中文摘要：

目的介绍裂-裂区设计及其方差分析。方法结合实例介绍裂-裂区设计中实验单位、实验因素级别的正确识别和合理安排,以便正确选择方差分析模型。结果只有正确识别裂-裂区设计中实验单位、实验因素的级别,才能获得正确的方差分析模型以及相应的分析结果。结论裂区及裂-裂区设计是采用区组化控制技术进行的多因素研究设计,是完全随机区组设计、拉丁方设计等的综合运用。

英文摘要：

Objective To introduce Split - split plot design and its analysis of variance （ANOVA）. Methods A case was used to illustrate the identification and discrimination of experimental units and factor levels in Split - split plot design, which is the base of correct selection of ANOVA model, Results Only if experimental units and factor levels are rationally identified and discriminated, can correct ANOVA model be selected and justfied results obtained. Conclusion Split - split plot design is multi - factor design with blocking control. It can be regarded as the synthesis of completely randomized block design and Latin-square design.