Scholarship on teaching and learning is research designed to test theories and gain knowledge that can be generalized to many teaching settings.
When Teachers Become Researchers and Students Become Subjects
Participation in research must always be voluntary. Ensuring that students’ consent is voluntary requires addressing the power dynamics in classrooms. Instructors have very real power in students’ lives. Even though students may be told that their decision about whether or not to participate in research will have no effect on their grades or their relationship with the instructor, they may not find that wholly believable.
Instructors have access to information about student performance such as their test scores and final grades in courses they teach. Even though they are entitled to have access to this information, they are not entitled to use it for research purposes. According to the Family Educational Rights and Privacy Act (FERPA), students must provide written permission for the use of test scores and grades for research purposes.
Addressing Student Choice
One way to address the concern that students may feel undue pressure to participate in research is to use strategies that ensure instructors will not know which students choose to take part, or to make sure students understand the process of concealing their choices. Examples:
- If a researcher would like to use a required written class activity or survey for research purposes, students may be asked to indicate at the end of the activity or survey, whether they will also allow the researcher to also use the data for research. No identifying information should be collected, ensuring that the instructor will not know which students provided permission.
- If student responses to surveys or other measures will be tracked over time, students can be asked to make up their own unique identification numbers based on a set of questions provided by the researchers, for example, “What is the name of your first best friend?” The students can put their number on the measures.
- If it is necessary to collect student names or Duke ID number to access and link data, for example FERPA-protected data, someone other than the researcher may secure-informed consent, access data, and de-identify the data before giving it to the researcher.
Risks to students include an inadvertent disclosure of identifiable information, or deductive disclosure of their identities based on demographic data. Student data containing demographics are susceptible to deductive disclosure. Data protection plans approved by Duke’s Information Technology Service Office (ITSO) may need to be included in the protocol, and when appropriate, demonstrate how the risk of deductive disclosure is managed.
Using Identifiable Data
If materials used in the classroom cannot be de-identified, for example, journals or video recordings of students, consent to use the materials for research purposes should be obtained after issuing final grades. This timing ensures that students know that they are (or are not) consenting to the use of their identifiable data for research purposes, and without fear of retribution while enrolled in the class. The consent form should make clear how the data will be reported and who else other than the researcher will have access to the material.
Other kinds of identifiable student data that researchers/instructors may wish to use for research include the contents of an optional web-chat room where students interact informally. Team activities that are part of a curriculum may be used for research purposes too. In all cases, researchers should ensure that students do not feel unduly influenced to allow their responses to be used for research purposes. Consent must be secured at an appropriate time, for example at the beginning of the semester for web chat rooms, and data should be anonymized when possible.
Risks to students include the use of identifiable data by people other than the researcher, for example, a video tape or journal that students released for research purposes as presented at conference proceedings. The researcher will need to address how they will manage the risk of unintended use by others without participant consent. The consent process should explicitly state whether data/materials will be made public or available to other researchers.