A design effect for the mean of a variable = 1.144 will the variance be under clustering be smaller or bigger than under a simple random sample
Given the calculated design effect, will the standard errors under clustering be smaller or bigger under simple random sample design?