Peltier M R, Wilcox C J, Sharp D C
Dept. of Dairy and Poult. Sci., University of Florida, Gainesville 32611, USA.
J Anim Sci. 1998 Mar;76(3):847-9. doi: 10.2527/1998.763847x.
In the use of ANOVA for hypothesis testing in animal science experiments, the assumption of homogeneity of errors often is violated because of scale effects and the nature of the measurements. We demonstrate a method for transforming data so that the assumptions of ANOVA are met (or violated to a lesser degree) and apply it in analysis of data from a physiology experiment. Our study examined whether melatonin implantation would affect progesterone secretion in cycling pony mares. Overall treatment variances were greater in the melatonin-treated group, and several common transformation procedures failed. Application of the Box-Cox transformation algorithm reduced the heterogeneity of error and permitted the assumption of equal variance to be met.
在动物科学实验中使用方差分析进行假设检验时，由于尺度效应和测量的性质，误差齐性的假设常常被违反。我们展示了一种转换数据的方法，以便满足（或在较小程度上违反）方差分析的假设，并将其应用于生理学实验数据的分析。我们的研究考察了褪黑素植入是否会影响处于发情周期的母马的孕酮分泌。褪黑素处理组的总体处理方差更大，几种常见的转换程序均未成功。应用Box-Cox转换算法减少了误差的异质性，并满足了方差齐性的假设。
