topic Re: Direction of significance for continuous variables in Statistical Procedures

Direction of significance for continuous variables

GS2 — Tue, 02 Jun 2020 13:15:43 GMT

Using SAS 9.4

I have 2 continuous variables and am reporting a p-value via a wilcoxon ranked sum test due to the non-parametric nature of my data. I need to report the direction of significance as well. What is the best method for programming this in SAS? Thank you

My code for the ranked sum test is as follows:

proc univariate data=have normal plots;
var diff_flex; *signed rank;

run;

Re: Direction of significance for continuous variables

SteveDenham — Tue, 02 Jun 2020 14:32:06 GMT

A bit confused about the use of PROC UNIVARIATE in this context, It doesn't appear that you have a CLASS statement, which indicates that you aren't comparing two groups using the ranked sum statistic. That would probably mean that you are testing against a defined value, which UNIVARIATE implements with the mu0 option. If that is the case, then the sign of the statistic will indicate the direction. Note that the p values presented are two-tailed and do not consider directionality,

SteveDenham

Re: Direction of significance for continuous variables

Reeza — Tue, 02 Jun 2020 15:57:57 GMT

Are you sure you posted the correct code for your signed rank test?

That doesn't look correct to me.