18th January 2012 at 7:48 pm #2831manal saeedMember
would like to ask about mixed methods ,is it possible to use explanatory sequential methods when you want to look at perception of patient regarding hospital care?? . So first will distribute satisfaction questionnaire and then follow it up with semi structure interview ,, This my plan but with more reading I find out that it should be exploratory since it is looking at patient experience of hospital care and must be the first phase interview then the second phase questionnaire ??
Appreciating kind advices and help
mino26th January 2012 at 5:54 pm #2836Muhammad Mazhar KhanMember
To my shallow knowledge, a study best starts from the understanding of the researcher – it is he RQs that decide which methods/instruments to use and how to approach a research setting and population/subjects.
I doubt if there is any definite “design” that requires 1st interviews and then questionnaire. What if the researcher is trained only in the qualiTATIVE research. She cannot use questionnaire to be statistically analyzed.
Thus, honestly, I would suggest that be open to the research context. Look at your AIMS of the research more closely. You may come up with the best route into the patient’s opinions/attitude you’d want to explore.
Mk26th January 2012 at 7:50 pm #2835manal saeedMember
thanks for the reply27th January 2012 at 12:20 pm #2834Sharon CahillMember
Do some reading try Cresswell book on mixed methods … There are some great explanations and diagrams that you might be able to use to solve your problem….any method book or even the journal of mixed methods should demonstrate different approaches to this problem. You need to be clear what it is exactly you are asking and why … Again this will help answer your question. Why not use a flow diagram yourself to start with? Start with the main research question … Is it exploratory or explanatory, what question comes next … Is it logical … An important aspect of mixed methods research is ensuring that you integrate the ‘results’ again this might give some clues to what should come first. Good luck27th January 2012 at 10:37 pm #2833Laurence MoseleyMember
No, it’s the other way round.
Do the semi-structured interviewing first to discover what might be the case. Then do the development work necessary to produce something more structured, and finally do that structured work to try to see whether what you suspected might have been the case really is the case. The serious and critical work comes in that middle development phase. I usually reckon that, unless you are using a well-tested questionnaire (and there are not many of them around), you should allocate at least 30% of the project time to that phase. I have seen projects in which over 60% of the time was devoted to that phase, but in those cases we were fairly confident that we were accurately measuring the phenomenon concerned. It is very easy to mislead yourself and your readers. Even apparently trivial research problems usually have a search space of the order of 10^6 or 10^7 while your unaided brain can handle only about 7 items at once i.e. 10^0.
The greatest danger is thinking that you have found what you hoped to find before you started, and designing research should, at the very least, protect you against that danger.27th January 2012 at 11:33 pm #2832Pat BazeleyParticipant
What you propose is fine – as what you are seeking to do is explain (or at least, better understand) their level of satisfaction by talking to them about it afterwards. While you would want to pilot test your survey, to do a major qual component first and then a survey to follow is a design for generalisation of findings from the qual, using a survey will only help you identify the levels of satisfaction, it will not explain why they were or were not satisfied.
You might find the following paper interesting – Green and her colleagues did the survey and interview at the same time and found some interesting discrepancies in the answers given. You can download it directly from the site:
Green, J., Statham, H., & Solomou, W. (2008). Assessing satisfaction: insights from a multi-methods study. Journal, 462. Retrieved from http://eprints.ncrm.ac.uk/462/
- You must be logged in to reply to this topic.