Teaching-Focused Philosophy PhD Programs
Which philosophy PhD programs focus on training students to teach and getting them placed into permanent teaching-oriented jobs (with some success)?
Philippe Baudelocque, “Universe”
A philosophy professor wrote in saying that there are regularly students in his MA program who want to pursue a teaching “track” but that they’ve been unable to get a good sense of whether any PhD programs are particularly oriented towards that, or have options to pursue something like that, and if so which ones.
If you’re aware of departments that are particularly teaching-focused, or have a teaching-focused option, let us know about them.
Related: “An Approach to Teacher Training in Philosophy Departments“, “How to Teach (Philosophy): Readings Sought“, “Diverse Teaching Experiences and the Philosophy Job Market“, “Preparing Teaching Assistants“
I am one of those students. I would like to know if there are any such programs, particularly because I study philosophy of education.
Surely one part of what it is for a PhD program to be teaching-focused is for it to be a program where graduate students get a fair amount of experience in teaching a variety of courses, with sufficient preparation and constructive feedback on the quality of their work.
We do this at Florida State, where graduate students are encouraged to teach their own courses during the academic year (just one per semester) and summer once they have received their MA. It comes at a cost, of course, namely that teaching takes work! And we do encourage students to seek opportunities for fellowships that will free them up for research, with the understanding that there is a cost to these as well, namely having less teaching experience to advertise when they are on the job market and build on early in their careers. (I was a *disastrous* teacher of undergraduates during my first year or two in a TT job.)
I’d be very interested to hear from faculty in other departments about the sorts of things they do to improve graduate students’ teaching: what we have right now is a pedagogy seminar that I believe all our PhD students are required to take, plus regular supervision of graduate student teaching that results in written feedback that the student instructors then respond to.
University of Oregon train their PhD students from the very first years of PhD to teach.
Here at Bowling Green, all first-year PhD students take a graduate seminar on teaching philosophy. Once graduate students begin teaching their own class in their third year, they are supervised once every semester by a faculty member. That faculty member writes up detailed feedback and meets with the grad student to discuss what they are doing well and how they can improve. Many of our PhD students are actively publishing and seeking academic jobs with a research component (and we’ve been successful in placing students in such jobs–we out-kick our coverage). Other students of ours are seeking permanent teaching-only positions. They love philosophy and want to teach it. And they are damn good at it.
Just this year, one of our graduating PhD students won every teaching award the university hands out and landed a tenure-track job at a community college. This is not an isolated success story. We are very proud to train and place students into secure jobs where they can do what they love and focus on how best to teach philosophy. I’m often embarrassed by how much more they know about pedagogical methods and the latest empirical research. We surely aren’t unique in this regard, but I’ve never been around a group of grad students who love teaching more than ours do.
Northwestern had a robust program even back in the 1990s under the direction of Ken Bain. Georgia Tech has a department faculty member fully dedicated to the training of graduate student instructors. Western Ontario is supposed to be good as well.
Sorry, it’s Georgia State, not Georgia Tech, that has the dedicated position.
University of Hawai’i at Manoa has a number of students who take coursework in philosophy for children (P4C) and work closely with the Uehiro Academy for Philosophy and Ethics in Education (p4chawaii.org). Other universities that have strong P4C programs would be good places to add to your list as well.
As discussed before at Daily Nous, the philosophy department at the University of South Florida runs a pedagogy seminar for its graduate students: http://dailynous.com/2019/01/22/approach-teacher-training-philosophy-departments-guest-post-colin-heydt/
We do not have a “teaching-track” for our PhD students (and I’m skeptical that would be successful given the market), but we’re developing a very strong departmental culture focused on constant improvement in teaching.
Looking at the 2018 APDA survey data, 5 or more past graduates and current students gave an average rating of “very satisfied” for teaching preparation at the following programs: Villanova University, University of Kansas, Saint Louis University, Fordham University, Michigan State University, Bowling Green State University, University of Illinois at Chicago, University of Hawai’i at Manoa.
For research preparation, students gave this rating for the following programs: Carnegie Mellon University, Princeton University, University of Southern California, Macquarie University, Massachusetts Institute of Technology, University of California, Irvine (LPS), University of Pittsburgh (HPS), University of Sheffield, Rutgers University, Saint Louis University, University of California, Los Angeles.
For financial support: University of Pittsburgh (HPS), Princeton University, New York University, Cornell University, University of Notre Dame, Stanford University, Emory University, University of Southern California, William Marsh Rice University, University of Pittsburgh, Vanderbilt University, University of Connecticut, Duke University.
