Conducting, Recording and Transcribing Interviews
Four slides of instructions - five considerations, ten obligations, eight preparations and eleven at-interview rules - and the collisions worth knowing before you sit down opposite anyone: script against probe, eye contact against verbatim notes, a promised recording against a negotiated one.
What the source says an interview is - and it says it four ways
Three interviewer qualities are named there - skilled, adaptable, sensitive - and none is elaborated. The second sentence sits awkwardly against the first: a series of open-ended questions does not produce standardised quantifiable data without a coding step, and no coding step is mentioned anywhere in the data collection week.
FOUR DEFINITIONS OF INTERVIEWING IN ONE DECK
| Where | What it says an interview is | What that excludes |
|---|---|---|
| The survey construction slide | A subdivision of survey research "involving interviewer completing what respondent says" | Any interview in which the respondent records anything themselves |
| The types of questions slide | Governed by the pair structured / unstructured, defined as respondent writes reply / interviewer writes reply | Makes every interview unstructured by definition |
| The types of interviews slide | Ranging from structured to unstructured, and including a "structured open-ended" type | Contradicts the definition three slides earlier |
| The conducting interviews slide | "asking respondents a series of open-ended questions" | Excludes the structured types the previous slide names, and any interview using a scaled or yes/no response format |
Four slides define interviewing four ways and none acknowledges the others. The kinds of interview, and the same collision seen from the other side, are on the interview types page.
Five considerations for interviewers
- "First of all, establish credibility and trust with respondents"
- "Listen carefully to responses"
- "Practise interviewing to develop skills"
- "Observe the respondent carefully"
- "Use probes to elicit richer responses, but don't get side-tracked"
Consideration three is the most useful of the five and the least developed: practise. Nothing in the material says what to practise on, how many rehearsals, or what a competent rehearsal looks like. The one adjacent instruction elsewhere in the week is that question wording should be rehearsed with someone else before you begin, which is treated on The Wording of Questions.
Ten obligations, six of them ethical
The second conduct slide is a list of ten instructions. Six of them are research-ethics obligations, and this is the densest concentration of ethics content anywhere in the supplied material.
THE TEN INSTRUCTIONS AND WHAT EACH ASSUMES
| Instruction, as stated | Kind | What the material supplies to do it with |
|---|---|---|
| "Adapt interview techniques to the person, e.g. those with hearing difficulties, children, various stake-holders" | Technique | No adaptation is described for any of the three groups named |
| "Explain your purpose and methods" | Disclosure | No script, no length, no required elements |
| "Keep any promises you make e.g. sending a copy of the tape" | Conduct | Assumes recording as a matter of course |
| "Explain any risks to the respondent e.g. organisational repercussions" | Ethics | No risk categories, no assessment, no threshold |
| "Ensure confidentiality of information they give" | Ethics | Not defined and not distinguished from anonymity |
| "Ensure anonymity" | Ethics | Not defined; no anonymisation technique named |
| "Make sure you have informed consent" | Ethics | No consent form, no consent elements, no approval process |
| "Explain who has access to the data" | Ethics | No access model, no retention period |
| "Stay focussed on the primary focus and aims" | Conduct | No method for keeping an interview on track beyond the instruction itself |
| "Protect the data" | Ethics | No storage medium, no security measure, no procedure |
Preparing for the encounter
The eight preparations, in the order given
Confirm the arrangement
"Follow up your meeting arrangement before the interview to confirm the venue and time".
Be prepared and punctual
"the respondent is graciously giving you their time" - the material's only statement of what the respondent is owed as a matter of courtesy rather than ethics.
Notebook and pen
"Have a professional looking notebook and pen at the ready". The preparation list is entirely paper-based.
Copies of the survey instrument
The only occurrence of the phrase in the week. No instrument is shown, templated or specified anywhere in the supplied material.
Send written background in advance
"give some written information on the project background and purpose" when arranging a time at the respondent's convenience.
Confirm familiarity
"Confirm that the respondent is familiar with your project intentions".
Take an official identifier
A student card or a business card are the two examples given.
Take a cover letter from your supervisor
Supervision is mentioned nowhere else in any week of the supplied material.
At the interview: eleven instructions that do not all hold together
The eleven, as stated
- First impressions count; appearance professional and in keeping with the respondent's expectations
- Establish a comfortable atmosphere and rapport
- Begin with easy, friendly general remarks and thank them for their time
- Keep the project explanation short and simple
- Ask easy questions first and lead up to the more difficult ones at the end
- Be sensitive to any questions that appear to cause discomfort
- Memorise your first few questions so you can maintain eye contact
- If using structured questions, stay with the script
- Do not interrupt or supply words; silence can be a helpful probe
- Thoughtfully ask for elaboration or clarification where appropriate
- Do not offer opinions or make judgements
Read as a set, three pairs pull in opposite directions and the deck reconciles none of them. This is worth knowing in advance, because you will meet all three inside the first ten minutes of a real interview.
