Closed acoppock closed 7 years ago
Need to consider implications for inferring designs in declare_ra or declare_rs
This would be great, if it doesn't screw up declare_ra/declare_rs as you raise.
fixed in 51d273366b016a786b0037342d3b9b08e43ac9c1
strata_var <- rep(c("A", "B","C"), times = c(50, 100, 200))
Z <- strata_rs(strata_var = strata_var, n = 20)
table(strata_var, Z)
with result:
> table(strata_var, Z)
Z
strata_var 0 1
A 30 20
B 80 20
C 180 20
Proposed syntax for sampling
10 As, 10 bs, 10 cs.
for random assignment
treat 10 of 50 As, 10 of 100 Bs, etc.