Hello,
I have a data of a retrospective study and one of the variables is the measurement of HIV viral load in patients. The data is not normally distributed according to Shapiro–Wilk test of normality and Kolmogorov–Smirnov tests of normality.
So, I am using Median (IQR). The problem is that these values are equal to zero (most patients had viral load of zero).
My question: Is it ok to write median and IQR to have zero value? Or should I use median and SD?
If I use median and SD, then I should use parametric tests, right?
No viral load is qualitatively different to amount of viral load, since zero really isn't an amount. So the concept of mean viral load is only meaningful, I propose, in those who have any viral load.
I would report the proportion who had no viral load and then, for those who had viral load, report median and IQR.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication