MATLAB functions to analyze directional (azimuthal) data—III: q-Sample inference

Data that represent azimuthal directions cannot be analyzed in the same manner as ordinary linear variables that measure, for example, size or quality. Although many statistical techniques have been developed for dealing with circular (directional) data, software is scarce for the unique calculations. This paper describes a MATLAB script, located on the IAMG server, that performs analyses that are analogous to linear-data one-way ANOVA, but on groups of azimuthal data. The data are restricted to vectorial, not axial, directions. Calculations include tests of hypotheses that the groups’ vector means are all equal, and that the groups’ concentrations of vector directions about the vector means are all equal. These tests use large-sample approximations or resampling methods.