Dear Listers,

I am considering using the -mhodds- command for an expvar with more than 2 levels in Stata 15. The help file states that in this case, -mhodds- calculates a 1-degree-of-freedom test for trend.

Is this the same as (or equivalent to) the Cochrane-Armitage test for trend? I am asking as the study was powered for a trend analysis using the -power- command which uses the Cochrane Armitage test.

Thanks,
Laura