The effect of photoperiod and prey density on long‐term survival of larvae—a data analytic approach where no parametric model is assumed and individual follow‐up is not available

A data analytic approach for analysing differences in long‐term survival and identifying treatment combinations which provide high survival in a factorial design set‐up is presented. The methods are particularly appropriate when individual follow‐up data are not available and no simple model fits the data. Upper quantiles of the survival distributions are used as the response variable and within and between‐treatment differences are analysed through both a structured ANOVA model and subset‐selection procedure. The methods are used to study the effects of several photoperiods and prey density treatments on the long‐term survival of the larvae of gilthead seabream, Sparus aurata. The two approaches used are shown to lead to similar conclusions.