I performed association study of SNPs with type 2 diabetes and diabetic retinopathy. I am confused whether I need to apply Bonferroni correction after multivariate logistic analysis. Even, for haplotype analysis is it necessary to consider Bonferroni correction? Because I am getting p value less than 0.05 but greater than threshold p value after correction in both the cases. So how to interpret? Any suggestion?
The correction would give you the possibility to control the probability of "rejecting at least one tested hypotheses under H0".
Is this important in your research?
And if so: what level would be sensible? Or: Is 5% a sensible choice?
If you screen several different SNPs for their correlation with diabetes, one can expect getting some small p-values even without assuming that diabetes is linked to any of the SNPs at all. Picking out the smallest p-values will therefore give a wrong impression of the (statistical) significance of the data. So in this case giving a corrected p-value is less misleading.
Thank you, Jochen Wilhelm, for the suggestion. So, what if p is less than 0.05 and greater than corrected threshold p? Can I say the SNP is weakly associated with diabetes?
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS