使用dplyr为指定列计算符合条件的行均值

原始数据 #

df <- data.frame(name = c("Bill","Sally","John","Lucy","Jake","Rob","Sarah"),
                 score1 = c(1,2,1,4,5,3,4),
                 score2 = c(4,2,3,2,1,3,NA),
                 score3 = c(3,4,1,4,NA,NA,NA),
                 score4 = c(4,3,4,NA,NA,NA,NA),
                 score5 = c(4,4,NA,NA,NA,NA,NA),
                 score6 = c(1,NA,NA,NA,NA,NA,NA))

需求 #

仅针对score开头的列中非NA值数量≥4的行，计算分数均值并生成avg_score列；其余行的avg_score设为NA。

解决方案 #

方法1：行分组处理（逻辑直观） #

通过行分组实现逐行判断与计算，适合理解行级操作逻辑：

library(dplyr)

df <- df %>%
  rowwise() %>%
  mutate(
    # 统计每行score列的非NA值数量
    non_na_count = sum(!is.na(c_across(starts_with("score")))),
    # 满足条件则计算均值（忽略NA），否则设为NA
    avg_score = ifelse(non_na_count >= 4, mean(c_across(starts_with("score")), na.rm = TRUE), NA)
  ) %>%
  select(-non_na_count) %>%  # 移除中间计算用的计数列
  ungroup() %>%
  mutate(avg_score = round(avg_score, 3))  # 保留三位小数匹配示例输出

方法2：向量化处理（高效版） #

无需行分组，直接用向量级函数实现，处理大数据集时效率更高：

library(dplyr)

df <- df %>%
  mutate(
    # 统计score列的非NA值数量
    non_na_count = rowSums(!is.na(select(., starts_with("score")))),
    # 条件判断并计算均值
    avg_score = ifelse(non_na_count >= 4, rowMeans(select(., starts_with("score")), na.rm = TRUE), NA)
  ) %>%
  select(-non_na_count) %>%
  mutate(avg_score = round(avg_score, 3))

最终输出 #

df
#    name score1 score2 score3 score4 score5 score6 avg_score
# 1  Bill      1      4      3      4      4      1     2.833
# 2 Sally      2      2      4      3      4     NA     3.000
# 3  John      1      3      1      4     NA     NA     2.250
# 4  Lucy      4      2      4     NA     NA     NA        NA
# 5  Jake      5      1     NA     NA     NA     NA        NA
# 6   Rob      3      3     NA     NA     NA     NA        NA
# 7 Sarah      4     NA     NA     NA     NA     NA        NA

内容的提问来源于stack exchange，提问作者Bradley Allf