I want to test for correlation between dataset A and dataset B (both have numerical data). For example, dataset B consist of hemoglobin values which have different reference ranges for male and female. How to correlate this kind of data with data from set A? Should I use converted hemoglobin values based on gender or some other approach?
Why not split the dataset by gender and correlate each subset with A? You will get two values: one for males and one for females.
That was my thinking, but I wanted to see if it is possible to do it in some way together.
Correlation requires you to measure something for the same entity, is that what you are doing? They have to be paired in some way.
Kindly check https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3116565/
Also check http://www.sthda.com/english/wiki/correlation-test-between-two-variables-in-r
- Badges
- Science topic