The relationship between harvest and survival rates of mallards: A straightforward approach with partitioned data sets

We randomly partitioned mallard (Anas platyrhynchos) bandings and recoveries from each of a number of selected reference areas into 2 groups and estimated survival and harvest rates for each area and group. This procedure produced independent vectors of survival- and harvest-rate estimates, which were used to test the general hypothesis that mallard survival and harvest rates were inversely related. We used Spearman rank correlation analysis and z-test contrasts between survival rates from years of high vs. low harvest rates. We also conducted computer simulation experiments to gain insight into the ability of these analyses to detect the relationship of interest. The data analyses suggested that survival and harvest rates of young females were inversely related, at least for the 5 areas included in the analysis. However, for young males and adults of both sexes, the analyses provided no evidence of an inverse relationship between survival and harvest rates, except possibly in a few specific areas.