不明白我在反向编码方面做错了什么？

Question

I'm trying to reverse code some of my data, as 24 of the columns are negatively keyed and needs to be reversed.

I use this dataset: https://www.kaggle.com/tunguz/big-five-personality-test This is the background documentation for the data, and explains how the different questions are keyed: https://ipip.ori.org/New_IPIP-50-item-scale.htm When I try to use the reverse code from psych I just get this error:

Error in items %*% keys.d : 
  requires numeric/complex matrix/vector arguments

I also get an error telling me that I can't do it with lists.... for some reason, I can't seem to produce that error now...

library("pacman", "psych")
dataIPIP <- data.final[, c(1:50)]

col_names2 <- c(paste("E", 1:10, sep = ""), paste("N", 1:10, sep = ""), paste("A", 1:10, sep = ""), paste("C", 1:10, sep = ""), paste("O", 1:10, sep = ""))
names(dataIPIP) <- col_names2

keys <- c(1,-1,1,-1,1,-1,1,-1,1,-1,
         -1,1,-1,1,-1,-1,-1,-1,-1,-1,
         -1,1,-1,1,-1,1,-1,1,1,1,
          1,-1,1,-1,1,-1,1,-1,1,1,
          1,-1,1,-1,1,-1,1,1,1,1)
spm_rev <- reverse.code(keys, items = dataIPIP[,-1], mini = rep(1,50), maxi = rep(5,50))

The structure for the dataIPIP dataframe is:

'data.frame':   1015341 obs. of  50 variables:
 $ E1 : chr  "4" "3" "2" "2" ...
 $ E2 : chr  "1" "5" "3" "2" ...
 $ E3 : chr  "5" "3" "4" "2" ...
 $ E4 : chr  "2" "4" "4" "3" ...
 and it goes like this through E5-10, N1-10, A1-10, C1-10 and finally O1-10.

Answer 1

It looks like your data has been read in as character variables (the "chr" in the structure overview of your data). R doesn't know how to "reverse code a letter", so you need to change the variables to actual numerical values.

The way the dataframe looks on kaggle it should be straight forward to just convert each column in the data frame.

One version to do this in base R would be

dataIPIP <- as.data.frame(sapply(dataIPIP, as.numeric))

Hope that helps. Cheers