Samplers often distrust model-based approaches to survey inference due to concerns about model misspecification when applied to large samples from complex populations. We suggest that the model-based paradigm can work very successfully in survey settings, provided models are chosen that take into account the sample design and avoid strong parametric assumptions. The Horvitz-Thompson (HT) estimator is a simple design-unbiased estimator of the finite population total in probability sampling designs. From a modeling perspective, the HT estimator performs well when the ratios of the outcome values and the inclusion probabilities are exchangeable. When this assumption is not met, the HT estimator can be very inefficient. In Zheng and Little (2002a, 2002b) we used penalized splines (p-splines) to model smoothly -varying relationships between the outcome and the inclusion probabilities in one-stage probability proportional to size (PPS) samples. We showed that p-spline model-based estimators are in general more efficient than the HT estimator, and can be used to provide narrower confidence intervals with close to nominal confidence coverage. In this article, we extend this approach to two-stage sampling designs. We use a p-spline based mixed model that fits a nonparametric relationship between the primary sampling unit (PSU) means and a measure of PSU size, and incorporates random effects to model clustering. For variance estimation we consider the empirical Bayes model-based variance, the jackknife and balanced repeated replication. Simulation studies on simulated data and on samples drawn from public use microdata in the 1990 census demonstrate gains for the model-based p-spline estimator over the HT estimator and linear model-assisted estimators. Simulations also show the variance estimation methods yield confidence intervals with satisfactory confidence coverage. Interestingly, these gains can be seen in an equal probability design, where the first stage selection is PPS and the second stage selection probabilities are proportional to the inverse of the first stage inclusion probabilities, and the HT estimator leads to the unweighted mean. In situations that most favor the HT estimator, the model-based estimators have comparable efficiency.


Design of Experiments and Sample Surveys | Statistical Methodology | Statistical Models | Statistical Theory