Dissertation interviews work when the conversation is designed to produce evidence that the research question actually needs. A long list of questions is not automatically a strong interview guide.
Confirm that interviews fit the research question
Interviews are often useful in qualitative research when the study needs participants to explain experiences, meanings, decisions, perceptions or processes. They may be less suitable when the main evidence needed is observation, documentary records or precise measurement.
Interviews are strongest when the participant has knowledge or experience that cannot be captured adequately through a simple measurement. Questions about how a decision was made, why an event was experienced in a particular way, or how a process operates in practice often benefit from explanation and follow-up.
They are less appropriate when the dissertation primarily needs precise numerical estimates or evidence that participants are unlikely to know accurately. Method choice should therefore begin with the evidence need rather than the assumption that interviews are automatically suitable for qualitative work.
Choose the interview structure deliberately
Structured, semi-structured and open formats provide different levels of consistency and flexibility. The choice should follow the research purpose and the sampling strategy used to identify relevant participants.
Build questions from the research objectives
Each interview question should have a reason for being there. Remove items that are interesting but unrelated to the study and avoid double questions or leading wording.
A useful interview guide can be mapped back to the objectives. Each main question should contribute evidence to at least one objective, while probes can explore examples, reasons, sequence, consequences or differences in experience. Questions that serve no research purpose can make interviews longer without improving the analysis.
Begin with wording participants can understand rather than academic terminology from the dissertation. Technical concepts can still guide the researcher’s thinking, but participants should not need to understand the theoretical framework in order to answer.
Use probes without leading
Probes can ask for examples, clarification, sequence or explanation. They should deepen the participant's account without inserting the answer the researcher expects.
Neutral probes include requests such as asking the participant to explain what happened next, provide an example, clarify what a term means to them, or describe why something mattered. A leading probe instead introduces the researcher’s preferred interpretation and can change the evidence being collected.
The interviewer should also tolerate short pauses. Filling every silence with another question can prevent participants from developing a fuller response.
Plan procedure and recording
Decide how participants will be approached, where interviews will happen, how consent will be handled, whether recording is permitted and how data will be stored.
Prepare for transcription and analysis
The interview process should anticipate how the material will be analysed. For thematic-analysis depth, use thematic analysis help.
The dissertation should decide what level of transcription is required for the intended analysis. Some projects need essentially verbatim transcripts, while others may not require detailed notation of pauses, tone or overlap. The decision should be consistent with the analytical approach.
File naming, anonymisation, correction of obvious transcription errors and the handling of inaudible material should also be planned consistently. These choices create the dataset that will later be coded and interpreted.
Protect participants and sensitive information
Privacy, confidentiality and sensitive topics should be considered before data collection. For deeper ethical review, use dissertation ethics help.
Review the guide before data collection
Read the guide from the participant’s perspective before using it. Check whether the sequence is understandable, whether questions overlap, whether any wording leads the participant, and whether every question contributes evidence that the research objectives actually require.
For broader methodology integration, return to Methodology Chapter Help.