For various reasons, I’ve found myself this semester talking with colleagues about the migration of students through our curriculum: the courses where they busily cluster, the lonely cobwebbed courses, the majors and courses that follow regular oscillating cycles of interest. We’ve been trying to figure out which classes are interchangeable and which are not from a student perspective, about what our students see when they look at the curriculum.
I don’t think that any of us really know what kinds of decision rules students are consciously and unconsciously employing. Each department and each individual faculty or staff member has his or her own folkloric narrative that explains some or all of the patterns in enrollment. Sometimes that’s based on a smidgen of hard data: real enrollment numbers over a five or ten-year period, some kind of assessment data or evaluation from students, frank conversations with a handful of perceptive students. Some faculty and staff work in contexts where they get more insight into these questions, and others (such as the education faculty) have special expertise that’s relevant for thinking about the problem. But I honestly don’t think anyone has a really systematic handle on the issue at any scale, whether it’s guessing about the total movement of students across the entire curriculum or about their presence or lack of presence in any individual class.
There are reasons why it’s a hard problem to investigate. It’s not uncommon for faculty to misperceive (in either direction) their own enrollments in relationship to the overall distributions, even when they have good data to consult. In part, that’s because the workload involved in teaching doesn’t necessarily scale to the number of students nor is it the same across departments or even between any two individual faculty members. And at least one of the reasons why students flock to some classes and avoid others has to do with their perceptions (and perhaps sometimes misperceptions) of faculty quality and that is a subject that’s nearly impossible to talk about openly in any official context without quickly descending into cruelty and recriminations.
But there’s also no way to completely avoid trying to figure out some of what’s going on. If students are pounding down the doors of a single professor’s courses but not of department colleagues or faculty teaching similar subjects, it might be safe to mark that off as a case of pedagogical charisma, which has no further institutional implications (save that you want to figure out how it’s done and build some of that into a vision of best practices). If an entire program is getting hammered by enrollments, or a single course is constantly over-enrolled regardless of who teaches it, then it’s imperative to figure out why that is. On the flip side, if a course or program is in relative terms under-enrolled (not because of a requirement of small class sizes), it’s important to figure out if that’s because there is a consistent movement of students away from the subject matter, because the course or program is doing a poor job of labelling or framing the subject matter, or because of student antipathy to a particular faculty member. In all of those cases, there are big implications for long-term planning–and big risks to just accepting whatever explanatory mythology comes most readily to mind. When all of that information is put into the structures of a real curriculum with all of its moving parts, the possible explanations for enrollment patterns quickly multiply into near-incomprehensibility. General education requirements and major requirements, various subtle and gross devices that departments and divisions put into place in order to manage, route, repel or capture enrollments (and all their unintended effects), leave cycles and temporary faculty, new courses that are poorly promoted and old courses that are abruptly cancelled, and so on, all exert serious influence over what students take and avoid.
I accept, therefore, that there’s going to be a pretty hard limit to any model that accounts for (and tries to predict) student interest in courses and majors over a five or ten-year period. But what I’d love to be able to do is speak with a bit more confidence, based on a robust mix of qualitative and quantitative data (especially quantitative data that tracks the most common patterns of total enrollment over four years, rather than data about isolated courses or departments), about the relative weight of the following factors:
1) What students (and their parents) believe about the match between particular subjects or disciplines and particular careers or the likely job market at the time of graduation.
2) What students believe the content of particular disciplines or courses is before they begin their studies and how those beliefs change over four years of study.
3) How much of a role the titles, descriptions and “marketing” of particular courses plays in the decision to sign up for a course.
4) How much students are driven by strategies that respond to “traffic management” within the curriculum (trying to secure places in desirable mid-level courses by pursuing entry to an undesirable entry-level required course, for example). Equally, how often curricular barriers such as requirements prevent students from taking courses that they believe they would like to take.
5) How often students believe their enrollment decisions to be driven by a strong attraction to a particular topic, idea, methodology, discipline that they have developed after beginning their studies at the college. (Especially when this represents a change from the initial perceptions relevant to 1 and 2.)
6) How often reputation of individual faculty members (quality, difficulty of grading, etc.) plays a major role in the decision to enroll.
7) What courses students consider to be interchangeable. (E.g., if a student is lotteried out of a course, what courses will they regard as reasonable substitutions and why?)
Having at least an approximate model covering these factors isn’t just for planning. It’s also required for persuading students that the vision of liberal arts on offer to them in a curriculum is one that they should accept and embrace. If we don’t know how our students see the curriculum, we can’t really talk with them about what we believe they should be seeing, let alone what they might consider choosing. And this swings both ways, potentially: we might find out that what we think the curriculum contains or says is palpably not what students experience in their actual coursework, or that students are seeking a plausibly and wholly legitimate different curriculum that’s still completely commensurable with the spirit of the liberal arts, and it’s the faculty that should be persuaded to nudge or move the emphasis of their teaching in some new directions.
You don’t even consider time slots in your discussion; I know my crowning achievement as a class planner was a semester where all of my classes were on Tuesday, Wednesday, and Thursday afternoons. As someone who routinely went to sleep around 4 am I looked very negatively at any course meeting before 10:30.
