Comparing P value and confidence interval_Weighted data

[Mariame SY]

I am analyzing weighted data. But I notice that when comparing proportions using the Wald svy test, some confidence intervals overlap, so there is no significant difference between 2 modalities. But if I do the adjusted Wald 2 to 2 test, I notice significant differences. When I look at the power of the tests, I see values of over 20.

I'd like to know which is the best test in this case to compare weighted data.

[Rich Goldstein]

no, your understanding is incorrect - CI's can overlap and yet the p-value can be quite small - you could search this forum as this has been discussed several times; or, you could take a look at the following:

Wolfe, R and Hanley, J (2002), "If we're so different why do we keep overlapping? When 1 plus 1 doesn't make 2", Canadian Medical Association Journal, 166(1): 65-66 (not theoretical)

Schenker, N & Gentleman. JF (2001) "On Judging the Significance of Differences by Examining the Overlap Between Confidence Intervals", The American Statistician, 55:3, 182-186, DOI: [COLOR=rgb(0.000000%, 0.000000%, 100.000000%)]10.1198/000313001317097960 (somewhat theoretical)[/COLOR]

[Mariame SY]

Thanks Rich Goldstein for your help.