Methods and Issues in data collection

Methods and Issues in data collection

1. Five major methods: Interviews, video recording, media data, internet data, published texts

   • video recording and interviews not the same thing, analyse it differently

2. Think about how the answer to the question will be derived from the data AND how data will be analysed

3. Semi-structured interviews: the most common means of data collection

   • used to explore how people talk about almost any topic, but esp where interest is in accounting, understanding, constructing, or views. e.g. Widdicombe & Wooffitt (1995) interviewed young people about identity Easy to do allow probing follow up and numbers of participants need not be large

4. Two kinds of interviewing

   • 1. Traditionally, empathic interviewing has been used as a means to access info (hugh jones 2010) lang is a vehicle to express what is inside 'true' attitudes/experiences 2. But qualitative researchers adopting a social constructionist methodology, treat lang (hence interview) as a medium of construction and through which objects, experiences etc are constituted (in terms of their meaning) Some also treat interview as interaction or co-consturcuted (is as active as the participant)

5. Designing s-s interviews: always be well prepared!

   • Be clear about what you want to know. Q address these aims. order youll ask Q's. Q's clear focused, non leading. Ask yourself if you could answer them. Think of responses and how you would handle them (pos neg, what do you mean?). add probes and follw up q's, (can u tell me any more?)
   • Appropriate Q's
     • What kinds?: neutral, non directive (open, get people to talk, produce accounts) aim of q's stimulate thinking generate accounts. Avoid jargon use familiar lang (everyday terms, non threatening). Prepare general chat/intro to put particpant at ease Add general Q at end inviting them to say if theres anything they want to add
   • Preparing for interviews
     • Practical issues: Pilot and learn schedule, speak properly. Familarise yourself wit recoridng equipment. audio or video. Don't panis if not according to plan in interview
     • Potential issues
       • participant rambles or goes off topic (ask them to summarize, thank them).  asks a Q or advice (leaflets) is silent (give time then move on) gets upset (tissues) interrupts Q's may have to be simplified or avoid probing Online interviews (evans, et al 2008) + overcome geographical/mobility issues. anon for senstive topics. low cost, more equal - loss of non-verbal communication. access to/competacne in using tech. difficulties establishign rapport, miscommunication more likely

6. Focus groups ex (Guise et al 2008)

   • increasing popular alternative to individual interview Researcher acts as moderator 'steering' convo advan of mobilising participants to respond and comment on others contribution so statements can be challenged, extended, undermined tch richer data esp useful if itnerest is how people jointly construct or argue abotu issue disadvan if topic is sensitive and particpants feel uncomfortable and transcription if harder
   • 4-6 people some advan in knowing each other. Prinicples of constructiong Q's for s-s interviews apply here too; audio or video

7. Audio or video recordin (naturalistic data)

   • Some studies record talk-in-interaction (everyday, institiutional, telephone, face-to-face). this is nauralistic data (Potters 2004 dead scientist test) Useful where interest is in actual interactions or how things get done or work (e.g. therapy) Increasingly preferred in discursive research  But accesibility is often an issue and Q's arise about the presence of the recording device (Hutchy & Speer, 2003)

8. Natural vs Contrived data?

   • Opene nd interviews have been the most popular means of data collection  But one problems is that interviews Q's are driven by social sciene agenda (not partipants concerns) (potter & Hepburn, 2005) Theirs analysis is more complicated because various footings (or identities) are relevant The results can't be easily extrapolated to real life situations 9accoutn of discrimination not the same as observing it as it happens)

9. Benefits of nauralistic data (wiggens & Potter 2008)

   • 1. avoid imposing researchers categorsies/assumptions  2.research situated within settings of everyday life 3. no need to try to apply finding from interview setting to other setting 4. research can be guided by participants issues surprising topics) 5. analyses life as it happens in normal situations
   • Responses to them
     • Not necessarily problems (Hollways 2005), can avoid using social science categories in interviews. Fotting, stake are topics for analysis. Depends on topic: sensitive issues may require interviews. some topics (national iden) rarely mobilised in everyday talk Depends on purpose (Willig 2001): no wrong or wrong methods just more or less appropriate

10. Comprimise?: treat interviews as interactions

    • In which it is as important to analyse the interviewers Q's as it is to anlyse responses e.g (speer 2001)

11. Media data

    • TV, radio talk(naturalistic and institutionalised, newspapers (textual data) Used to address issues: How particular issues or people are represented or constructed management of identity of accountibility (eg. wallwork & dixon 2004, construst of britshness to argue again fox hunting in paper Requires a systematic, feasible and manageable means of data collection (type, time period) on a particular topic.  Online databases can be used Think carefully about kinds if claims you can make on the basis of this data
    • Internet data: examples
      • Increasinly popular, wide range of data: online interviews, discussion thread on forumsie.e vegans negotitate health issues (Sneider & teMolder 2005) Commentaries of newspaper articles
      • Issues
        • useful for accessing people who might otherwise be hard to get or inaccessible data Data should be collected systematically and practically ie random sample, time period Think about effects for analysis of sequencin issues Ethical issues need carefu lconsiderstion
      • Ethical issues
        • Blurs boundaries between public and private (accessible but doesn't mean public disclosure) Problem of gaining informed consent where identity and email hidden by avatars Need to think carefully about possible hard of study (withdrawal from online support groups) Preserving anonymity and ensuring that data is not traceable (e.e. via search engine) Bower & Tuffin 2004) blurs private/publics and protecting identies difficult

12. Ethics (all qualitative)

    • 1. Seek informed consent 2.Avoid deceptions 3. make sure participants know they can withdraw 4. Prepare a constent form 5. Debrief and offer a summary report 6. Mainain confidentiality

13. Politics of research

    • Whose interests wil be served?, how wil lthe knowledge i produce be used? Some argue that qualitative research should deliver benefits for participants (not just protect) eg in challenging social injustice

14. Choosing method

    • Carefully consider topic and purpose of study  Consider strengths and weaknesses of methods (inrelation to Q) Choose methods best fits aims How answer to Q will be deived from the data specific ex/ issues:interviews Think about how the data will be analysed