How big is PER

Some common questions about the size and shape of PER, aimed at curious physics departments.


Eleanor C Sayre


June 6, 2018

Starting with my editorship at PERjobs and continuing through my work with PhysPort, I end up talking to faculty at institutions who are curious about starting a PER group, or hiring their first PER person. Those faculty are usually curious about whether having a PER person is right for their department, whether that person would be tenurable in their department, and how they can make a case to their faculty that a PER person is the best next hire.

This FAQ is aimed at those faculty. If you’re one of them and you have more questions, please send me an email.

0. How big is PER, and how can you tell?

These are complicated questions to answer because the demarcation problem is hard and there’s not a lot of data collected on it.

Some people do a little bit of PER in addition to their ordinary physics research. Some people do a little bit of PER and a lot of teaching. Some people do a little bit of research and a lot of curriculum development and evaluation. Some people don’t identify as PER researchers, but do a little bit of research or evaluation on the side. And of course, some people do a lot of research.

Additionally, some PER researchers don’t hold faculty positions – they are staff (lab directors, center directors, faculty development people, etc) – and some PER researchers hold positions in Schools of Education or industry.

1. What is the approximate size of the PER community?

Depending on how you count, the size of the community is measured in the hundreds. There are about 400 people every summer registered for PERC (our central research conference) and about 1200 for AAPT (which also draws high school teachers). There are about 700 members of the PER topical group of the AAPT, and somewhat less than that for the APS group on PER. In the last 30 years, about 1000 people have published in the top three journals for PER. The top three journals capture about half of all PER papers.

For faculty or other people who have a PhD and are active PER researchers, and have gotten their degrees in the last 15 years or so, there is a mailing list with about 150 people on it. The list doesn’t have “older” people and it’s somewhat spotty with recent grads, so it definitely undercounts the number of faculty-like researchers.

1a. What is its growth rate?

This is also dependent on how you count, because there are people who convert old positions into PER and PERers hired for generalist positions with little/no research expectation. It’s also hard to measure because there’s no tracking for this. Generally, my sense is that the field generates about 10-20 faculty-level new hires each year, though with big variation. I don’t have hard data to back this up; it is only my sense.

2. Of those faculty, how many received their Ph.D.’s in PER vs. how many switched into PER (esp. later, after tenure)?

Most young people got PhDs in PER or did postdocs in it (PhDs younger than ~20 years). This fraction is increasing. Most people who switch after tenure are more dabblers than researchers, but there are a few core researchers who joined the field later in their academic lives. Nobody tracks this directly, as far as I know. Though there are efforts to do so.

3. What is the breakdown of PER faculty according to small liberal arts colleges, community colleges, large research universities, non-university, etc.?

I don’t know. I don’t know if this is knowable. The demarcation problem gets to be really hard here: do we count someone who got a PhD in PER as a PERer if they haven’t published in 10 years? 3 years? What about someone who publishes in AMO and SoTL? Someone who doesn’t publish, but is deeply involved in evaluation of the the pedagogical mission of their college? The reason why the demarcation problem is harder for non-research schools is that people at smaller schools tend to wear more hats.

I know that SLACs are more likely to have deeply-engaged teaching faculty who don’t publish often, and they’re also more likely to have generalists. My sense of the TYC community is that they are more focused on curriculum development than basic research.

There are about 10 populous, high-output, prestigious, graduate-degree-granting groups: (from west to east) University of Washington, CU-Boulder, Kansas State, UIUC, Michigan State, Ohio State, NC State, Florida International University, University of Maryland, University of Maine. There are about another dozen groups that regularly put out PhD students or postdocs, but do so less frequently.

4. Of those at institutions with graduate programs, how many graduate students does a PER faculty member typically support and how many graduate per year?

The numbers here are pretty variable. My group is steady state at about 6-8 students (1-2/year) and 2 faculty. Maine is three faculty and probably 6-8 students (1-2/year), Rutgers is 1-2 students and 1 faculty (1 student / 2-3 years).

Probably an average group is slightly less than 1 student / year / faculty member, but the error bars on that are huge. A typical PhD takes about 5-6 years, same as the rest of physics. Postdocs in PER are common but not required, and they usually last 2-3 years.

5. Are the demographics of any of the above significantly different from the larger physics research community, e.g. in terms of gender or underrepresented minorities?

Yes! This is something for which there is actual data. Ben Van Dusen and Ramon Bartholemy conducted a study of graduate students in PER in the early-2010s. At the graduate student level, PER is about 50% women and mostly white. A separate study at the faculty level (early-2000s – this is old data), said it was about 30-40% women.

6. What can you tell me about the genesis and development of PER programs across the country?

In the last 10 years or so, several institutions have decided to start new (or transplant old) research groups in PER through multiple hires. Other institutions have decided to capitalize on other DBER positions to create new STEM centers with PER as new lines. A few research institutions have hired a solo person to do PER, but that person is lonely and I don’t think it lasts: either they hire more people, or the person leaves. If you decide to start a PER group, you’re in good company.

7. My department is concerned about solo researcher loneliness and why faculty leave. What can you tell me about the cases where people left?

Mostly, these are not my stories to tell.

