How much should interviewer speak in a fine interview?

Speak as minimal as possible on the interviewer's part & ask open questions to solicit maximum sharing from the interviewee. The objective is to collect quality data from interviewee & our open questions are to facilitate / guide them to share more. Also finding the time spent by interviewer in talking can gradually reduced through performing more interviews.

My view - Minimum possible. The interviewers questions must be concise & clear so as to convey to the respondent what exactly he is asking. After this, the interviewer needs to speak only if there is inadequate/ incomplete response from the respondent. That's the only way to ensure getting reliable information.

Interviewer can get more out of an interview when the interviewee is relaxed. Therefore first he/she should put the candidate/interviewee at ease in an interview by proper greeting and smile.

Then the interviewer should try to get more from the candidate. I agree with Emmanuel Vijayanand Murray and Han Ping Fung. Interviewer should talk lesser than the candidate.

Thank you guys. So I guess we all agree NECESSARY MINIMUM is the correct answer. But is it possible to place this qualitative issue into more quantitative perspective? Has anyone studied it on a decent sample of interviews and could give % value for NECESSARY MINIMUM and ACCEPTABLE MAXIMUM?

No, I don't think you can define a desirable minimum or acceptable maximum, but I certainly agree with the "as little as necessary" standard. What matters most, however, is what the interviewers said and whether this had an impact on how the participants responded. For example, if the participant did not have an initial answer to a question, did the interviewer ask leading questions in order to produce a response?

One thing that makes a big difference in getting participants to talk more is establishing rapport in the beginning of the interview. I find that the best strategy for this is to ask a first question that is both easy for the participant to answer and interesting to the participant. Too often, researchers rely on questions that are too narrowly centered on their own interests, so that participant are either confused or bored.

The following sources should also be helpful, particularly: listen more, talk less, and ask real questions (Seidman, 2006, pp. 78-93); suggestions/preparation for interview (Turner III, 2010; Doody and Noonan, 2013).

• Doody, O. and Noonan, M. (2013). Preparing and conducting interviews to collect data. Nurse researcher, 20, 5, pp. 28-32.
• Jovchelovitch, S. and Bauer, M. W. (2000). Narrative interviewing. In: Bauer, M. and Gaskell, G. (eds.) Qualitative Researching with Text, Image and Sound: A Practical Handbook for Social Research. England: SAGE Publications, pp. 57-74.
• Seidman, I. (2006). Interviewing as qualitative research: A guide for researchers in education and the social sciences, New York, NY, Teachers college press.
• Turner III, D. W. (2010). Qualitative interview design: A practical guide for novice investigators. The qualitative report, 15, 3, pp. 754-760.

I guess my contribution would be:

1. always ask open ended questions so there is not a yes/no response

2. ask for 'feelings about' or 'emotion when' as well as ideas ie thinking. Not because I am opposed to thinking but feeling type responses evoke the recollection of the event more thoroughly that thinking.

3. two ears & one mouth -- a good guide for who speaks how often.

Although, the intro section might be more wordy from interviewer to start the ball rolling.

4. Often in a pause the participant is thinking .... so let them think. If re-wording or redirection is needed, a word or even a gesture might suffice rather that repeating the whole question/paragraph.

5. The whole purpose of an interview is to gather data form participant, hence the talk of the interviewer is actually wasted time [OK, not too serious, but you get my point?]

6. Lastly, what might your reaction be is you be interviewed and you have something to say: .. how might you want to be heard?

There are no quantitative rules. however, the interviewer should intervene as little as possible. His/her function is to generate rapport first and then get a fluid flow of information from the interviewee.

There is no absolute rule on how much the interviewer should intervene; be guided by how the interview is going relative to researcher's purpose in conducting it---whether quantitative or qualitative.

It's a judgement-call.

Indeed, the "as little as necessary" approach is the most common. However, in the end it very much depends on your research question and methodology of choice. There are methodolodies wherein one must be as silent as possible and those wherein one should be guiding and participate to get the information needed. Luckily, most published method guides specify their approach to data gathering, including interview participation.