An approximate design effect for unequal weighting when measurements may correlate with selection probabilities

It is common practice to estimate the design effect due to weighting by 1 plus the relative variance of the weights in the sample. This formula has been justified when the selection probabilities are uncorrelated with the variable of interest. An approximation to the design effect is provided to accommodate the situation in which correlation is present.