dplyr::mean_sd_se 统计量的计算

测试数据

data <- data. frame(label = sample(LETTERS[1:5],500,replace = T), value = rnorm(500))

1、根据公式计算

data %>% group_by(label) %>% summarise(mean = mean(value),sd= sd(value),se = sd(value)/sqrt(n()))
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
# A tibble: 5 × 4
  label    mean    sd     se
  <chr>   <dbl> <dbl>  <dbl>
1 A      0.148  1.01  0.108 
2 B      0.0789 0.946 0.0936
3 C     -0.0888 1.02  0.0998
4 D     -0.102  0.958 0.0983
5 E      0.0893 0.823 0.0785

2、rstatix 软件包计算各数据指标(计算精度偏低)

data %>% group_by(label) %>% rstatix::get_summary_stats(value,show = c("mean", "sd","se")) #### 精度偏低
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
# A tibble: 5 × 6
  label variable     n   mean    sd    se
  <chr> <fct>    <dbl>  <dbl> <dbl> <dbl>
1 A     value       88  0.148 1.01  0.108
2 B     value      102  0.079 0.946 0.094
3 C     value      105 -0.089 1.02  0.1  
4 D     value       95 -0.102 0.958 0.098
5 E     value      110  0.089 0.823 0.078

3、plotrix 软件包计算标准误

data %>% group_by(label) %>% summarise(mean = mean(value),sd= sd(value),se = plotrix::std.error(value))
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
# A tibble: 5 × 4
  label    mean    sd     se
  <chr>   <dbl> <dbl>  <dbl>
1 A      0.148  1.01  0.108 
2 B      0.0789 0.946 0.0936
3 C     -0.0888 1.02  0.0998
4 D     -0.102  0.958 0.0983
5 E      0.0893 0.823 0.0785