In the statistical analysis of this paper I have some questions regarding their approach.


“Variables with non-normal distributions were either log2 or rank transformed, whichever was appropriate. WBCs and morphometric variables were standardized in all regression analyses. A rank transformation itself produces a standardized variable, but those that were not transformed or underwent log2 transformation were subsequently standardized for analysis”

Then later, when addressing Cox regressions:

” Hazard ratios (HRs) were expressed for a 1 SD increment in the distribution of each WBC variable.”

I want some clarification and my coffee is running out (and my skills as a lowly undergradute are being exposed). Are the hazard ratios expressed in the table for the variables log transformed then subsequently standardized? My understanding is that the standardizing to mean 0 and sd of 1 had two effects: helped facilitate comparisons AND also normalized the data? But My understanding is that normal variables are not necessary for Cox regression. Would like some insight. Let me know if I can clarify my question.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.