Semi-Structured Interview

The semistructured interview offers a means for understanding human experience through the engagement of two persons in shared inquiry. It is a qualitative research method involving open-ended questions interspersed with questions of clarification, further meaning-making, and critical reflection within the interview. Semistructured interviews offer much promise across disciplines and research topics.

The investment in preparation of a semistructured interview and attention to the nature of the interactions as they are unfolding in the interview contribute to the [Page 1486]quality of the data gathered. Formulating questions and ordering them requires careful thought and some field-testing of the protocol, the set of questions that guide the interview. Some researchers will engage in a preliminary study of the context or involve individuals who share some connection to the topic or are insiders within the community of study. This level of collaboration can strengthen the formulation of questions and the structure of the interview and assist the researcher in anticipating ethical issues.

The order in which the questions are placed within the interview protocol reflects a deliberate progression toward increasing depth and exploration of the phenomenon under study. However, the instrumentation is not entirely contained in the interview protocol. The researchers themselves act as the primary instrument, as they extend questions and pursue ideas conveyed in the participants’ responses, probe particular statements, and reflect back to the participant what the interviewer is understanding. The researcher is acting in the role of a co-inquirer with the participant in working toward answering a broader set of research questions. In this way, the complexity of knowledge produced stems from the intersubjective nature of the social exchange between individuals engaged in inquiry together within an interview.

The semistructured interview offers great versatility in the arrangement and types of questions. It is dependent on the skills of the interviewer, which include but are not limited to the following: (1) the quality of questions structured into the interview protocol; (2) the responsive construction of additional questions as the interview progresses; (3) the degree of care in listening and in understanding what is said; (4) the ability to stay focused on the narrative as it is unfolding while keeping track of unfinished stories, interrupted sequences, and potential insights to which the interviewer might return; (5) the capacity for intersubjective engagement toward grasping another’s experience or perspective; (6) the encouragement of description of experience through examples, stories, and details of lived experience; (7) and the creation of generous space for the participant to introduce tangents that may be rich and complex while redirecting the exchange should it move too far out of focus.

As the interview progresses, the interviewer is increasingly engaging in meaning-making with the participant. Questions the interviewer might be conscious of include the following: How is the lived experience as narrated by the participant informing the phenomenon under study? Is there a story, metaphor, or a particular phrase pregnant with meaning within the interview thus far that needs further exploration? Are there contradictions in the participant’s narrative that might be raised with care and perhaps reveal important tensions? As interviewers proceed in the interview, they move in a manner that invites depth but attends to indications on the part of the participant that the topic has been exhausted, is inaccessible, or is off-limits. This can be difficult to ascertain. It relies on one’s ability to probe to open up areas the participant may not have directly considered while also reading body language, facial expression, and tone of voice to determine participant desire to move on to the next questions. A useful way to gauge the situation is to close out each major area of questioning by asking, “Anything else?”

...