New Method Estimates Survey Errors, Improving Accuracy of Population Data.
The article presents a method to estimate variance components in sample surveys. The researchers developed formulas to calculate variances of target parameter estimates in multistage surveys. They focused on surveys where the last stage units are selected with equal probabilities. By using recent techniques from mixed linear models, they were able to account for finite population corrections in the estimates. The study found that these formulas can effectively estimate variance components in complex survey designs.