Students gave an average overall rating of “definitely would recommend” for: University of California, Berkeley, University of Massachusetts Amherst, Australian National University, University of Southern California, Carnegie Mellon University, University of California, Irvine (LPS), Massachusetts Institute of Technology, Saint Louis University, Rutgers University, University of California, Riverside, University of North Carolina at Chapel Hill, University of Sheffield, St Andrews and Stirling Graduate Programme in Philosophy.
For more details on the programs I have covered so far in detail, including these average ratings and public comments, see links in the running tally, here: placementdata.com:8182/phd-programs-running-tally
A final report for this survey will be provided to the APA by the end of this summer.
Thanks Carolyn. I was trying to use your placement data to find an answer to the second half of the question posted here–which programs have had good success at placing their students in permanent teaching positions (tenure-track or lecturer). Of course, most of them end up placing most people in such positions. But it looks like some are better at it than others, including some that are not PGR-ranked (or high ranked). Have you been able to breakdown the data in a way that makes this information maximally salient?
That would be hard to see from the public data. From the survey data, I looked at programs with three or more participants now in permanent positions who have provided information about the distribution of their expected working hours (between research, teaching, and service)–a bit over 100 programs total. Of those, I looked at the percentage who report that 50% or more of their time is expected to go toward teaching-related activities. This is 19% of overall. The programs with more than a third of their graduates like this (of those graduates now in permanent positions) are as follows: University of Kansas, University of Kentucky, University of California, Irvine, University of Cincinnati, University of Calgary, University of Iowa, University of Maryland, College Park, Arizona State University, Binghamton University, University of Miami, William Marsh Rice University, University of South Florida, Bowling Green State University, Baylor University, University of Hawai’i at Manoa, Duquesne University, Stony Brook University, Ohio State University, Syracuse University, The New School, University of California, Santa Barbara, University of Illinois at Chicago, University of Missouri, University of Minnesota Twin Cities, Boston University.
Kansas, Bowling Green, Hawai’i, and Illinois are on both lists (graduates are very satisfied with teaching preparation and a third or greater of graduates now in permanent positions report 50% of more time teaching).
Perhaps this is also helpful–for the following programs more than a third who answered our survey say their *ideal* distribution of working hours would have 50% or more go to teaching-related activities (9% of graduates overall say this): University of Kansas, University of California, Irvine, Binghamton University, University of Memphis, University of Iowa.
Also: if you haven’t seen it, the “academic model” in this post is close to what you are after: http://placementdata.com:8182/the-philosophical-gourmet-report-and-placement/
You need to be a bit careful about significance here. If 19% of graduates say ‘more than 50% of my time is expected to go towards teaching’, and a given program has 5 people who answered your survey, there’s about a 25% chance that a third or more say that even if that program has a totally typical propensity to place students in teaching posts. If you want p23 participants. (The pattern is a bit noisy because ‘more than a third’ gets more demanding in discrete intervals.)
Hi David,
I am not sure how to respond to this comment, as I can’t tell if you are actually trying to have a productive exchange with me.
If you are intending to make a productive suggestion here, such as that you think my comments would have more weight if they included statistical significance, it would be more effective as a suggestion if it didn’t start with “You need to…”. Imagine if I said “You need to be a bit more careful how you treat your junior colleagues.” Not effective, right? Or, at least, I would probably have to be someone you know and trust for it to be effective.
If instead you are just giving everyone a head’s up that some of the above programs may have had a third or more of their graduates who served as participants in our survey report 50% or more of their time in a permanent position as going toward teaching without that fact having been established as statistically significant–fine, noted.
If you think there is some convention I am not abiding by, please do spell it out for me. If I include a study like this in a paper or research report I would, of course, aim to use the appropriate statistical method. I don’t think I have to run statistical tests for every blog post comment. I don’t think I implied that the above passed a test of statistical significance. Here is my view about statistical tests: they are tools we can use to try to discover if a finding counts as tracking a true feature of the world. Such a test cannot prove this on its own–it has to be the right test, the data have to be good and sufficiently powered, and one can’t have already engaged in p-hacking, such as running a test on every idea that one might have had along the way. I am trying to be careful about all of this, but I am open to advice from those who are knowledgeable about the use of statistical methods in the social sciences. As it happens, I already count on the advice of such people, but if others have thoughts or advice on these matters. please feel free to reach out to me.
To Caroline:
The second, basically: it’s an observation that people shouldn’t read too much into the results you give because it’s not at all clear they’re more than noise.
Over and above that I think it’s preferable that when anyone quotes statistical data, even in informal contexts, they at least do a quick sanity check to see if the data is unlikely to be noise. Otherwise the data being quoted isn’t useful. That doesn’t need to be time-consuming or technically sophisticated – in this case it’s just the binomial distribution (we’re asking, “what’s the probability that 1/3 or more people drawn at random from a large population have feature X, given that 19% of the population in general have X.”) Often people don’t, in which case I end up doing it myself if it’s quick and straightforward, and (on blogs) then usually post it.
