Advice sort on how to provide a more in-depth view of differences between groups

[Sharon Kibble]

I am conducting research into the changes in clinical thinking overtime between 4 groups of medical students.

They were asked to provide Mind Maps of their thoughts about a clinical case at 2 different time periods.

From these Mind Maps I have created a coding manual on all the items found present within the maps and have these codes under theme headings in two different types of  clinical thinking. These are clinical reasoning and deliberation.

On the frequencies of appearances of these codes I have conducted related samples t tests  and in some areas I have found some changes. However, I am concerned that due to small sample size these findings are limited and so I was wondering if anyone had any suggestions as to how else I could approach this large data set to report my findings.

I am looking to report the differences in the appearances of the two main themes of clinical reasoning and deliberation between maps 1 & 2 within each group and between all 4 groups.

All advice will be gratefully received.

[Robert Adams]

Your research is a qualitative study, so the small sample size should not be a concern. The design of a qualitative study is such that the results are detailed and very specific to a relatively small population from which the sample is drawn. The depth of information gathered is the strength of qualitative research. The trade-off for this detail is the small sample size, which prevents the results from being generalized to a larger population. T-tests are a statistical tool used with quantitative data. Quantitative studies have larger sample sizes that allows such tests to be useful. For a small sample, p must be extremely low, generally .01 or less, for the differences to be significant. I suggest reporting your qualitative findings in detail, and being very careful about generalizing your results to a larger population using t-tests.

[Author]

Posts