Multivariate analysis of variance (MANOVA) is a statistical procedure commonly used in fields such as education and psychology. However, MANOVA’s popularity may actually be for the wrong reasons. The large majority of published research using MANOVA focus on univariate research questions rather than on the multivariate questions that MANOVA is said to specifically address. Given the more complicated and limited nature of interpreting MANOVA effects (which researchers may not actually be interested in given the actual post hoc strategies employed) and that various flexible and well-known statistical alternatives are available, I suggest that researchers consult these better known, robust, and flexible procedures instead, given the proper match with the research question of interest. Just because a researcher has multiple dependent variables of interest does not mean that a MANOVA should be used at all.

中文翻译：

---

MANOVA：时间已经过去的程序？

多元方差分析 (MANOVA) 是一种常用于教育和心理学等领域的统计程序。然而，MANOVA 的流行实际上可能是出于错误的原因。使用 MANOVA 发表的大部分研究都集中在单变量研究问题上，而不是 MANOVA 据说专门解决的多变量问题。鉴于解释 MANOVA 效应的性质更加复杂和有限（考虑到所采用的实际事后策略，研究人员实际上可能并不感兴趣）以及各种灵活和众所周知的统计替代方案可用，我建议研究人员参考这些更知名的，相反，如果与感兴趣的研究问题适当匹配，则程序是稳健而灵活的。