基于3水平condition列重排44行df：交替排序需求及报错解决

问题描述 #

有一个44行的数据框df，condition列包含3个水平：

• neutral：11行
• fem：11行
• filler：22行

需要完成两步操作：

1. 随机打乱每个水平内的行
2. 按filler→neutral→filler→fem的交替顺序重排整个数据框

数据框结构 #

Rows: 44
Columns: 11
$ display              <chr> "experimental_eye_tracking", "experimental_eye_tracking", "experimental_eye_tracking"…
$ ANSWER               <lgl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
$ TopLeft              <chr> "1_row-1-column-1.png", "2_row-1-column-1.png", "3_row-1-column-1.png", "4_row-1-colu…
$ BottomLeft           <chr> "1_row-1-column-2.png", "2_row-1-column-2.png", "3_row-1-column-2.png", "4_row-1-colu…
$ TopRight             <chr> "1_row-2-column-1.png", "2_row-2-column-1.png", "3_row-2-column-1.png", "4_row-2-colu…
$ BottomRight          <chr> "1_row-2-column-2.png", "2_row-2-column-2.png", "3_row-2-column-2.png", "4_row-2-colu…
$ audio                <chr> "بعد الصيف تم انتخاب الرئيسة.mp3", "بعد الصباح تم إخراج الأم.mp3", "في المساء تم تنوي…
$ item_type            <chr> "experimental", "experimental", "experimental", "experimental", "experimental", "expe…
$ condition            <chr> "neutral", "neutral", "neutral", "neutral", "neutral", "neutral", "neutral", "neutral…
$ comprehension_Q      <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…

condition列分布 #

[1] "neutral" "neutral" "neutral" "neutral" "neutral" "neutral" "neutral" "neutral" "neutral" "neutral" "neutral"
[12] "fem"     "fem"     "fem"     "fem"     "fem"     "fem"     "fem"     "fem"     "fem"     "fem"     "fem"    
[23] "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler" 
[34] "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler"  "filler" 

condition列dput数据 #

c("neutral", "neutral", "neutral", "neutral", "neutral", "neutral", 
"neutral", "neutral", "neutral", "neutral", "neutral", "fem", 
"fem", "fem", "fem", "fem", "fem", "fem", "fem", "fem", "fem", 
"fem", "filler", "filler", "filler", "filler", "filler", "filler", 
"filler", "filler", "filler", "filler", "filler", "filler", "filler", 
"filler", "filler", "filler", "filler", "filler", "filler", "filler", 
"filler", "filler")

已尝试操作 #

使用相关方案时出现错误：Arguments have differing numbers of rows

期望结果 #

最终数据框按以下顺序排列（示例）：

Rows: 10
Columns: 2
$ condition <chr> "filler", "neutral", "filler", "fem", "filler", "neutral", "filler", "fem", "filler", "neutral"
$ trial     <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10

可行解决方案 #

通过dplyr和tibble包实现，步骤如下：

1. 加载所需包：

library(dplyr)
library(tibble)

2. 拆分数据框并打乱各水平内的行：

# 拆分并随机打乱每个组
neutral_shuffled <- df %>% filter(condition == "neutral") %>% sample_frac(1)
fem_shuffled <- df %>% filter(condition == "fem") %>% sample_frac(1)
filler_shuffled <- df %>% filter(condition == "filler") %>% sample_frac(1)

# 将filler拆分为两个等长组（各11行）
filler1 <- filler_shuffled[1:11, ]
filler2 <- filler_shuffled[12:22, ]

3. 按指定顺序拼接数据框：

# 创建交替顺序的索引映射
order_df <- tibble(
  group = rep(c("filler1", "neutral", "filler2", "fem"), each = 11),
  idx = rep(1:11, 4)
)

# 按索引拼接并整理最终数据框
final_df <- order_df %>%
  group_by(group) %>%
  mutate(row = list(
    if (group == "filler1") filler1[idx, ]
    else if (group == "neutral") neutral_shuffled[idx, ]
    else if (group == "filler2") filler2[idx, ]
    else fem_shuffled[idx, ]
  )) %>%
  unnest(row) %>%
  ungroup() %>%
  select(-group, -idx)

4. 验证结果：

# 查看前10行的condition列顺序
head(final_df$condition, 10)

内容的提问来源于stack exchange，提问作者Alaa