Probability vs. Nonprobability Sampling: From the Birth of Survey Sampling to the Present Day

At the beginning of the 20th century, there was an active debate about random selection of units versus purposive selection of groups of units for survey samples. Neyman’s (1934) paper tilted the balance strongly towards varieties of probability sampling combined with design-based inference, and most national statistical offices have adopted this method for their major surveys. However, nonprobability sampling has remained in widespread use in many areas of application, and over time there have been challenges to the Neyman paradigm. In recent years, the balance has tilted towards greater use of nonprobability sampling for several reasons, including: the growing imperfections and costs in applying probability sample designs; the emergence of the internet and other sources for obtaining survey data from very large samples at low cost and at high speed; and the current ability to apply advanced methods for calibrating nonprobability samples to conform to external population controls. This paper presents an overview of the history of the use of probability and nonprobability sampling from the birth of survey sampling at the time of A. N. Kiær (1895) to the present day.