Interviewing in social research Interview overview •qualitative & quantifiable •interview or survey? •resource implications of interviewing •style of the interview -structured? -in depth? -semi-structured? •interview techniques -face to face -email -telephone -focus groups Type of data collection •Interviewing is one of a range of ways of collecting qualitative data •The information collected via interviews can also be quantified (via thematic analysis and counting up frequency of occurance of themes etc) •Important to decide early on in the research proposal whether interviews (often qualitative) or surveys (often quantitative) are the best way of obtaining data: obviously you could do both and triangulate •There are different interview styles that can be utilised, depending on whether your work is strictly ethnographic (e.g. in-depth and exploratory) or hypothesis-led (in which case you may conduct structured interviews and examine transcripts for evidence of your theory) Interviews vs surveys •Interviews can result in rich and in-depth knowledge about specific and particulars, a substantial look at the views, attitudes, beliefs, opinions, experiences of individuals or groups of people. Considerable costs involved, but excellent if there are a number of key people whose experiences or attitudes need to be explored in more detail then a survey would allow. •Surveys on the other hand reach a wide range of people cheaply, so are crucial if your sample is large, but by necessity limit and restrict the response that participants might be capable of making. Easy to manage and quantify, easy to ensure reliability and validity. Resource implications •In terms of resourcing, including time, collecting data via interviews involves: •setting up interview times with sample •designing a schedule or list of areas to be covered •travelling and or organising interview space away from others •holding interviews (tape recording, shorthand) •transcribing tapes or drafts •designing a way of coding the text •analysing the texts in order to code them •interpretation and analysis: what do the texts tell us? •presenting the information succinctly and intelligibly: this allows people to grasp headlines: bullet points if text, tables or graphs if quantified. Interview styles •Structured: a preordained list of questions, interviewer takes control •Semi –structured: a list of themes to cover, interviewer guides the content •In-depth: explores more thoroughly that coming from the interviewee Structured: to use a schedule of pre-ordained questions, some of which may be quite closed. Allows original responses (rather than choosing a survey item option that ‘fits’ – it often doesn’t!) but ensures that researcher’s objectives are covered. –useful if there is a bigger sample to interview, as take less time than in-depth interviews –researcher sets the agenda rather than the interviewee – something important might be missed because the researcher decides in advance what is to be covered –useful if you want individual responses to specific questions, a bit like an open ended questionnaire (not like a survey). –there is usually an ‘anything else you’d like to say’ option to account for this –reliability of a common approach: ensures you allow everyone a equal chance to cover the themes that you will, after all be looking for in the analysis, if your research is hypothesis driven. Structured cont. –useful if there is a bigger sample to interview, as take less time than in-depth interviews –researcher sets the agenda rather than the interviewee – something important might be missed because the researcher decides in advance what is to be covered –useful if you want individual responses to specific questions, a bit like an open ended questionnaire (not like a survey). –there is usually an ‘anything else you’d like to say’ option to account for this –reliability of a common approach: ensures you allow everyone a equal chance to cover the themes that you will, after all be looking for in the analysis, if your research is hypothesis driven. • In-depth / unstructured: – A mainstay of ethnography: •you can take advantage of things as they arise and gets a inside-out view of what it is like to be the person. •the interviewee has more control in setting the agenda - the researcher explores with open-ended questions and / or prompts, allowing free development of the interview. •in this way the interviewee’s voice is dominant, and the researcher facilitates, often by using the interviewee’s own words and phrases. In-depth interviewing cont. – In-depth interviews are particularly useful if part of the research strategy is to develop a questionnaire: in-depth interviews with a small sub-group of the sample can be useful in flagging up themes and items when developing a survey tool (questionnaire) or structured interview schedule. • In-depth interviews cont. –In-depth interviews might not be the answer outside of ethnographic research (‘what is it like to be this person?’) as there is little researcher control over the direction that the interview takes. –A costly option as it can take a long time to hold the interview, in addition to the costs involved in transcribing, coding and analysing texts of considerable length. 45 minute tape can take a proficient typist or transcriber 3-4 hours to transcribe – literally! –If the interview is conducted in order to test a hypothesis, (for example, that participation out of school hours activities positively impacts of self esteem), how can you ensure that you have covered the areas and given interviewees a chance to comment on the areas intend to analyse the text for. –Difficult to pick out common categories when coding: perhaps people didn’t mention something because they were not prompted, so again, perhaps not a good option if you are going to investigate the text for your impression of what should be there (hypothesis led). • Semi-structured: Uses a range of open-ended questions to explore attitudes or perceptions (for eg) to pre-set themes. In this way the interviewee can make original responses which can be explored in further depth • –This approach combines some of the advantages of structured interviewing (common approach, fairness in flagging up themes that you are then going to look for in your analysis) –with the some advantages of in-depth interviewing: having prompted the issue that you want to explore, you can investigate further: researcher interests AND interviewee interests –A list of themes to covered facilitates coding and analysis procedures –therefore, is frequently the preferred method. Interview procedures •Face to face: –high cost, but the best option for in-depth interviews –personal: this can be an advantage if the person needs reassurance –personal: this can be a disadvantage - social desirability effect: wanting to say the right thing (face to face is a very social interaction) –recording is straightforward: shorthand, laptop, tape recorder • Procedures continued •email: –advantages: is cheap to administer, you just both need to be online at the same time –the respondent is doing the recording – you print off –is impersonal: this can be both an advantage and a disadvantage –opportunity to explore issues further if you correspond in a Q by Q basis (otherwise is just a emailed questionnaire!) procedures continued •telephone: –one advantage is can reach a bigger no. of people more quickly –is suitable for structured interviews where there is a list of pre-ordained questions –recording needs organising: you need a head set if you are writing shorthand or tape facilities linked up to the phone –can have negative connotations with cold-calling sales people so is less intrusive if the first call is to try and arrange a convenient time to do the interview later –some issues re. contacting people at work: whether receptionists screen all calls / time etc. procedures continued •focus group: –consensus can be an issue –some people happier talking in a group, provides a focus, a joint platform –some happier speaking alone – don’t want to share their views –more difficult to transcribe – you need to use people’s names a lot to tag them on tape –can be useful if you want a general overview – Process and conduct issues •Design: objectives, variables, questions, sample and analysis •Pilot: test and revise •Preamble: introduction and information •Questioning: closed, open, scale, listening •Probes and prompts: clarification and suggestions, checking •Analysis: transcription, description, interpretation and reflexivity • To recap: •Purpose: to generate knowledge via conversation •Difference in interview is the degree of structure • To recap continued: •Validity and reliability: interpersonal, social interaction, respondent validity, triangulation with other data sources, transcription accuracy. •Ethical issues and power: consent, confidentiality, consequences, power. •Practical considerations: resourcing etc, also notes, taping, conditions, venue etc •