Here's a familiar situation: a professor teaching a large course gets approached by a graduate student. The grad student (who may or may not be in the same department as the faculty member) describes a research study he or she would like to conduct for his/her thesis. However, there's one problem—recruiting subjects to come into the lab for an hour. Wouldn't the professor please give extra credit to students to come to the study? Pretty please? Science (or at least a thesis) hangs in the balance!
I have a number of frustrations with this situation. First, and perhaps unfairly, I ususally perceive most students in this situation are just looking for a way out of their very practical problem, that is, how to graduate. They see a huge pile of subjects, and they don't understand why the professor won't help. After all, what's it to her or him if the students get a few extra points?
On the flip side, the professor usually feels like the target of a bulk marketing ploy. Their duty is to teach the students and ensure a good learning experience, and what on earth could they possibly gain from an hour being a guinea pig? Perhaps due to the frequency of unjustifiable requests, perhaps just because it's a hassle, many faculty will deny such requests out of hand, or even try to get a department policy preventing such extra credit. The most extreme case is one in which a department (Psych 1 pools are famous for this) try to institutionalize the use of subjects for credit but only by 'designated' experimenters. This risks trouncing the academic freedom of the instructor, and the benefit-to-subjects justification for why the students should get credit for participating in favor of a sheer power grab. "My subjects, mitts off!"
The reality is that there are a number of reasons research can benefit students' learning, and therefore can fit comfortably into the realm of course activities, even course activities for credit. I'm a faculty member who does educational research and advises many grad students doing various kinds of technology, education, and social science research projects. And I've seen a lot of cases where providing points for participating in a research study is sound teaching. What are some of the good reasons for including research in a class? And what are some of the ways of doing it?
As I see it there are three main justifications for including research studies into a course: providing a window onto where knowledge comes from, giving students additional training in the domain, or improving the course the students are taking. I take each of these in turn.
In this case, and this is most commonly the justification for Psych 1 subject pools, the idea is that students learn by being exposed to the process of research. This can take many forms; a psychology student might be benefitted in a class on cognitive psychology by participating in cognitive psychology experiments, because they have a direct experience with the methods that are used to derive the information in their textbooks. But what about a developmental psychology course--is there benefit here if the student is participating in an experiment pretty different from the course content? I would say, sure, as long as the student is learning something about the lives of psychologists, the ways of reasoning used in the domain generally, and so on. Indeed, one can make an argument that learning about the methods of other niches in the field is more valuable, because it gives the student something to compare and contrast with what they are reading about in their textbooks or hearing about in their lectures. Indeed, many psych 1 pools at various universities not only give extra credit for participation in studies, they require it as a mandatory part of the course, because they believe the process of experiencing psychology research is crucial to understanding the subject.
This rationale is not limited to psychologists. For instance, when I ask my students in information science to participate in studies on technology, they are understanding how research in IT proceeds, and I'd feel comfortable exposing them to this even if the study was a human-computer interface study and the course was on the economics of IT with nary an HCI finding in sight. Why? Because the types of inferencing are still relevant to the course topic, even if it's not part of the mandatory content.
In some cases there's a nice mapping between a research study and the content of a course. For instance, consider an experiment in which subjects are asked to solve math problems, then play a computer game, then solve more math problems. The experiment is designed to test some ideas about computer gaming and psychology, but regardless the students get practice taking math tests. If I were a math instructor, I'd be comfortable with my students doing this study for credit, figuring that practicing math is good for them, even if the problems weren't specifically drawn from my homework sets.
But this argument can also in some cases be made more generally. For instance, with the same study I might feel comfortable giving credit for participation to students in a freshman seminar on study skills, because it would give me an example to discuss in class about test-taking skills, or practice in taking tests where you don't know every answer.
When a study improves the course a student is taking, I feel comfortable giving credit for participation, on the theory of 'give a little get a lot.' As an instructor, I spend my valuable class time giving mid-semester surveys. Why? Because it helps me improve the class, and give the students a better experience. I also feel comfortable assigning end-of-semester surveys. Even though this crop of students may not benefit directly, they experience the benefits of students before them having completed final course feedback forms, and they probably get something out of the act of reflecting on the course and on their own learning. Likewise, although the feedback may not benefit them as individuals, it does benefit them collectively, and that's okay--I should be able to incentivize this just as I do "class participation."
