I have a data set containing data of patients with cancer. There are 60 subjects in this study, from which 30 got the standard treatment of chemotherapy, and 30 got chemotherapy along with nutritional supplements such as Turmeric and others.

For each subject, the white blood count level was measured at 2 time points, I will call them T1 and T2.

Chemotherapy reduces the level of the white blood count, and the aim of the study is to show that in the treatment group, the decrease is more moderate, that the level of white blood count did not reduce as much as in the control group.

I have a question regarding the best analysis:

Which of the following should be the dependent variable:

  1. The WBC level at time 2
  2. The absolute change from baseline (WBC2 -WBC1)
  3. The relative change from baseline (WBC2/WBC1)

Choosing the first option, doesn't account for different baseline levels, and therefore leads to ANCOVA with baseline as covariate. Same goes for option 2. Option 3 fixes the different baseline levels problem, but it doesn't feel right to me. I read some reviews against relative change from baseline.

One more thing, the data is not normally distributed, it has a right skewed distribution. With a sample size of 30 for each group, how robust is the normality assumption for t-test and ANCOVA?


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.