Writing in the Disciplines Knowledge Base

Students conduct their own research

This approach is highest in authenticity and also in workload. Having students conduct their own research means you must provide them guidance for the whole experience of being student-researchers, from proposing answerable questions to designing methods constrained by the time frame of the class to writing the report.

• Advantages: This can be a great choice if the whole class is about writing research reports or you wish to add writing to a methods class.
  • produces high student motivation
  • produces high student involvement
  • produces more genuine discipline-specific learning
• Disadvantages: This can be a poor choice if you don’t have time (or desire) to help students set up an entire research project.
  • produces high student stress
  • produces significantly more work for teacher
    • teacher must develop a meta-vocabulary about the discipline — i.e., in addition to “what” (content), you also have to add “why” and “how”.
    • Students will have to understand how to ask an answerable question using valid methods in the time allotted for class. If students are led through this process carefully, then they can often get through data collection very quickly. However, the teacher will have to set it up so that methods are tightly controlled, and only allow time-intensive approaches (qualitative, interview-based, some simulations) to go ahead with special permission or not at all.

Students will need training and access to research tools

I’ll assume that you are not teaching a lab class; if this is the case, all the “tools” needed have been provided! For the rest of us, conducting original research means that a student will have to perform some kind of methodology. Students will also need training in choosing methods, choosing variations within methods, and training/support in how to do the method itself.

• Be reasonable with environmental constraints
  • Time — how much time will students have to gather data? In a single semester, it’s usually about a week, maximum. Students will have to ask a question and design a method that can be accomplished in a week’s time.
    • Our students can usually gather 50-300 survey responses in 3 days, but qualitative research takes much longer to both gather and analyze, so should be reserved for those students advanced enough to do the work (I’ve met only a handful of undergraduates who are intellectually ready to perform thematic analyses, and most classes, whether skill or content, do not leave enough time to teach this).
    • It goes without saying that cross-sectional designs will be the only options here; there isn’t time for anything longitudinal!
  • Money — will it cost the student money? If students need to make copies or buy something, was that cost figured into the class and explained at the beginning of the semester?
  • Location — if students need to be physically present for any part of the method, have you worked this into the schedule, including weather-induced variations?
    • Provide instructions for how to do observations, too, whether participant-observer or counting instances or visual inspections.
• Limit the methods students can use to reduce the burden on you
  • Since you’ll have to train students to do the work, limit the possibilities, e.g., surveys or interviews or observations. Keep in mind this is an exercise in disciplinary thinking, not a graduate degree. The goal is for students to make all the decisions someone in your field makes, from question to data collection to communication. They do not need to become expert researchers.
  • For every allowable method, offer some training in how to do it. For less experienced students (which is most undergraduates), review their method and FIX it so that the data will work better. This is not coddling or cheating. Instead, it supports the goal of ensuring that students have a disciplinary experience: they cannot write a discussion section if there are no patterns in their data. Your expertise supports this goal.
  • Constrain the length and complexity of what is allowed. I’ve learned the hard way that smart, ambitious students will unintentionally create huge workloads for themselves by designing a really good 40 question survey that includes both closed and open questions. Even with modern software, that’s a lot of data to sift through for a class paper. Analyzing that much material also shifts the focus too much towards being a researcher when we want the focus on being a writer.
    • It’s hard to do this because we know that a solid paper usually has a ton of data behind it. If it helps, construct a scenario in which a shorter study makes more sense, e.g. a pilot study, feasibility study, or focus group. Or, use a scenario where highly specific information is needed, such as feedback from a specific neighborhood about a particular construction issue or convenience sampling only the patients who come to the office during 2 weeks of flu season.

Students must conduct ethical research

IRB — Officially, research that is conducted for a class is considered non-IRB controlled b/c the results won’t be publicly disseminated, though UF has moved to a stance where ALL research, even that not intended for publication, should have IRB approval. Check with your institution to find out. You’ll then have to make a decision about whether to pursue IRB approval for each project in the class. If you intend to have students publish or share information in any way that is construed as “public”, then you are ethically obligated to seek IRB approval. If there is no sharing component, then IRB approval is less necessary (though in many institutions, strongly preferred).

Nonetheless, our personal position is that students should not use this as an excuse to conduct research that would not be IRB-exempt or fast-tracked. This means that teachers must act as the IRB board, and explain IRB to students, provide the categories, and insist students do IRB-exempt or fast-tracked research. Students often cannot simply ask people the kinds of questions for which they want answers nor can they experiment upon the public; instead, they have to find ways of asking ‘around’ the situations or go for a less incendiary topic. For example, students who want to investigate trauma, such as sexual abuse, cannot do so directly. Nor should they: there are strict protocols on such research that include availability of counseling. However, students can find out “about” traumatic topics that do not question participants directly on whether they have been personally affected.

• Teach and require consent forms — I find this the single most common piece of ethical research practice left out of coursework.
  • If a student wants to work in ANY capacity with a minor, they MUST get parental consent. If they don’t have time, they’ll need a new research question.
  • If a student wants to work in ANY capacity with a public official or employee, they MUST create a consent form that explains exactly how information will be used.

Page 2: Students conduct their own research
Page 3: Instructor provides part of the research
Page 4: Scheduling and Scaffolding