I need to test if the difference between male and female program coverage rates (number treated/target pop) is more than 5%.
The null and alternative hypothesis are set as follows:
Null hypothesis: Male coverage rate – female coverage rate = ± 0.05
Alternative hypothesis: Male coverage rate – female coverage rate (not euqal to) ± 0.05
Can I use the following Stata command?
gen comp=.05
gen diff =abs(pcvg_m - pcvg_f) // pcvg=program coverage rate; m for male and f for female
prtesti target_pop diff comp
local p = r(p)
gen pval=`p’
Related Posts with Testing if the difference between two rates is more than 5%.
xtlogit and excluding ineligible participants- Dropping every ID when respondent go from "work" to "retirement"
- Is there a way to use panel data rolling regressions with driscoll-Kraay standard errors?
- Obtain same results with regress and rifhdreg
- Coefplot: Displaying each level of a factor variable in one plot
Subscribe to:
Post Comments (Atom)
0 Response to Testing if the difference between two rates is more than 5%.
Post a Comment