Similarly, when I do an educational research study in which I'm finding out something about learning that may benefit not only this particular course, but courses of this ilk, or even courses in general, I feel quite comfortable giving credit to students to participate, even if the study involves, for instance, random assignment into control and treatment conditions. As long as everyone is likely to get something out of the experience, and I'm maintaining the 'standard of care' I would expect as a student, this is a reasonable place to receive credit. To be clear, if the control condition is sitting around twiddling thumbs, I wouldn't feel that meets the 'standard of care' test, but if there are a few different instructional techniques I might reasonably try, and I'm not aware of evidence that one is entirely inferior to the other, I believe it's to my students' benefit to try multiple things. In some cases, the data have really surprised me, and I've had a 'control' condition work better than the 'treatment'. Students often get angry if they feel they are getting a different learning experience than other students, but I don't believe anybody benefits by homogenization. I've seen huge undergraduate courses where some department committee decides that all the exams, and all the homeworks must be standardized. Students then get mad even if their lectures vary slightly. But this tends to produce at best a mediocre learning experience, and it removes any possibility of playing to the strengths of individual instructors. It also limits the academic freedom and professional judgment of individual instructors.
In general, research participation must follow two precepts: it must be truly voluntary, and the consent should be informed consent. How, then, can research studies play a role in a course setting, where grades can coerce students (in fact, grades are explicitly designed to coerce students) into doing what the instructor feels best?
One possibility is the extra credit model. With extra credit the instructor feels something is educationally valuable, but makes it optional. At Penn State, any instructor giving extra credit for study participation is required to also give equivalent opportunities for extra credit that don't require being a research subject. I myself have given extra credit for many activities that, while they are not essential enough for learning that I would feel comfortable requiring them, I feel they have learning benefit and merit some reward. In other cases, I've been able to make opportunities available (such as a field trip) that I'd be delighted to require for everyone, but logistics prevent it, i.e., the field trip meets when some students have other, required, class meetings. With research studies, I consider whether the experience is something I'd like my students to do for their own benefit, whether there are enough opportunities for credit to leave something for everyone, and so on. In some cases, I might consider making the points contingent on the study plus something else, for instance a half-page essay on how the study relates to a chapter in the textbook, or something similar. In this way, I can be even more sure the study participation links to the course content.
Another possibility is the required-activity/optional-consent model, and I've used this especially when conducting educational research on or in my own courses. In this case, I might as an instructor decide a certain homework assignment is required for all students. The data I collect as a researcher might consist of their homework assignment and a post-test I administer in class. The activities are normal parts of my teaching. I then ask the students to optionally consent to letting me use their homework and test as research data. In this case, everyone has to do the activity for their grade because it's a core part of the course, and I as instructor have decided it's important for their learning. But they can freely choose to participate or not in the study--I usually just hand out a consent form with a checkbox of 'yes I consent' or 'no I do not consent', and then their work would be excluded from the study if they choose not to participate, and I make sure it doesn't affect their grade either way. (In many cases, it's practical for me to not even look at the consent forms until after the course is over.) Given that there is no extra effort, I've found most students are quite comfortable consenting. If they don't consent, they still learn from the homework, but I don't use their homework as data.
In cases where an instructor and a researcher are collaborating, it works just the same. Some instructors feel morally bound not to allow any researcher-suggested activities into their course, or any differences in how some students are treated (for instance if students are assigned to one of several learning conditions). Obviously, an instructor must use her or his best academic judgment about what constitutes a good learning experience for the students, but the mere fact that a suggestion comes from a researcher, or that its primary purpose is to yield information about the course should not be barriers--remember the end-of-semester feedback surveys or similar activities. Rather, the instructor should make sure all activities meet the 'standard of care' threshold, and that they have relevance within the scope of the course itself. I have been personally acquainted with teachers who felt they knew exactly what was best for their students until they started collaborating with researchers, at which point they both came to understand that their own picture of the class was quite incomplete, and that a small amount of attention to researching the class--even deliberate experiments--could improve it immensely with little effort on the part of the teacher.
Copyright © 2007 Christopher Hoadley. Last updated 17 February 2008.