Note: [1] Hubert M.P. Kersten studied mathematics at the State University of Leyden with a major in mathematical optimization methods and economics. After his doctoral examination in 1976 he joined the Department for Statistical Methods of the Netherlands Central Bureau of Statistics (CBS) as a research associate, working in the field of sampling theory.
Note: [2] Jelke G. Bethlehem studied mathematics at the University of Amsterdam with a major in mathematical statistics. After completing his examination in 1974, he worked as a research associate at the Mathematical Centre in Amsterdam. In 1978 he started to work at the Department for Statistical Methods of the CBS where he is engaged in statistical consulting as well as research in the field of multivariate statistics and sampling methods.
Note: [3] The views expressed in this paper are those of the authors and do not necessarily reflect the policies of the Netherlands Central Bureau of Statistics. The authors wish to thank the Department for Statistics of Income and Consumption for the data collection and J.G. Thijssen and J. van den Ende for their valuable comment on a preliminary version.
Abstract: The Basic Question Procedure (BQP) is a method to explore, estimate and reduce nonresponse bias. It consists of asking one question, the basic question of the survey, to nonrespondents who really cannot be persuaded to participate in the survey (refusal, not-at-home, illness, etc.). Estimates for all nonrespondents can be calculated by stratifying, according to type of nonresponse, the nonrespondents who answer the basic question. In this way insight is obtained in the bias which may affect the survey estimates. The method was applied in the Dutch Housing Demand Survey 1981.
