Abstract
Space sampling is a useful substitute for classical random sampling if the sampling universe is unknown. It is a family of sampling methods that uses geographical features such as patches on maps or geographical points. So far, the methods have been applied to problems in which the spatial dimensions of the sampling units are irrelevant (i.e., where the probability of sampling any unit is independent of its size, for example, persons surveyed with the street intercept method). If, however, space sampling is applied to problems in which the spatial dimensions of the sampling units matter, the existing evaluation procedures fail. This article describes evaluation procedures that work in such cases.
Get full access to this article
View all access options for this article.
