I have three-diffrent groups of patient who were diagnosed with different types of cancer (aggregates). One can have more than one type. I wonder if anyone can suggest me what is the best statistical method that I can use for analysing such data.
Easy technique: Only use the data of the most severe cancer of each patient! Then analyse using Kruskal Wallis ANOVA.
BTW, what is your research interest with this analysis? Easy technique might be sufficient.
Hard technique: For analysis incorporating the repeated measures, first the data must be re-formatted indicating which observations are from the same subject. No repeated measures analysis can be done in the current form (i.e. the CxCa group needs more information added as to who had which combination of cancers). Then it will need to be analysed. My guess is that the analysis would need to be a multinomial regression or repeated measures regression but I'm not sure. Also, your sample size is small.
Dear Anan
It is important to state the aim of the statistical test. If you want to test the differences across the three groups and three types of cancers you can use MANOVA, where the sub categories of groups and types of cancer will act as the independent variable.
Good luck with your research
Dear all,
I'd like to restate that I have three-different independent groups of patients' tissues which were histologically examined for types of lymphoid aggregates. One sample can have more than one type.
So that my aim is to test the significance of different in frequencies of lymphoid aggregates in both between and within groups.
Dear Anan,
As you are comparing frequencies across groups, I think Chi-square test will fit.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques