The utility of a variety of objective, quantitative and robust methods of assessing similarities between antenna measurement data have already been highlighted in a number of recent publications (1, 2, 3). These techniques involved the extraction of interval and ordinal features from the data sets that can then be effectively compared to establish their adjacency. However, frequently such is the volume and complexities of the data involved that a single comparison methodology is inadequate to effectively classify all types of data. Within this paper we intend to compare and contrast several techniques for obtaining a quantitative, holistic, measure of similarity between data sets as well as introducing a new hybrid technique. In addition to more conventional interval techniques, e.g. crosscorrelation coefficient, two newer more sophisticated techniques will be presented. These are an ordinal and an interval-ordinal technique. In addition to these newer statistical image classification techniques, a novel hybrid categorical ordinal technique is developed that retains the advantages of the interval-ordinal technique but removes the requirement for interpolation and facilitates the comparison of two, or higher, dimensional data sets of differing sizes. These techniques will be illustrated with reference to a number of data sets that will be examined assessed and classified to obtain measures of adjacency that relate global features of the data sets. This data is derived from the output of partial scan techniques (4) that attempt to reduce truncation errors in planar near field antenna measurements by the construction of bespoke polyhedral sampling surfaces that aim to enclose all the current sources.