Qualitative data analysis

Home Forums Methodspace discussion Qualitative data analysis

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #5269

    I am a new member to methodspace. I am a PhD part time student. I have collected data and I do not know where to start and how to analyse it. I am reading books but they do not provide enough guidance in terms of the nitty gritties of the data analysis process. Where and what do I start with? One book mentioned the use of typology, analysing chronologically and the other third method. I do not know what to choose. My data composes of algebraic tasks that teachers responded to, interviews and video tapes.

    #5271
    Diane Roberts
    Participant

    Hi

    You’ve obviously looked at the research methods literature and I’m sure will have used your ideas about the topic to inform the interviews etc so to make a decision as to how you proceed can I suggest you think about:
    1. Your research question(s). What, exactly, do you want to know? What sort of things might inform you in answering the question(s).
    2. Can all your data be analysed in the same way? Are the videos recordings of the interviews? If not, what was their purpose and does it represent a different perspective on the same points or is it something totally different? How do the responses to algebraic questions fit with the qualitative data? If this is quantitative, can the results be mapped on to the qualitative data?
    3. Are you adopting a particular conceptual approach within the study? If so, this too will indicate a way forward in establishing a basic analytical framework.
    4. Ideas will develop as you go and the process of analysis needs to be sufficiently flexible to allow you to record these and to re-analyse as relevant.
    5. Never forget that one extract, kept in context, may actually be ‘codable’ under more than one heading.

    Good luck!

    #5270
    Roger Gomm
    Member

    Diane’s advice is very sensible. But however you proceed you are going to have to generate and test out typologies. Presumably your data on teachers carrying out algebraic tasks should generate at least a three dimensional typology of the relations between different tasks, different teachers and different responses: ie between teachers type one, teachers type 2 teachers type x, and task one, task 2, task x, and response type 1, response type 2 response type x. What criteria you use to create categories will depend on your research question(s), (if you have such) – for example if you were interested in gender differences in problem solving, ‘male’ and female would be obvious teacher types, or if there is a chronological dimension then ‘before’ and ‘after’ might be appropriate (as Diane implies there might be several ways of classifying a particular dimension). Or if your interest is in the heuristics subjects use , then different heuristic approaches might constitute the ‘responses’ classification. Or maybe you don’t have a research question, then you can try finding one in trying out different typologies for your data and cross classifying them.

Viewing 3 posts - 1 through 3 (of 3 total)
  • You must be logged in to reply to this topic.