Leiter Reports: A Philosophy Blog

News and views about philosophy, the academic profession, academic freedom, intellectual culture, and other topics. The world’s most popular philosophy blog, since 2003.

  1. Fool's avatar
  2. Santa Monica's avatar
  3. Charles Bakker's avatar
  4. Matty Silverstein's avatar
  5. Jason's avatar
  6. Nathan Meyvis's avatar
  7. Stefan Sciaraffa's avatar

    The McMaster Department of Philosophy has now put together the following notice commemorating Barry: Barry Allen: A Philosophical Life Barry…

A reliable guide to legal norms regarding academic hiring in the US?

Brian Leiter

Philosopher David Wallace (USC, previously Oxford) writes:

I was reminded to ask this by your blog post on the matter: is there any reasonably straightforward way to get a reliable guide as to what US (and, I guess, state) law permits and mandates on equal opportunities? It’s one thing I’ve found very weird coming out here: in the UK (well, in England, more exactly) I had a very clear idea (partly because university HR took it seriously and provided resources) just what the law permitted, and I was confident we were (basically) operating within it. In the US I have a much less clear idea, and resources seem harder to come by.

This is a very good question:  do any readers have a link to a *reliable* resource on this issue?  While "hidden criteria" searches, as I discussed them in my CHE piece, are illegal, my impression is that preferences for women, African-Americans, Hispanics, and some other demographic minorities are permissible.  (Much of Europe has a clearer demarcation:  "positive discrimination" is permitted in favor of underrepresented demographic groups when candidates are of otherwise equal qualifications.  This is, correctly, called "positive" discrimination because it does not involve demeaning the disfavored group.  But it also has the virtue of acknowledging that the candidate is being chosen not because of the merits related to the job, but because of the societal demographic considerations.  Although this is not the de jure rule in the U.S., it is often the de facto one.) 

Leave a Reply to Allen Partridge Cancel reply

Your email address will not be published. Required fields are marked *

14 responses to “A reliable guide to legal norms regarding academic hiring in the US?”

  1. Allen Partridge

    Affirmative action on the basis of race, sex, and so on is illegal at public institutions in the following states: California, Washington, Florida, Michigan, Nebraska, Arizona, New Hampshire and Oklahoma.* This applies to the following schools (organized according to the 2014 Gourmet Report):

    4 – University of Michigan, Ann Arbor
    10 – University of California, Berkeley
    10 – University of California, Los Angeles
    13 – University of Arizona
    23 – University of California, San Diego
    24 – University of California, Irvine
    28 – University of California, Riverside
    42 – University of California, Davis
    42 – University of California, Santa Barbara
    45 – Florida State University
    Unranked – Arizona State University
    Unranked – University of Washington, Seattle**

    Philosophy departments at several of these institutions are almost certainly not complying with the law. This is easiest to demonstrate for graduate admissions: take a look at the names or photographs of graduate students on the department website, and calculate what percentage are female. Since we know that women comprise about 30% of philosophy majors, if the proportion of graduate students listed on the department website who are female is significantly higher than this, it is likely that the department is favoring female applicants to its PhD program, in violation of state law. And if a department ignores the law when it comes to graduate admissions, it stands to reason that they will ignore it in hiring as well.

    There is an interesting ethical question here, namely, under what circumstances do we have a moral obligation to abide by laws we take to be unjust?

    *See http://www.pewresearch.org/fact-tank/2014/04/22/supreme-court-says-states-can-ban-affirmative-action-8-already-have/. Various other sources confirm that this information is accurate.

    **Note that the University of Southern California and the University of Miami are private institutions and hence exempt.

    Reply
  2. Anon

    Allen Partridge writes: "Philosophy departments at several of these institutions are almost certainly not complying with the law. This is easiest to demonstrate for graduate admissions: take a look at the names or photographs of graduate students on the department website, and calculate what percentage are female. Since we know that women comprise about 30% of philosophy majors, if the proportion of graduate students listed on the department website who are female is significantly higher than this, it is likely that the department is favoring female applicants to its PhD program, in violation of state law."
    No demonstration of this is to be found by these means! Partridge's comment is ridiculous. There are many other explanations that are compatible with the data. And don't say that this is an inference to the best explanation, unless you are similarly willing to accept as an inference to the best explanation that it is very likely that many departments use straightforwardly racist hiring policies in order to explain why there are so few faculty members of color in the profession. For the inference in this case is very nearly equally justified.

    Reply
  3. Allen Partridge

    "And don't say that this is an inference to the best explanation, unless you are similarly willing to accept as an inference to the best explanation that it is very likely that many departments use straightforwardly racist hiring policies in order to explain why there are so few faculty members of color in the profession. For the inference in this case is very nearly equally justified."

    Your analogy is faulty. I was suggesting we compare the proportion of women at successive stages of the pipeline — % of philosophy majors versus % of philosophy PhD students at select institutions. You are comparing people of color at opposite ends of the pipeline, % of the US population versus % of philosophy faculty.