I would imagine athletes and other early risers have the opposite issue.
Oh yeah, completely. Huge issue.
This was going to be a few sentences. I got carried away.
1. “It’s not uncommon for faculty to misperceive (in either direction) their own enrollments in relationship to the overall distributions, even when they have good data to consult.” -Swarthmore (the institution/the administration) does not provide good data to consult, making it easy for faculty to misperceive their own enrollments. Faculty can peruse individual course enrollment data, but by making these data difficult to obtain, and by making opaque each faculty member’s realized course load, it’s easy to see why people don’t bother to study this in any detail, leading to these false narratives and unfounded conclusions.
2. “In part, that’s because the workload involved in teaching doesn’t necessarily scale to the number of students nor is it the same across departments or even between any two individual faculty members.” -Yes, you have the word “necessarily” in there, but the first half of this sentence frustrates me. I don’t think you’d find many people argue that teaching 10 people is twice is hard as teaching 5. Perhaps you’d even find people say that teaching 10 is the same or even easier than teaching 5. But surely, controlling for individuals, departments and subject matter, you wouldn’t find many people clamoring to teach an 80-student class over the 10-person class they’d been assigned. And while we talk about whether or not the College should recognize the work load differences due to enrollment differences, the College already recognizes this difference. Faculty who can break an 50-person class into two 25-student sections usually get two teaching credits even though they have no more preparation, no more grading, and no more office hours. While teaching two identical preparations does lead to some unevenness in workload across departments and individuals, I think we all agree that the value added in student experience makes it worth the cost. Allowing faculty to make this split into two sections is good for the students; giving faculty two teaching credits for it is partial recognition of this. (Where I get a bit queasy is when we talk about splitting 18-person classes into two 9-person classes.) Lastly, there is the problem of enrollment creep. At some point, each 3 or 5 extra students in your course probably doesn’t significantly alter your work load, but when you get 3 or 5 more the next year, then 3 to 5 more then year after that, suddenly you find you’re working way harder. Perhaps it’s because you’re trying to teaching the same way you taught for 20 students to 35 students, or perhaps it’s because the range of students is necessarily broader with 35 than 20, or perhaps it’s because you’re spending the necessary time to retool the course for the new steady state.
3. ” If an entire program is getting hammered by enrollments, or a single course is constantly over-enrolled regardless of who teaches it, then it’s imperative to figure out why that is. On the flip side, if a course or program is in relative terms under-enrolled (not because of a requirement of small class sizes), it’s important to figure out…” -Go ahead and do the research, find out why there’s an imbalance. From my vantage point, though, it seems like we tell departments to keep getting hammered while we do the research, but it’s not clear that anyone is actually doing this research. If someone is doing this research, it’s moving at a pace that’s too slow to be an effective release for these departments. If there are structural flaws that cause departments or classes to get hammered, we should act to remove the flaws. But your second sentence seems to imply that what we should thinking about ways to add structural incentives to more evenly distribute the workload, rather than removing the structural flaws and then follow student migration with resource allocations to the extent possible. We can do our best to inform students about what a liberal arts education is and why they should be excited about the entirety of the College program, not just their particular narrow interest, but we shouldn’t end up with a curricular prescription that forces students into departments with low enrollments.
4. “I accept, therefore, that there’s going to be a pretty hard limit to any model that accounts for (and tries to predict) student interest in courses and majors over a five or ten-year period. ” -True. What irks me, though, is that when we talk about allocating permanent resources (tenure lines), we’re always talking about whether or not we want to make a 40-year investment. If we can’t know what it’s going to look like in 5 years, why we do pretend to know what’s going to happen in 40? For any department with more than 3 or 4 faculty that we’re contemplating adding faculty to, will no one retire in the next 40 years? Won’t that be an additional moment of reflection where we can decide if the 40-year investment in that department is working out? If so, while we might be making a 40-year investment in an individual, we are almost never making 40-year investments in a department. Therefore, we shouldn’t be so tentative about reallocating resources away from one area and into another given our limited predictive abilities. We’ll have many opportunities to fix our mistakes when we make them. It’s naive to think we’re avoiding mistakes but not reallocating mistakes — we’re just setting ourselves up for different ones.
5. “What courses students consider to be interchangeable. (E.g., if a student is lotteried out of a course, what courses will they regard as reasonable substitutions and why?)” -My view of the curriculum may be too narrow, but I see this interchangeability more easily in courses outside the sciences (with the exception of introductory-level science courses where a large number of students enroll simply to fulfill a curricular requirement regardless of the particular discipline.) Is it that science departments haven’t done as good a job of setting up these equivalencies? Are these equivalencies not there past a certain point in the science curricula we’ve constructed, or are they just not there? On the other hand, I suspect that if you’re a strong student and decide to go to grad school in Biology after choosing to major in Chemistry, you’re not going to run into enormous stumbling blocks. But if you’re a student with an interest in Biology who keeps getting lotteried out of the courses, you are unlikely to be comforted knowing that at least you’re gaining many of the same research competencies through the Chemistry courses you find yourself in.
