隔离具有相同值的列

Question

I am trying to isolate those columns of a dataframe for which all observations have the same value (ignoring NAs). See below for a hypothetical example:

ForestName <- rep("Planige", 4)
TreeNumber <- c(1:4)
Height <- c(2.3, 2, 2.1, 2.9)
Type <- c("AA", "AA", NA, "AA")
df <- data.frame(ForestName, TreeNumber, Height, Type)
df

The new dataframe should contain ForestName and Type. The columns with unequal values (TreeNumber and Height) should be contained in another dataframe.

Answer 1

You can use unique and check if this reduces to a single element:

df[sapply(df,function(x) length(unique(x[!is.na(x)])))==1]
  ForestName Type
1    Planige   AA
2    Planige   AA
3    Planige <NA>
4    Planige   AA

Or test that all elements are equal to the first non- NA :

df[sapply(df, function(x) all(x==na.omit(x)[1],na.rm=T))]
  ForestName Type
1    Planige   AA
2    Planige   AA
3    Planige <NA>
4    Planige   AA

Answer 2

Among many other ways, I'm sure:

df[,sapply(df,function(x) {length(unique(x[!is.na(x)])) > 1})]

   TreeNumber Height
1          1    2.3
2          2    2.0
3          3    2.1
4          4    2.9

And you can negate the sapply expression to get the other columns.

Answer 3

使用相同的基本原理的更紧凑的方法

 Filter(function(x){length(unique(x[!is.na(x)])) <=1}, df)