Hi, I have a qualitative data and would like to apply statistical analysis or any credible data testing technique to test or validate the result. For instance, my investigation is identifying constraints to service delivery and a host of constraints, say about 12, have come up and there are a number of case studies on similar subject matter and I want to compare them against other research findings for validity and credibility. I need guidance on the type of statistical analysis or test I would carry out to validate and strengthen my result.
If you only have qualitative data, there is no basis for quantitative analysis. Use the information you have to design a quantitative study. Attempt to constrain the study to as few parameters as possible. If you want to compare qualitative data from different studies you have no basis for comparison as you have no basis for ordering the results from the different studies.
I have been working on the cyber risk modeling where we convert qualitative data into quantitative for the purpose of analysis. Please do write to me with more details and perhaps I can give you some pointers.
There is a tool available that allows researchers to test their finding for credibility and trustworthiness. so far it has been working fine.
My research is basically about investigating barriers to service delivery in Public Private Partnerships. I have discovered about 11 or 12 barriers ranging from corruption to limited capacity to inexperience financial market. I have been searching for the right model or tool to enable me test the efficacy and credibility of the findings. Hope this helps and I look forward to hearing from you soonest.
I think you need to provide more detail. Qualitative has at least two completely different meanings in this context. Some statisticians use it to refer to categorical data - often dummy coded as 0/1 etc. Social scientists and most others use qualitative data analysis to refer to a range of approaches that focus on meaning using narrative, text, images etc. directly rather than quantifying them.
For statistical advice you need to give an idea of the structure of the data how you code constraints and so forth. It is possible you could use techniques such as logistic regression, latent class analysis or even simple graphical methods.
First of all, the research was structured to extract informants opinions through a semi-structured interview. A narrative analysis then followed to relate to the experiences of respondents by narrating their views, and real life situation on the subject that was under investigation. Prior to the interview, I handpicked three themes from the literature reviews and adopted it as a conceptual framework that guided me along the lines of 6 questions. The themes included Factors and barriers ;Service delivery and enabling environment. I split the findings sections into three parts, representing each theme in order to come up with a clear and concise findings. The findings, as I indicated earlier, ranges from corruption, limited institutional capacity to weak financial management system. How to validate and test these findings is now my concern.
Do the findings suggest a structure for a quantitative investigation? The only way to qualify your results to make them more quantitative is for you to assign values based on your judgement.
I am not aware of any statistical tool for testing validity in qualitative data. I think you have the option of critical review of literature and compare - contrast your findings with what is already known. Your results may be your contribution to knowledge on the subject matter.
I'm aware with Sogunle. Acritical review maybe is the best practice.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis