Evaluation of accuracy of digital map data via multiple comparisons

Digital map data sets (or geo-databases) are an important part of the spatial data infrastructure (SDI) of the European Community. Different methods of producing large-scale map data are described in the paper, and the aim is to compare the accuracy of these methods. Our analysis is based on statistical tools belonging to the multiple comparisons theory. The first method is the well-known analysis of variance (ANOVA), and the second one is the rank-based method. The latter approach, which is rarely used in geodetic research, allows us to determine the order of the considered methods with respect to the positional accuracy of digital map data that they produce. Using this approach, one can identify the least accurate set of map data or a fragment of a map that should be updated by a new direct survey. The rank-based methods can also be rather easily applied to other technical (engineering) disciplines, e.g. geodesy and cartography.