    If we set up the analogy correctly, it becomes clear that an inference to racial bias would, in fact, be appropriate. If, for instance, under-represented minorities earned 30% of philosophy PhDs — in line with their share of the US population — but then made up only (say) 20% of junior hires, this would be solid evidence of pervasive racial discrimination in hiring in our discipline. There would be calls for investigations and lawsuits, and rightfully so. That isn't what's happening, though. Only 8% of philosophy PhDs are awarded to historically under-represented minorities*, which means that black and hispanic philosophy students are leaving the pipeline in huge numbers well before they get to the hiring stage. This is still, in my view, a major problem for the discipline, but the culprit probably isn't going to be anything as simple as search committees discriminating against minority applicants.

    In any case, your analogy, once corrected, supports the original point: if significantly more than 30% of a program's PhD students are female, we should take this as evidence that the department is practicing affirmative action on the basis of sex, just as we would take a large disparity between the share of PhDs awarded to minorities and the share of minorities entering tenure-track jobs to be evidence of racial discrimination.

    *See https://www.humanitiesindicators.org/content/indicatordoc.aspx?i=266.

    Reply
  4. Anon

    What I wrote may be incorrect. But if it is, you certainly haven't explained why.
    You've misunderstood the analogy. Because of this, you *think* what I wrote is incorrect. You compared the percentage of women grads *within a department* with the percentage of women in the profession, and your explanation of the disparity is gender discrimination *within the department, in violation of the law*. Analogously, you might compare the percentage of faculty members of color *within a department* with the percentage of PhDs of color in the profession–notice, many, many departments have 0% faculty members of color–and your explanation of the disparity is racial discrimination *within a department, in violation of the law*.
    The inference in either case is ridiculous. Neither is an inference to the best explanation.

    Reply
  5. Janet Broughton

    The Office of General Counsel for the University of California has provided written guidelines on this general subject, taking into account federal law, California law, and university policies. This link should take you to a PDF of the guidelines:

    http://www.ucop.edu/general-counsel/_files/guidelines-equity.pdf

    Reply
  6. Allen Partridge

    Hold up: you seem to be claiming that it is never possible to infer that a department is practicing affirmative action on the basis of its student demographics. Doesn't that strike you as a little crazy? Suppose that every graduate student in one department were a gay Fijian. Do you really think this would not give us any reason to believe that the department was favoring gay Fijians in admissions?

    "Analogously, you might compare the percentage of faculty members of color *within a department* with the percentage of PhDs of color in the profession–notice, many, many departments have 0% faculty members of color–and your explanation of the disparity is racial discrimination *within a department, in violation of the law*."

    Your revised analogy is also flawed. For 2016, around 5% of the APA's regular members were black or hispanic.* With 15-30 philosophy faculty per school, this means that we should expect to see, in the absence of racial bias, between 0.75 and 1.5 under-represented minorities in each department. Since the fewest minorities it is possible for each institution to have on its faculty is zero, this means that we can never be confident when looking at a single school in isolation that minority under-representation isn't due to chance alone. You are comparing a situation in which it is possible to get a significant result to one where it is not.

    *See http://www.apaonline.org/page/demographics.

    Reply
  7. Anon

    None of what you've written speaks to my point. In fact, this latest comment of yours has completely missed the point. My claim is simply about the claim you make about all of the departments that you're original comment mentions. Your claim is that each of those departments is likely not complying with certain laws if *certain facts* support there being a non-trivial disparity between the percentage of women in the profession and the percentage of women grads in the relevant department. But it is absurd to infer from *these facts* supporting the disparity to the high probability that the department is not complying with certain laws. It is as preposterous as a similar inference from very similar facts supporting a non-trivial disparity between the percentage of PhDs of color in the profession and the percentage of faculty members of color in the relevant department to the high probability that the department is not complying with certain laws.

    Reply
  8. harry b

    Allen

    Its nearly a decade since I analysed my own department, but, whereas the proportion of female grad students in the department was at that time slightly higher than the proportion applying, they had, on average, 1.4 refereed publications for every 1 publication that a male graduate student had (they had at that time 1.8 conference presentations for every 1 male conference presentation). This suggested to me that either we were better at teaching female students, or were in some way discriminating against them in admissions, despite the fact that we were enrolling them in slightly greater proportions than they were applying. I did not believe the former.

    What you need to know, for the departments you mention, is whether the number of women enrolled is proportional to the number of high quality female applicants.

    Final comment. About 30% of philosophy majors are women. It would be fallacious to assume that, among the most talented and skilled 20% of majors only 30% are women. That has not been my experience at all. But others may testify otherwise.

    Reply
  9. Allen Partridge

    Anon:

    "In fact, this latest comment of yours has completely missed the point."

    Your point is clear enough. The problem is that it's wrong. The strongest conceivable case you could make that a department is discriminating against black and hispanic applicants in hiring would be if a department had 40 members, none of whom were under-represented minorities. But this isn't a statistically significant difference from the expected number of under-represented minorities, which is 2, given the APA figures. A chi-square test comparing an observed frequency of 0/40 to an expected frequency of 2/30 falls well short of significance, p = 0.15. So, because only 5% of the profession is black or hispanic to begin with, and because a single department is a small sample size, it's impossible to ever be confident just by looking at one department that minority under-representation isn't due to chance alone.

