order()函数失效求助：按数据框首行升序排序失败

Hey there! Let's break down why your order() function is giving unexpected results, and fix it step by step.

The Root Cause #

Your issue comes down to data type mismatch in your data frame. When you use rbind(values, labels):

• values is a numeric vector, but labels is a character vector
• rbind() forces all elements to the same type (character, since it's more flexible), so your first row (df[1,]) ends up being a character vector instead of numeric.

When you sort character strings like "11", "10", "2", R compares them lexicographically (like dictionary order): "1" comes before "2", so "10" and "11" end up before "2"—that's exactly the weird order you're seeing!

The Fix #

We need two key adjustments:

1. Create the data frame in a way that preserves the numeric type of the first row
2. Use order() correctly on the numeric values

Step 1: Correctly Build the Data Frame

Instead of rbind(), use t() (transpose) on a column-bound dataset to keep types intact:

values <- c(11,10,9,8,7,6,5,4,3,2,1)
labels <- c("A","B","C","D","E","F","G","H","I","J","K")

# Bind as columns first, then transpose to get rows
df <- as.data.frame(t(cbind(values, labels)), stringsAsFactors = FALSE)
# Set meaningful row names
row.names(df) <- c("values", "labels")

Now df[1,] is a numeric vector, which is what we need for proper sorting.

Step 2: Sort with order()

Now you can use either of your original order() calls (they'll work correctly now):

# Option 1: Direct order call
newdf <- df[, order(df[1,])]

# Option 2: Using with() (though not necessary here, but still valid)
newdf <- df[, with(df, order(df[1,]))]

# Print the result
print(newdf)

Expected Output #

You'll now get the correct ascending sort based on the first row's numeric values:

X11 X10 X9 X8 X7 X6 X5 X4 X3 X2 X1
values   1   2  3  4  5  6  7  8  9 10 11
labels   K   J  I  H  G  F  E  D  C  B  A

Quick Check (If You Want to Verify Types) #

To confirm your first row is numeric, run:

class(df[1,])
# Should return "numeric" instead of "character"

内容的提问来源于stack exchange，提问作者Rodney Beard