improve their approach to an issue, it could be legitimate to use them in other cases where the problem is motivation to take a good cognitive approach to an issue.
(E)You are such a case."

I don't know what you mean by "a good cognitive approach." I politely requested evidence and then explicitly thanked the person who provided me with it. I didn't press the matter further. (All I have done since is fend off attacks.) In general, I don't uncritically accept every claim I come across. If I don't know the reasons or evidence behind it, I will ask for them.

The "asshole" comment was unnecessary and I retract it.

When the intellectual climate on an issue has reached the point where one cannot make a polite request for evidence to sustantiate a claim without being subject to rude and persistent ad hominem remarks and straw man arguments, something has gone amiss.

That is the last thing I will say.

Mr. Zero said...

Hi 7:31,

I did not interpret your remark as an honest question. I interpreted it as moderately hostile, though I wasn't sure if it was because you're some kind of jerk or if you just weren't familiar with the data. If I was wrong about that hostility, and it seems that I was, I apologize.

However--and I realize it's lousy to apologize and then say, 'however'--the evidence you were asking for had already been provided: somebody asked for evidence that it's harder for women to get published, and somebody else provided evidence that peer-review is often not anonymous. The reason you didn't recognize that this was the asked-for evidence was that you didn't know enough about the topic to know what you were looking at. And the information that would have helped you connect the dots is out there and relatively easy to come by, and it generally comes up whenever this topic arises on this blog. But again--and this is my bad--I was too grouchy about it. I could have been much nicer. Sorry.

And I also think that if anyone is responsible for derailing this thread, it's me. Sorry.

R. Kevin Hill said...

I've said this before, perhaps somewhere else on this blog: many hiring departments use the method of bring four people out to campus to give a talk and perhaps teach a class, and then rank order the candidates. If you apply "objective" criteria like department of origin ranking, and pick your four favorite candidates, and you are a low ranked institution, then you run the risk of having all four of your finalists taken from you by better departments of hire. If you fail to hire anyone, you risk losing the line. Consequently, over time, low-ranked departments learn that it is unwise to let prestige be your guide; you will be its victim. It takes several searches to learn this, but once it is learned it does influence which candidates are chosen. We routinely now do not invite out our favorite candidates; it is a demonstrable waste of time, and a serious risk. Before we even compile our APA interview list, I now make a rough comparison of our applicants to competing jobs, pairing them off best to best, second best to second best, etc. until we reach the line below which we find our "gettable" candidates. This does not prevent us from momentary delusions, but deep down we know we will hire from below that line, not by preference but by necessity.

R. Kevin Hill said...

[Just lost a long comment to "Blogger", so a briefer one must suffice]

When we are hiring, I rank order the candidates who apply by department of origin, then rank order jobs in the JFP in our AOS, pair the lists off and largely ignore any applicant who pairs off with a position higher ranked than ours. To do otherwise risks wasting recruitment funds on four on-campus interviews of people who will not accept our offer, and worse, risk losing the line by failing to fill it. Whenever we have invited someone out who fell above that line and made them an offer, they turned us down for someone better (three different searches so far have gone that way). If you're at a top program and do not get interviews, do not assume that it must be even worse farther down the foodchain; if you are not near the top, don't despair, as some hiring departments are focused on you.

Anonymous said...

R Kevin Hill. Interesting. I'm wondering how you look at candidates who come from an unranked department. I come from a good department, but it's in continental Europe, so not on Leiter's list by definition (only Anglophone departments are listed). I have published well and have good teaching evals. But I'm afraid many American universities won't look at my file because I'm European (and not British).
As a matter of fact, I got two first-round interviews from American departments where I would have loved to work. Throughout the interviews, I noticed the SC were concerned that I would not want to stay in their small-town university, that I'm too European and cosmopolitan, and that I would return, culture starved, to Europe. I failed to convince them that I just want a TT position (they hardly exist in continental Europe, most positions are full professor level, so people are in postdocs for sometimes decades).
I'm wondering if there is anything I can do to mitigate possible prejudices when applying to the American market.

Anonymous said...

Kevin,

While I fully understand your methodology and why it is pragmatically necessary, I can't help but feel sore about being disqualified for psychological presumptions that, at least in my case, are false. After a couple of years on the market, several interviews, and still no job, I would feel just shy of moral outrage if I found out that I wasn't offered a flyout or job on the basis of these kinds of (in my case false) psychological assumptions.

Again, I'm not really targeting you as I understand the danger of losing the line. The fact that your reasoning makes pragmatic sense is what I find infuriating. My anger is targeted not at you but at the market.

Anonymous said...

The practice that R Kevin Hill describes is necessary (and I suspect common) because there is little to no coordination among departments.

The two best ways to improve things for professional philosophers and for the philosophy discipline are: (1) Decrease # of job-seekers so that it is closer to the # of jobs by cutting the # of grad students admitted to PhD programs; (2) ensure that all or almost all of the best candidates find jobs. (1) and (2) are virtually impossible unless and until departments become better coordinated. This is a collective action problem.

Carolyn Dicey Jennings said...

On gender and publications, it might be worthwhile considering this finding (posted here: http://prophilosophy.wordpress.com/2012/07/13/tenure-track-hiring-analysis-stage-three/#comments): "Noting that a smaller proportion of women receive postdoctoral positions and the correlation between priors and number of publications, I looked at the proportion of women with prior positions as compared to men with prior positions, and it was 30% and 57%, respectively, for tenure-track jobs. I hadn’t thought of this before, but this might explain the difference in average number of publications by itself. If that is right, we might need to be more careful as a field in distributing postdoctoral positions. It seems to me that these are sometimes awarded less democratically, which may disadvantage women and other minorities in the field."