Open d-callan opened 2 years ago
@d-callan is this specific for mbio? if there aren't already, i can see future studies where a subset of samples were tested by one assay and not the other (for example, if the lab ran out of the sample). It would be a bad idea to impute 0s for this case
It's per variable/ collection. I imagine clinepi should be fine.
I think if its possible without giving ourselves a major headache we should have histograms in the subsetting tab also impute 0s for NULL values. this would get us consistent across subsetting and viz tabs.