Dear all,
I plan to develop a tool specific to a clinical population using a sequential exploratory mixed method design- the first phase will be Qual (FGDs and in-depth interviews and literature review). Second Quan phase will be focused on constructing the instrument and psychometric properties.
However, there isn't any specific theory from which I can entirely borrow the tool items or even define the variables. So I plan to consider different existing theories for item generation and definition of my variables.
I would like to know if I need to develop a particular model based on my qualitative analysis to build my tool on. If yes, how can it be done? If not, do we call it a pan-theoretical scale or what?
Eager to get your valuable inputs!
Thanks :)
The design of the qualitative component in a qual --> QUAN design should be driven by the needs of the QUAN component. So, if the desired QUAN instrument is not theory-based, then there is no reason for the qual study to be theory-based.
This is thinking ahead but may be helpful for you to think of when you are constructing this tool. Once you have a quantitative tool, you want to be able to test the reliability and validity of your instrument. This involves various forms of reliability (e.g., test-retest, internal consistency), and validity (e.g., construct, content and criterion). In order to have good construct validity, you need to compare the test to other tests that measure similar qualities. As such, you will have to review previous instruments and identify the ones that you think would be most relevant. Looking over these can also allow you to identify strengths and weaknesses and identify areas of improvement you would want to integrate in your own instrument. Good luck!
Hi there,
There is nothing wrong with taking/applying constructs from various models (more than one) towards building your own model. However, if you really wish to develop a model and claim it as a new model, it would usually come from the quantitative results rather than the qualitative. This is because it would require significant testing such as Structural Equation Modeling to confirm if the model would work. You can however use qualitative statistics to support the quantitative results/model for added validation.
Sachin Suknunan I think that when one is building a new quantitative instrument, it will always be necessary to assess the reliability and validity of that instrument, regardless of whether it is devloped from qualitative research, and regardless of whether it based on theoretical models.
Shikha Ahuja with agreement to all of the valuable answers above, as it sounds from your question like you want to develop one or more new quantitative instruments for a specific population group (clinical in your case), first you must focus on your phenomenon of interest or variables of interest especially drawing content (definitions) from relevant literature and then can conduct focus group discussions to clearly define the constructs and to generate a pool of scale items. After generating an agreed upon definition based set of items you can collect initial data for reliability-validity tests such as EFA, concurrent and incremental validity etc. In most of the cases researchers take mixed-method and multi-method as interchangeable terms, which in my opinion is wrong as triangulation of the results in mixed method is really hard. In short, you need relevant literature, expert opinion, and many more while developing new definitions and measurement scales. Tq.
Shikha Ahuja one most important issue is that the scope of a qualitative inquiry is very subjective and too broad while a quantitative research model is always developed with some quantitatively measurable constructs chosen based on a solid theoretical foundation. Tq.
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments