ABSTRACT
This chapter describes probability and nonprobability sampling techniques and the methods of inference that can be used for both. It reviews techniques used in probability sampling, particularly for official statistics. The chapter enumerates the types of nonprobability samples, based on a classification by the American Association for Public Opinion Research that practitioners have used. Since nonprobability samples are often obtained in a poorly controlled or uncontrolled way, they can be subject to a number of biases when the goal is inference to a specific finite population. The chapter reviews the fundamentals of probability sampling, including key design features (e.g., stratification) and traditional methods for inference. It provides some of the details of approaches to estimation with nonprobability samples, including how information from probability samples can be used to potentially improve inference from nonprobability samples. The chapter focuses on surveys of households or persons, but the issues raised apply to other types of units like businesses or schools.
