Piece of advice for anyone beginning a research project: if your supervisor says anything to you that sounds even remotely like, "that's really ambitious" or "you may want to focus ..." or "why do you want to collect...", you should stop and think about what you will need to do when you begin your data analysis.
I am conducting a mixed-methods (I have found Creswell to be a great beginning primer on this), cross case research project. I am still working through the survey data that I gathered from one case study. Ugh! Part of the challenge I am having is that I hadn't really thought through all the interrelationships that I might find - to give myself a bit of a break, how could I have possibly known everything that would come up in an exploration, right? Interesting trends are appearing, and I am now fearful that my approach of "the more data the better" may lead to "paralysis by analysis".
I recently came across an article by Sullivan and Artino (2013) as I was trying to find a very clear explanation of working with Likert-type scales. In addition to the discussion of parametric tests for Likert-type scales, their parting thoughts on researchers taking the time before they begin their studies to explicitly consider which tests will be done on data (e.g. ttest or Mann-Whitney U) are really good food for thought.
I am not a statistical expert, but I'm learning. And boy-oh-boy do I have enough data for practice.
References
I am conducting a mixed-methods (I have found Creswell to be a great beginning primer on this), cross case research project. I am still working through the survey data that I gathered from one case study. Ugh! Part of the challenge I am having is that I hadn't really thought through all the interrelationships that I might find - to give myself a bit of a break, how could I have possibly known everything that would come up in an exploration, right? Interesting trends are appearing, and I am now fearful that my approach of "the more data the better" may lead to "paralysis by analysis".
I recently came across an article by Sullivan and Artino (2013) as I was trying to find a very clear explanation of working with Likert-type scales. In addition to the discussion of parametric tests for Likert-type scales, their parting thoughts on researchers taking the time before they begin their studies to explicitly consider which tests will be done on data (e.g. ttest or Mann-Whitney U) are really good food for thought.
I am not a statistical expert, but I'm learning. And boy-oh-boy do I have enough data for practice.
References
- Creswell, J.W. (2009). Research design: qualitative, quantitative, and mixed methods approaches. Thousand Oaks: CA, SAGE Publications.
- Sullivan, G. M., & Artino, A. R. (2013). Analyzing and Interpreting Data From Likert-Type Scales. Journal of Graduate Medical Education, 5(4), 541–542. http://doi.org/10.4300/JGME-5-4-18