I want to compare two diagnostic test with 3 different categories (mild, moderate,severe). what statistical analysis could I used?
should I join the categories into 2 and use chi square?
thanks
What I can recommend depends on what you mean by "compare two diagnostic tests." If you mean find out if they are related, then you can use a 3 x 3 Chi Square or use Spearman's rho to determine the level of association. If, instead, you want to compare the mean scores on the two tests, then I suggest using Wilcoxen's t test. I do not advise making the values dichotomous before conducting a statistical test.
I assume you want to measure the level of agreement between the two tests (and not something like test the null hypothesis that there is no association). Weighted kappa is often used for this, but there are other measures that might fit your specific context.
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Agronomy
- Crop Science
- Crop