Association indices were originally developed to describe species co-occurrences, but have been used increasingly to measure associations between individuals. However, no statistical method has been published which allow testing the extent to which the observed association index values differ from those of a randomly associating population. Here, we describe an adaptation of a test developed by Manly (1995), which uses the observed association data as a basis for a computer-generated randomisation. The observed pattern of association is tested against a randomly created one while retaining important features of the original data, e.g., group size and sighting frequency. We applied this new method to test four data sets of associations from two different Hector's dolphin populations. The test demonstrated that populations with similar median Half Weight Index values showed clear differences in association patterns, i.e. some were associating non-randomly whereas others showed no evidence thereof. Hence, we argue that the magnitude of an index value is not useful to describe the strength of an association unless tested against a randomised data set.