方差分析的数据要符合正态分布

library(rstatix)
iris %>%
  group_by(Species)%>%
shapiro_test(Sepal.Length)

## # A tibble: 3 × 4
##   Species    variable     statistic     p
##   <fct>      <chr>            <dbl> <dbl>
## 1 setosa     Sepal.Length     0.978 0.460
## 2 versicolor Sepal.Length     0.978 0.465
## 3 virginica  Sepal.Length     0.971 0.258

非常严格，不通过的话，需要转换或非参数方差分析

方差分析的数据要符合方差齐性

```r
library(rstatix)
iris %>%
  # group_by(Species)%>%
levene_test(Sepal.Length ~ Species,center = median)
```

```
## # A tibble: 1 × 4
##     df1   df2 statistic       p
##            
## 1     2   147      6.35 0.00226
```

```r
iris %>%
  # group_by(Species)%>%
levene_test(Sepal.Length ~ Species,center = mean)
```

```
## # A tibble: 1 × 4
##     df1   df2 statistic        p
##             
## 1     2   147      7.38 0.000882
```

要求不严，不通过的话，可以进行方差不齐的方差分析

By default, median, provides a more robust test.

其他注意事项

异常值要按规定进行处理

重复次数至少要达到3次

方差分析与事后比较

In a scientific study, post hoc analysis (from Latin post hoc, “after this”) consists of statistical analyses that were specified after the data were seen. They are usually used to uncover specific differences between three or more group means when an analysis of variance (ANOVA) test is significant–FROM Wikipedia

永远先做方差分析，然后再做事后比较。

事后比较方法选择

官方说明

多重比较与齐性子集

The relevant literature seems to support the conclusion that the tests available in the multiple comparisons table generally have better properties than the homogenous subset tests and are the preferred focus for post hoc analysis.