Purpose
Description
A structured interview is one in which a sequence of questions or "probes", often accompanied by a set of tasks which students are asked to complete, is carefully chosen beforehand. The props, tasks, questions, and responses to students' statements (e.g., the particular wording of follow-up questions, requests for clarification, and focusing prompts) are planned in advance to insure consistency across students and groups of students. Structured interviews are typically audiotaped or videotaped and then transcribed for later analysis; though the interviewer can take limited notes during the interaction, its important that it does not distract the interviewee or disrupt the interview protocol.
Because interviews can be time consuming, researchers often select a sample of students from within the larger course enrollment population. A variety of sampling strategies can be used, depending on the goal of the analysis. For example, selecting a broad range of students that represents a cross-section of students who take the course can provide a portrait of how different students within a class are faring with the instruction. In another context, however, it may be more beneficial to select students based on course grades or performance – often some equal percentage of high-achieving and low-achieving students – in order to reveal how different types of problem-solving strategies, study habits, and/or interpretations of course content are related to overall performance.
The data collected via structured interviews can be analyzed in multiple ways. Scoring-rubrics are often applied to interview data in order to test hypotheses about the relationship between various pedagogical techniques, technological tools, or student strategies and differences in learning outcomes.
General Requirements
All interview methods, including structured interviews, work best when the participant is relaxed and at ease. When working with students, it's often useful to conduct the interviews somewhere outside of class, particularly when the interviewer is the instructor. Generally, interviews are conducted individually, though variations on this can be used where appropriate.
Limitations
Structured interviews, like all social interactions, are co-constructed, meaning that both the interviewee and the interviewer shape the context of the dialogue and what is (and is not) said; researchers drawing inferences from such data should by mindful of this fact. For example, inferences regarding what a particular student does or does not understand should be qualified by an acknowledgement of the variety of other reasons why the student may have behaved a certain way – the interviewer may have unknowingly made the student uncomfortable, the particular features of the task or props used may have led the student down a garden path, etc.
Variations
Instances Interviews, which shed light on the extent to which students can recognize an instance of a given concept or when a given procedure or technique is applicable. Here, the student is presented a set of examples and counterexamples of a particular concept or problem-solving scenario, one by one, and is asked to identify which examples illustrate the concept or are appropriate for a particular strategy and why.
Prediction Interviews, in which students are presented a set of situations and are asked to predict the outcome and their reasoning behind it. Such interviews are frequently used in science courses.
Sorting Task Interviews, in which students are presented with a group of objects (typically on cards) and are then asked to sort them in a way they find meaningful. Identical sorting task interviews given to both novices and experts within a particular domain have shed light on the nature of both expertise and learning with a variety of domains (an excellent example is the work of Chi, Feltovich, and Glaser, 19811).
Problem Solving Interviews, which focus on students' performance and conceptual understanding related to a given task. Here, the participant is asked to “think aloud” while attempting to solve a selected problem or set of problems, explaining what he or she is doing and why.
Additional Resources
- Chi, M. T., H., Feltovich, P. J., & Glaser, R. (1981). Categorization and representation of physics problems by experts and novices. Cognitive Science, 5, 121-152.7
- Ericsson, K. A., & Simon, H. A. (1980). Verbal reports as data. Psychological Review, 87, 215-251.16
- Fredette, N., & Clement, J. (1981). Student misconcepts of an electric current: What do they mean? Journal of College Science Teaching , 10, 280-285.20
- Smith, M. U. (1992). Expertise and the organization of knowledge: Unexpected differences among genetic counselors, faculty, and students on problem categorization tasks. Journal of Research in Science Teaching, 29, 179-205.47
- White, R., & Gunstone, R. (1992). Probing understanding. New York: The Falmer Press.51