Three methods to measure parasite aggregation using examples from Australian fish parasites

The distribution of parasites in host populations is integral to simulations of parasite populations. We compared three methods to measure parasite aggregation, namely the Hoover, coefficient of variation and I-10 indices, using field data from a range of fish parasites. These included anisakid nematodes, trypanorhynch cestodes and polyopisthocotylean monogeneans from 10 previously published studies and one unpublished study. Positive correlations among the three methods were high. Standard errors and estimated biases obtained by bootstrap methods were smallest for the Hoover index. There was no evidence for a bias-variance trade-off. The analyses suggest that the Hoover index is to be preferred as a measure of aggregation although difficulties remain in deriving theoretical and operational definitions of aggregation.