在R中合并含NA值的特定行：移除重复时间的NA行

嘿，我来帮你搞定这个数据集的问题！从你提供的数据和截图来看，整点时刻会出现两行重复的Time记录，其中一行有有效的elevation值，另一行是NA。下面给你两个实用的R语言解决方案，按需选择就行：

方法1：直接移除elevation为NA的重复行 #

这个方法简单直接，过滤掉那些elevation为NA且属于重复Time的行，保留有有效数值的那一行。如果某个Time对应的两行elevation都是NA，那就保留其中一行（完全符合你说的“其他列保留NA也无妨”的需求）。

用dplyr包实现的代码：

library(dplyr)

cleaned_data <- stackoverflow %>%
  # 保留两种行：elevation非NA的，或者是该Time的唯一行（避免过滤掉所有NA的情况）
  filter(!is.na(elevation) | !duplicated(Time, fromLast = TRUE))

如果你习惯用base R，也可以这样写：

# 生成保留行的索引
keep_rows <- !is.na(stackoverflow$elevation) | !duplicated(stackoverflow$Time, fromLast = TRUE)
cleaned_data <- stackoverflow[keep_rows, ]

方法2：按时间分组合并（保留所有列的非NA值） #

如果你的其他列也可能存在不同的非NA值，想要把同一Time的行合并，取每列的第一个非NA值，这个方法更通用：

library(dplyr)

cleaned_data <- stackoverflow %>%
  group_by(Time) %>%
  # 对每一列，合并取第一个非NA值
  summarise(across(everything(), ~ coalesce(!!!.)),
            .groups = "drop")

这个方法会自动把同一Time的所有行合并成一行，elevation列会保留有效的数值，其他列也会保留各自的非NA值，完美解决你的需求。

验证结果 #

处理完之后，你可以用head(cleaned_data)或者查看整点时刻的记录，比如filter(cleaned_data, Time == as.POSIXct("2015-05-24 14:00:00 UTC"))，就能看到只保留了elevation为282.27的那一行啦。

你的数据集示例（前100行）：

structure(list(Time = structure(c(1432425600, 1432426500, 1432427400, 
1432428300, 1432429200, 1432430100, 1432431000, 1432431900, 1432432800, 
1432433700, 1432434600, 1432435500, 1432436400, 1432437300, 1432438200, 
1432439100, 1432440000, 1432440900, 1432441800, 1432442700, 1432443600, 
1432444500, 1432445400, 1432446300, 1432447200, 1432448100, 1432449000, 
1432449900, 1432450800, 1432450800, 1432451700, 1432452600, 1432453500, 
1432454400, 1432454400, 1432455300, 1432456200, 1432457100, 1432458000, 
1432458000, 1432458900, 1432459800, 1432460700, 1432461600, 1432461600, 
1432462500, 1432463400, 1432464300, 1432465200, 1432465200, 1432466100, 
1432467000, 1432467900, 1432468800, 1432468800, 1432469700, 1432470600, 
1432471500, 1432472400, 1432472400, 1432473300, 1432474200, 1432475100, 
1432476000, 1432476000, 1432476900, 1432477800, 1432478700, 1432479600, 
1432479600, 1432480500, 1432481400, 1432482300, 1432483200, 1432483200, 
1432484100, 1432485000, 1432485900, 1432486800, 1432486800, 1432487700, 
1432488600, 1432489500, 1432490400, 1432490400, 1432491300, 1432492200, 
1432493100, 1432494000, 1432494000, 1432494900, 1432495800, 1432496700, 
1432497600, 1432498500, 1432499400, 1432500300, 1432501200, 1432502100, 
1432503000), tzone = "UTC", class = c("POSIXct", "POSIXt")), 
    Turtle = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), levels = c("R3L1", "R3L11", "R3L12", "R3L2", "R3L4", 
    "R3L8", "R3L9", "R4L8", "R8L1", "R8L4", "R8NAT123"), class = "factor"), 
    elevation = c(NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, 282.27, NA, NA, NA, NA, 283.21, NA, NA, NA, NA, 282.14, 
    NA, NA, NA, NA, 281.63, NA, NA, NA, NA, 281.63, NA, NA, NA, 
    NA, 281.63, NA, NA, NA, NA, 282.63, NA, NA, NA, NA, 281.63, 
    NA, NA, NA, NA, 282.14, NA, NA, NA, NA, 281.63, NA, NA, NA, 
    NA, 282.14, NA, NA, NA, NA, 281.36, NA, NA, NA, NA, 282.14, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA)), row.names = c(NA, 
-100L), class = c("tbl_df", "tbl", "data.frame"))

内容的提问来源于stack exchange，提问作者Cassidy