• experimental design;
  • modelling;
  • requirements;
  • statistical analysis

A limited survey of published reports on dietary nutrient requirement estimates for fish (three journals, 46 papers) indicates that broken-line analysis and analysis of variance ( ANOVA) are often used to estimate nutrient requirements from dose–response data. The application of regression models using published treatment mean values to re-evaluate estimates was possible using 33 of these reports. Re-evaluation suggests that the broken-line method and ANOVA frequently underestimate the requirement. Regression produced estimates that averaged approximately twice, but were up to five times the published requirement. Additional problems that prevented re-evaluation or produced errors in the original estimates were: failure to include nutrient levels high enough to produce a maximum response, failure to space nutrient input levels closely enough to adequately model the dose–response relationship, an apparent failure to screen data before analysis, and insufficient model diagnosis. Examples from the literature are presented to illustrate how design, method of analysis and the choice of model affect the requirement estimate. The effects of measurement frequency and the experiment duration on the resulting requirement estimate are discussed. A set of protocols is presented to help improve nutrient requirement estimates.