So on #1, the thing I’m thinking about is that you can show some colleagues the relative teaching loads in a department and that person will nevertheless continue to imagine that he or she is the most burdened person (or sometimes the most needing more students) even if the data doesn’t show that. A lot of us get sufficiently invested in a belief about our relative workloads that we can’t be convinced to think otherwise.
On #2? The point is that the work of a course doesn’t scale on a linear basis. Nobody at Swarthmore that I know of teaches with multiple-choice scanned tests and all classroom time as lecture, but if you did, 50, 80, 200, who cares? Additional bodies are pure efficiency at that point, and no additional burden to the instructor. Which is why some higher education institutions have moved in that direction, and why online education will probably eat their lunch sooner or later. Automate the assessment (or outsource it) and the main teacher doesn’t feel the burden of more bodies. In our environment, if I’m teaching 30 students in a lecture-discussion class with one paper and a final exam, my workload is not three times the person who is teaching ten people with four papers, a research assignment, a journal and a team project. In fact, it’s probably the other way around: the intensity of the workload in the smaller class is almost certainly greater (and the teaching outcomes are probably better). Most of us wouldn’t approach the smaller class that way precisely because it would be too demanding (and probably repel students for that reason) but it’s possible that we might selectively do so for various reasons. I do think there’s kind of a magic number where the size of the class means you can no longer keep track of individuals very well (I gather there’s some good psychological and sociological work on this very subject) and that almost certainly means that you have to rethink workload and learning ambitions at around that number.
3. I agree that you can only engineer the choices of students so far even if you’re a control freak, which I’m not. But (just for one example) it’s really important to try and understand: am I seeing something in high enrollment contexts that’s about a group of faculty doing an unusually exemplary job or am I seeing something that’s culturally and economically way bigger in scope, well beyond our borders? The first is something you can control to some degree (by identifying best practices and looking to spread them) the second is something you can only adapt to and discuss.
4. I think there’s a pretty good argument for giving up the hubris of aspiring to make forty-year decisions. But there’s a collateral point that comes along with that: it might mean that schools like Swarthmore are talking about the wrong things when they try to decide whether students will want a particular field, discipline or specialization for the next forty years–the mistake is not just the arrogance of thinking you can know what that will be, but in thinking that’s what you’re looking for at any time scale. Maybe job #1 is figuring out what intellectual attributes and training encourage a tenure-track faculty member to grow, adapt and retool so that the institution is rarely at risk in its resource allocations. Then you’re not trying to decide if in forty years you’ll still need an expert in phlogiston, because you’re confident that what you’re recruiting is not an expert in phlogiston but a person who happens to be thinking about phlogiston at the moment. (or teaching students who at the moment are flocking to classes in phlogiston studies)
5. Interchangeability is a pretty dense meeting zone between what students think they know about subjects and what we think we know about subjects. E.g., we may know that in real life, some classes and disciplines that look very proximate or similar are not no matter where you are. We also may know that some classes and subjects that in other institutions have close relationships or easy movement between them might not here for all sorts of artificial or structural reasons. Students perhaps imagine or think differently–and of course sometimes they discover real interoperability between subjects, professors or classes that we aren’t aware of and didn’t plan for.
I like the framing of your questions (“the movement of students”) but would like to encourage you (gently, I hope) to expand the frame to include student diasporas, as well. The Swarthmores of the world can more or less assume a stable student body, but how do we address these crucial questions at institutions where transfer students are the rule and not the exception? If anything, the need to be clear about the narratives circulating about general education and the curriculum in general is even greater in such an institution than it is at a place where immersion for four years in a particular culture probably counts for something.
So, I would for example want to know if students who have been to 3 or 4 institutions on their way through college are in a “better” position to understand the relevance or potential impact of questions about how a curriculum is arranged.
Also, would it make a difference that instead of factoring in the narratives offered by parents of the students, my institution might want to factor in the narratives offered by the children (who are sometimes themselves in college) of our students, or anyway the expectations of people at their workplace?
Fred, I will see your schedule and raise: double credit seminars on Tuesday and Wednesday afternoons.
Now, that alone may prove that I made some bad decisions as a college student, but in response to 6, I think professor was the single most important factor in many of my class choices. Especially because the history department had fairly loose constraints in terms of what courses needed to be taken and in what order, I was able to double down on my favorite professors and skip otherwise interesting sounding classes taught by professors I didn’t like as much, or – and this is worse – professors I had merely heard bad things about.
This is basically irrelevant to your local discussion, but…
“Nobody at Swarthmore that I know of teaches with multiple-choice scanned tests and all classroom time as lecture, but if you did, 50, 80, 200, who cares? Additional bodies are pure efficiency at that point, and no additional burden to the instructor.”
This is *almost* true, but not quite. Think of the very occasional exceptional problems that come up with a particular student – that maybe you encounter once every few years, but which end up being very time-consuming. When you have 2000 students in a class (which I’ve seen), they’re no longer occasional problems – you have to deal with them every single time you teach that class, because even at very low probabilities, *some* student will have that sort of problem.
Chris: Full House, kings high. Double credit seminars Monday evening and Thursday evening.