As for language: if you said to me “you need to be a bit more careful how you treat your junior colleagues”, then yes, that wouldn’t be effective. If you said to me “you need to carry the 2′, or “you need to allow for the fact that those variables are strongly correlated”, or “you need to work in spherical polar coordinates”, it would be. (But in any case, it was really aimed at the reader in general, so the ‘you’ is impersonal – read it as ‘one needs to’ or ‘there is a need to’ if you like.)
To any reader: the second-last sentence of my post got mangled (I think because I used the greater-than symbol, which gets picked up as HTML code, though it could just have been fat fingers). It was supposed to say “If you want p<0.05, you need 15, 18, 21, 22, or more than 23 participants. "
Once again, it’s “Carolyn.”
I guess in both of our cases we thought the other person “needed to” do something, without providing reasons. You have your preferences, and I have mine.
I don’t know why I keep mistyping ‘Carolyn’ – I know perfectly well how your name is spelled. I’d like to blame my autocorrect but I think it’s just some mental blank of mine. I apologize, it’s probably really annoying. (I get irritated by ‘Dave’ and that’s at least a standard contraction of my name, not a flat misspelling.)
Well, to use your language, in the case that it helps you to see the point a bit more clearly:
I think it’s preferable that when anyone engages with another person, even in informal contexts, they at least do a quick check to see if they are using language that is respectful to that person. Things like checking to see if the name is correct. Otherwise the exchange isn’t likely to be useful. This doesn’t need to be time-consuming or socially sophisticated.
Touche!
But it wouldn’t help, I fear: like I say, I do know how your name is spelled, and I do proofread posts before publishing them. But it’s pretty well documented that people often don’t notice their own typos, even on repeat checking. I can only apologize again (for the misspelling, that is – I stand by the rest of the post!)
The name is one thing. The original post is another.
“You need to be a bit careful about significance here.”
If you had meant the general “you,” then a quick proofread would have picked up that this doesn’t convey that meaning. It conveys something else.
To see what else one’s words convey, sometimes it is helpful to try to read them from the other person’s perspective. People who don’t have much power, your junior colleagues, likely do this all the time when they talk to people with more power, their senior colleagues (like yourself); they can’t afford to be seen to show a lack of respect. They are punished, socially, when they do so. Being in a position of power, and I do take myself to have power in certain respects, means taking that extra moment to think about the way we talk to those with less power. To think about what our words might mean to them. Publicly telling them they need to do something, without providing explanation, comes across as a demand or instruction. In my view, in this case, as an inappropriate demand.
You offered your advice to me, and I am offering mine to you. May we both be better for it.
Okay, we’re clearly just talking past each other here. I was making an elementary point about statistical significance, of a kind and in a tone that I have been making on blog posts for the last decade (when I was at an earlier career stage than you are now, for whatever that’s worth). I would have spoken (and often did speak) in exactly the same way, as a postdoc, graduate student or undergraduate, in a discussion with anybody else, however ‘senior’, on a mathematical or otherwise-technical matter, and would have expected the same from them.
But you are obviously reading my comment very differently (I guess it’s even possible there’s a UK/North America style issue at play here) and from experience this kind of conversation becomes unproductive quite quickly, so I won’t comment further.
No, I understood your point. I just disagree with it in this case. I think you don’t understand my point. I am a British citizen and I did my undergrad at St Andrews. This is not a British thing. It also isn’t a point about statistical significance. It is about whether I needed to run statistical tests for my comment above, since I did not. Your point about the conversation being unproductive is the very point I have been making. If you really want it to be productive, in my view you should explain why you are telling someone to do something, not just tell them to do something, especially when it is a junior colleague. It comes across as needlessly bossy. I thought that was clear from my first comment, but I guess not.
I’m not sure I’m following this exchange. Aren’t both of you tenured philosophers of roughly the same age? I mean, I know USC is more prestigious than UC Merced, but from where most of us are sitting UC Merced is a pretty sweet job.
In any case, on the first-order issues: (a) I think “you need to” and “one needs to” often express the same content, and (b) I guess I’m still not sure why reporting statistical insignificant results as being interesting or important would be an appropriate thing to do, on a blog or otherwise. It’s fine if you didn’t realize they were statistically insignificant–I agree that one doesn’t *in general* need to run statistical significance tests before commenting on a blog. But I do think that if someone points out that the results one has reported aren’t statistically significant, that’s a fairly decisive rebuttal.
