2026 ASEE Annual Conference & Exposition

This methods/theory research brief examines how different types of interview prompts influence the depth, diversity, and relevance of responses in qualitative interviews about well-being and identity among engineering doctoral students (EDSs).

This study addressed two research questions:
RQ1: Do focused and broad prompts differ in coding density?
RQ2: Do focused and broad prompts differ in code and thematic distribution and co-occurrence patterns?

Multiple semi-structured interviews were conducted with nine biomedical EDSs about their experiences preparing for qualifying exams at a mid-sized private university (IRB approved). Given prior research linking qualifying exam preparation to EDS well-being [1], this context provided a naturalistic setting to examine how interview design affects data quality on sensitive topics. Interviews included both focused and broad prompts. A crossover experimental design [2] was employed to ensure approximately equal representation of each participant characteristic (e.g. presenting high- versus low-stress) across conditions. For the second interview, one group received broad prompts followed by focused follow-up prompts (Protocol A); the other group received focused then broad prompts (Protocol B). Groups switched protocols for the third interview.

Interviews were transcribed and analyzed using inductive thematic analysis [3,4]. A codebook was collaboratively developed by two coders through calibration sessions with a senior researcher. Inductively-derived codes were grouped to align with a modified PERMA-V framework [5,6]. The final codebook included 75 unique codes. Each quotation was classified by prompt type and protocol.

Differences between focused and broad sections were tested across three dimensions: the number of codes per excerpt, response length, and thematic breadth (number of codes per segment). Code co-occurrence patterns were also examined across protocol types.

Statistical analyses revealed no significant differences between prompting methods in response length per excerpt, thematic breadth, thematic coverage, or code co-occurrence patterns. Qualitative observation revealed that broad responses were often characterized by extended elaboration on single ideas—participants "thinking aloud" or refining their articulation—rather than introducing multiple distinct concepts. In contrast, focused prompts elicited more targeted, quotable statements addressing aspects of well-being.

Despite producing no statistical differences in information density or thematic diversity per excerpt, the two approaches shaped discourse patterns differently. Focused prompts yielded more discrete codable units per unit time, making them efficient when interview duration is constrained or comprehensive thematic coverage is prioritized. Open-ended prompts produced fewer excerpts but allowed participants to direct conversation toward topics they deemed important—a form of implicit prioritization. However, much of the additional length in open-ended responses reflected discursive processing rather than substantive elaboration, meaning researchers should anticipate needing more interview time to achieve equivalent thematic breadth. For EDS well-being research specifically, a hybrid approach is recommended: broad prompts early in interviews to build rapport and allow participant-driven prioritization of sensitive topics, followed by focused prompts to ensure systematic coverage of theoretically-relevant domains. Prompt type selection should be based on desired interaction style and project constraints rather than assumed data quality differences, as both approaches yield equivalent thematic content through different conversational pathways.