Aside from the theoretical issues involving the validity of inferences from surveys, the basic problem of producinq unbiased estimates of regression parameters and estimates of the associated standard errors has been a particularly difficult issue in dealing with results from stratified multistage sample designs such as the one used in the National Longitudinal Study of the High School Class of 1972 (NLS). The purpose of this report is to review some appropriate available techniques that may be useful in applying regression models to the NLS data. The first section provides a framework for evaluation and an appraisal of some alternate approaches within this framework. A preferred approach (combining the Horvitz-Thompson estimator and taylorized deviation) is compared to an Ordinary Least Squares approach, through a simulation procedure using actual NLS data. The several results are summarized. Formulae underlying the preferred approach are provided separately in Appendixes A and B, and details of the development and use of a computer program to implement the approach are provided in Appendixes C and D. (Author/BW)