In the past 20 years or so, I know of six unsuccessful tenure cases in the field, or cases where someone left pretenure after getting strong advice not to seek it. Four of them were solo PER faculty, and at least three of them had substantial research funding. In four of the cases, it seems like tenure was denied because the department or the dean felt that the research area was inappropriate to the institution, though in two of the cases it’s pretty clear that the researchers weren’t performing at a tenurable level for any institution. I can’t say very much about it because I don’t want to “out” the researchers.

It seems like the best possibility for success in a new group for PER is the same as for any group: hire several people, at least one of which should be senior; have clear expectations for success, both within the department and the university; have cognate groups in other fields already, or have good chances for starting them. If you do all of this, probably you will succeed. If you do some of it (and most places can’t do all of it), you can still succeed.

8. I have had several instances of people telling me “I was chatting about the use of (evidence-based pedagogical method) with my colleagues at (institution where it was pioneered), and they say the faculty and students there both hate it.” This is perhaps more common than not for the PER community, but I think it would help to know the details in advance. How should I respond?

I’ve heard this many times, and it’s depressing every time. I think there are several forces at play here:

  • Cranky students and faculty are louder than complacent ones. In polls of the whole class, students are a lot more likely to like innovations than dislike them. But, in teaching evaluations and in discussion, it’s the ones with an axe to grind who yell the loudest. Noah Finkelstein has done some work on this, so you might ask him. There’s growing anecdotal evidence that students appreciate reforms long after the course is over, when it’s not possible to go back and do teaching evaluations again. Personally, I would rather have good long-term results than bad long-term ones, but the reporting structures we have don’t support that kind of long-term measurement.
  • Some faculty think that teaching in a reformed way is more work than teaching in their ordinary way, and they chafe at being required to teach in a way that isn’t natural to them. Sometimes PER-based teaching is more work (especially if you’re reducing class sizes and therefore teaching more sections). But, sometimes it’s not more work, just a different kind of work than faculty are used to. I “grew up” in PER-influenced classes: as an undergrad my coursework was PER-based, as a graduate student my early teaching was in PER classes, and as faculty my natural teaching methods are PER-influenced. Because that’s natural to me, it’s less work than writing powerpoint lectures. Other faculty are different. I’m happy to help them see how I teach, but I’m not going to force it on them. I would be angry too, if I were required to teach in a way that felt unnatural to me.
  • Some people think that PER only means developing new ways to teach. While some PER research focuses on developing new curricula, that’s far from the only strand of research in the field. It just happens to be the most visible to students and ordinary faculty. Here’s an analogy: new curricula are to PER as lasers are to AMO. While lasers are important, not all AMO people are laser instrumentation researchers, and some AMO people don’t even use them directly.

In skeptical faculty, these forces combine to lend an impression that a newly-hired PER researcher would reform the introductory course and require everyone to teach like them. Students would hate it, and faculty would hate teaching it. There are some PER researchers who might do that, but I think most would not. As a department, you need to have extended conversations (possibly with some PER people) about what you want from a PER group, and what a PER group might want from the department. If you’re in the midst of these conversations and you’d like to talk to a PER person about the space of possibilities, I’m happy to chat with you personally, your department chair, or your department as a whole.

A small counterpoint to this narrative: Kansas State reformed the calc-based class in the late 1990s, and it is still going strong. As far as I know, everyone either likes it or doesn’t mind it. There is talk about reforming some of the other intro classes to match. It seems like the biggest barrier to implementing it in another class is having enough rooms, not cranky faculty or students.

9. Faculty in my department are expected to have a vibrant research program with our undergraduates as researchers. Can a PER person do that?

Emphatically, yes.

I regularly take undergraduates as researchers, both during the academic year and over the summer. Some of them are physics students who want to be PERers; some of them are psychology students who like working with adults; some of them are pre-service teachers who think PER will be more relevant to their intended careers; some of them are just physics students who want to do research before they apply to grad school. I’ve published papers with some of them, brought some to national conferences, and convinced some to go to grad school. PER research is everything a standard physics research experience is.

In fact, if your department is strapped for cash (who isn’t?), PER people are very cheap as experimentalists go.

10. Would a PER person even want to come here? My department is really afraid of having a failed search because there are no acceptable candidates.

There are PER people at all kinds of institutions. To maximize your chances for a successful search, you need to be honest with yourself about what you want a PER person to do and what kind of resources you have to support that person.

If you’re a teaching-focused institution, you don’t want someone who plans to aggressively pursue external funding. If you’re a top-tier research institution with high expectations for funding and publication, you should not hire a tenure-track person to reform your introductory sequence and teach a full load – there simply aren’t enough publications there, and not enough leftover time to do publishable research.

Conversely, if you are used to acceptable candidates having had 6-8 years of postdoctoral experience after their PhD, you need to renormalize your expectations. PER people tend to do 1-4 years of postdoc / VAP. You can expect 3ish papers per year for a typical postdoc, and 3ish papers total for a recent PhD. Exceptional candidates are, of course, better.

You should also think about being flexible about qualifications for your potential hire. If you need someone to teach 4 sections, you don’t need a physics PhD; a physics MS and an education PhD is fine. (Someone with that teaching load isn’t going to do much research: does it really matter what research field they’re in?) If your faculty tend to teach courses at all levels, you should look for someone with graduate coursework in physics; if this person is going to teach the methods class for teachers, you should look for someone with graduate coursework in education.

Back to top