Quality and reliability of household survey data

Abstract

The quality of statistics derived from a survey is determined by a number of factors, such as accuracy, their relevance to the needs of users, and timeliness of the results. This paper does not aim at dealing with all these factors, but concentrates on some of the errors which affect the accuracy of the results and more specifically on the sampling, non-response and response errors. In doing so, the main purpose is not to describe them in detail but to provide the conceptual framework for the Annex in which some guidelines for quality reporting are presented. There are a number, of excellent references which deal in-depth with this subject matter a list of some of them is given at the end of this paper.