    This is not true of graduate students and sex. Suppose that a department has 40 graduate students, 20 of whom are women. Because around 30% of philosophy majors are women, the expected number of female graduate students in a department this size is 12. A chi-square test comparing an observed frequency of 20/40 to an expected frequency of 12/40 is statistically significant, p = 0.001.

    So, as I said, you are comparing a situation in which it is impossible to get a significant result to a situation where it is possible to get a significant result. This is a poor analogy. The reason why we are unwilling to infer that a department with zero under-represented minorities on its faculty is racially biased is that we can't be confident that this distribution didn't come about by chance alone. But some departments have a M:F ratio among graduate students which almost certainly didn't come about by chance alone. This is why we can get evidence that a department is practicing affirmative action by looking at its graduate student demographics, but can't get evidence that a department is discriminating against black and hispanic job candidates by looking at its faculty demographics.

    harry b:

    "Its nearly a decade since I analysed my own department, but, whereas the proportion of female grad students in the department was at that time slightly higher than the proportion applying, they had, on average, 1.4 refereed publications for every 1 publication that a male graduate student had (they had at that time 1.8 conference presentations for every 1 male conference presentation)."

    The reverse appears to be true today, in general.* Carolyn Dicey Jennings's placement data suggests that men on the job market have about 2.2 publications, on average, while the average for women is around 1.1.

    *See https://bparsia.wordpress.com/2014/12/29/some-philosophy-hiring-data-analysis/.

    Reply
  10. Allen Partridge

    Correction:

    "A chi-square test comparing an observed frequency of 0/40 to an expected frequency of 2/30 falls well short of significance, p = 0.15."

    Should be:

    "A chi-square test comparing an observed frequency of 0/40 to an expected frequency of 2/40 falls well short of significance, p = 0.15."

    Here's another way of looking at it. I am claiming that having a six-sided die land on five or six 20 times out of 40 is evidence that the die is biased. Your response, Anon, is that this is bollocks, because we would never take a twenty-sided die landing on twenty 0 times out of 40 to be evidence that the twenty-sided die is biased. Do you see why the analogy doesn't work when put into these terms?

    Reply
  11. Anon

    The inference in the die example is fine; the best explanation of the evidence is that the die is biased, because there are no other available explanations of the evidence that are as good as bias. The same cannot be said for the examples of illegal departmental discrimination. In these cases, there are other explanations that are at least as good as illegal discrimination. This is obviously true (to everyone other than you, I guess), but Janet Broughton's link provides evidence of this.

    Contrary to everyone of your comments, my analogy is good precisely because, in each of the cases, there are other available explanations that are as good as (if not better than) illegal discrimination.

    Reply
  12. Allen Partridge

    "Contrary to everyone of your comments, my analogy is good precisely because, in each of the cases, there are other available explanations that are as good as (if not better than) illegal discrimination."

    My, you sure are determined to stay on board that sinking ship, aren't you? Here was the original structure of your analogy: inferring that racial discrimination has taken place from a department having no under-represented minorities on the faculty would be ridiculous, therefore, by parity of reasoning, inferring that affirmative action has taken place from a department having a surfeit of female graduate students is also ridiculous. But, for the reasons given above, this doesn't actually follow — it is indeed ridiculous to infer racial discrimination from a department having zero under-represented minorities on the faculty, but this isn't because there are "other available explanations of the evidence that are as good as bias", it's because the department may well have ended up with zero minority faculty members purely by chance.

    So, once again, let's correct the defect in your analogy. Suppose that a department has 20 faculty members, none of whom are women. Would we take this as evidence that the department is discriminating against women in hiring? It seems to me that the answer is clearly yes. At the very least, it would raise big red flags, and it would be incumbent upon the department to explain how they managed to end up with such an unlikely sex distribution without actively discriminating against women. And, given that we would be willing to infer bias from a 20:0 M:F ratio among faculty, I see no reason that we should be unwilling to infer that affirmative action is taking place from a 1:1 M:F ratio among graduate students. Could there conceivably be, in either case, a totally innocuous explanation for why the observed gender balance differs significantly from the expected gender balance? Sure. But this is just to say that the evidence is not *absolutely decisive*, which no one has ever denied.

    Reply
  13. Anon

    My original complaint–in comment 2–and every complaint afterward appealed to inference to the best explanation. Either you struggle with reading comprehension, or you don't understand inference to the best explanation. Given your comments, I'd guess both are true. Ask a lower-division undergrad in an Intro to Phil class for help understanding it. If they are anything like my intro students, they will be capable of explaining it to you.

    Reply
  14. Innf

    Wow, anon really has no understanding of statistics…at all.

    —–
    KEYWORDS:
    Primary Blog

    Reply

Designed with WordPress