JD–no one knows whether or not what I posted is statistically significant. Wallace is pointing out that some of it may not be, but someone would have to look at the data carefully to know for sure. No one has done that. But if it was a descriptive point that doesn’t matter at all. I clearly put it in descriptive terms above, not generalizing. Statistical significance applies when attempting to generalize. Further, not finding statistical significance *does not* make something noise. That is a simplistic understanding of statistical significance. As I say above, it could be the wrong test, you could have run too many tests, it could be underpowered, etc.
On power: I am tenured as of July 1st (yay!), But power differences go beyond the tenure, non tenure divide.
Since this now seems to have returned to an area where I think the conversation can be productive (and since I *think* I’d being told I have a ‘simplistic understanding of statistical significance’, though I could easily be overreading!) let me reply:
“not finding statistical significance *does not* make something noise”
Indeed not, but it does mean one should be skeptical about drawing any conclusions from that something that assume it’s not noise, e.g., letting it inform which PhD program to apply for, which is the topic of this thread.
“Statistical significance applies when attempting to generalize.”
It also applies when pulling particular elements out of a dataset, and saying or implying that those items have a particular property, e.g., being graduate programs that are unusually good at teaching.
“it could be the wrong test”
I didn’t even run a test, just did some elementary probability theory (though I imagine there is a ‘test’ functionally equivalent to what I did). But if you want to call it a ‘test’, it’s not the wrong test. There are areas where the right way to think about the probabilities is subtle and arcane; this isn’t one of them.
“no one knows whether or not what I posted is statistically significant.”
Indeed; that was my original point: “people shouldn’t read too much into the results you give because it’s not at all clear they’re more than noise”.
Many different standards can inform what programs one might select from a list in responding to a comment on a blog post, and a guess at significance making some pretty broad assumptions is one such standard. One reason *not* to use that standard in this case is it would mean never or almost never including small programs on such a list (the point about being underpowered I have made a few times now). This is one way social science differs from physics, making the required statistical methods more complicated: missing data, small data sets, etc. Again, that something hasn’t been shown to be statistically significant does not make it noise, does not render it meaningless. This is absolutely not a matter of “carrying a 2,” but of convention. You may prefer that I only post comments when I have checked them for significance, and that I only report those numbers that are demonstrably statistically significant. But that is your *preference.* Since false negatives are just as problematic as false positives if one *really* cares about truth, and not just point scoring, the reader should also be skeptical about people saying they should be skeptical when, really, the truth hasn’t been established *either way*. I reported a select group of facts and reported the way I selected them. If I had checked for significance I would have said so. If I thought we could draw broad conclusions from it I would have said so. I think I was careful with my language. How about you? Were you careful with yours? Was it fair to imply that I made a mathematical error here? That what I said is meaningless? That others should treat my comments as noise? Should we start assessing each other’s comments with this much scrutiny, or would it be better to just focus on ourselves and what we can bring to the table? You start. (General you.)
If a dataset is too small to meaningfully distinguish signal from noise, it’s too small to meaningfully distinguish signal from noise. A desire to include small programs does not affect that.
Otherwise, I think I’ve now stated my position as clearly as I’m going to be able to (in this medium at any rate) and I think there is enough material for anyone reading to draw their own conclusions (and I should probably do some work!), so I won’t reply further on this matter.
Carolyn: congratulations on your tenure.
“If a dataset is too small to meaningfully distinguish signal from noise, it’s too small to meaningfully distinguish signal from noise. ”
Sure. But I said: “a guess at significance making some pretty broad assumptions…would mean never or almost never including small programs…making the required statistical methods more complicated”. It isn’t that you *can’t* say anything meaningful in the social sciences, you just often can’t use the easy methods of physics. You have to try a bit harder, using more complicated techniques.
“Carolyn: congratulations on your tenure.”
Hey, thanks 🙂
I would be very interested to know what colleges and universities have or would create “teaching-track” tenure positions.
Primarily because I would apply in a heartbeat.
Many schools have these, they aren’t tenured positions, but they’re essentially permanent positions. “Teaching Assistant/Associate Professor,” “Professor of Practice,” “Retained Instructor,” etc. Lots of different titles for the same position.
This paper might be a good resource to answer this question. https://philpapers.org/rec/CONTSO-23
Baylor University has a teaching philosophy seminar that all PhD students must take in one of their first three years. As a grad assistant they tend to get additional teaching experience (which of course varies depending on who you are working with), then teach one course per semester years 4 and 5 (usually intro and logic, but the last year this might include ethics or other 1000-2000 level courses depending on student expertise and departmental needs). We ranked number 5 in the 2017 APDA placement rankings, and most of those jobs are at teaching universities and SLACs.
Thanks to everyone who commented. This is helpful information for our students. Carolyn, thank you very much for putting together the placement information, which does provide insights into this question.