What one instruction requires
- Stay with the script if using structured questions
- Ask easy questions first, difficult ones at the end
- Do not offer opinions or make judgements
What another requires at the same moment
- Use silence as a probe, and ask for elaboration where appropriate
- Design questions so they fall into a logical sequence, which the questionnaire slide requires
- Establish rapport and be sensitive - with no technique given for doing either without responding
Recording responses
Three instructions and one consent caution. The fourth - note who says what, including the interviewer - is the most valuable line in the week for anyone who will later analyse a transcript, and it is stated once and never developed.
WHAT THE RECORDING GUIDANCE SUPPLIES AND WITHHOLDS
| Decision you have to make | What the material says | What it withholds |
|---|---|---|
| Whether to record at all | Make sure the person feels comfortable with audio or video use | No recorded consent, no written consent, no statement of what happens to the recording afterwards |
| How to take notes | Record responses as they are being made; develop your own shorthand system | No system, convention, abbreviation set or worked example |
| How to attribute speech | Note who says what, including the interviewer | No notation, and nothing on how this is preserved into analysis |
| How to transcribe | Nothing operational anywhere in the material | No procedure, convention, notation, level of detail or software |
Transcription is named, quoted about, and never taught
The deck's response to transcription is an appendix. It reproduces a search result for an article on transcription in qualitative research - the article's abstract in full, its author, and a viewing date - with no framing sentence, no question, no task and no reading instruction attached.
The abstract's argument is that transcription is under-addressed: that little attention is given to it in research training, and that methods texts continue to foreground data collection methods such as interview conduct. That is a precise description of the four slides above it.
Where this sits against the week's objective
The week states a single learning objective: to evaluate methods of data collection, including interviewing structures and techniques among six named items. The delivery audit records interviewing as only partly delivered - structures as four unreconciled types, techniques as the instruction lists above - and records the objective's verb as undelivered, because the week performs no evaluation and compares no two methods on any criterion.
What to carry forward
- Write your own one-sentence definition of the interview you conducted. The material offers four incompatible ones and settles on none.
- Six ethics obligations are required in a single list and none is defined. Bring the definitions and the approval process from elsewhere, and say where from.
- Settle the script-versus-probe rule, and the difficulty-versus-logic ordering, before the first interview rather than during it.
- Document any shorthand you invent and keep the key with the data, because the next week's comparability requirement assumes a system you can hand over.
- Nothing here tells you how to close an interview, how to respond to distress, or how to transcribe. Those are three separate second-source jobs.
Frequently asked questions
Does this material tell me how to transcribe an interview?
No. It raises transcription in an appendix that reproduces an abstract arguing transcription is neglected in research training, and attaches no procedure, convention, notation, level of detail or software to it. Choose your transcription level yourself, state it in your method, and take the convention from a source that supplies one.
Do I need consent to record, or just the respondent's comfort?
The recording slide asks only that the person feels comfortable with audio or video use. A separate slide requires informed consent and an explanation of who has access to the data, and the two are never joined. Treat recording consent as part of informed consent, get it explicitly, and record what you told them about storage and access.
What is a probe, exactly?
The material requires probes and never defines one. Its only instance is the observation that silence can be a helpful probe. It also warns against getting side-tracked without giving any criterion for when that has happened, so the boundary is one you have to set and state yourself.
Should I stay with my script or follow what the respondent raises?
The material tells you to do both. One instruction says to stay with the script if using structured questions; two others require silence as a probe and thoughtful requests for elaboration. Decide in advance which questions are verbatim and where follow-up is allowed, and write that rule into your method.
How do I take verbatim notes and maintain eye contact at the same time?
You largely cannot, and the material asks for both without noticing. In practice this is the strongest argument for recording the interview with consent and taking only sparse notes, or for bringing a second person whose job is the record - neither of which the material discusses.
Is any of this guidance usable for group interviews?
Very little of it. Every instruction here assumes one interviewer and one respondent. Focus groups are named in the week's objective and described in a single sentence elsewhere in the deck, with nothing on moderation, size or recording, so group work needs a second source throughout.
References and source attribution
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Davidson, C., "Transcription: Imperatives for Qualitative Research", viewed 30 January 2013 - reproduced as an abstract on the deck's appendix slide. The journal is not named in the source, so a complete reference cannot be constructed from what is supplied.
- Gibbs, G. R., a six-part video series on interviewing and a further video on transcription - cited on two slides with no year and no publisher, and a single address given for six parts.
- Trochim, W. M. K. 2006, Research Methods Knowledge Base - the data collection week's only listed reference, given as a bare address that is the source of nothing in the deck.
- The supplied teaching source: the week 10 slide deck on collecting data, slides on conducting interviews, preparing for the interview, at the interview, recording responses and the transcription appendix, all reproduced or summarised on this page.
Suggested questions for Ask KEVOS
- Turn these eleven at-interview instructions into a one-page card I can take with me.
- Draft the script-versus-probe rule for my interview method section.
- What ethics obligations does this material require, and which of them does it define?
- Help me write a shorthand key I can lodge with my data.
- What must I get from a second source before I can transcribe and analyse my interviews?