用 R 中另一列的值替换一列中的空单元格

Question

I am trying to pull the cell values from the StudyID column to the empty cells SigmaID column, but I am running into an odd issue with the output.

This is how my data looks before running commands.

StudyID    Gender  Region  SigmaID
LM24008      1       20    LM24008  
LM82993      1       16    LM28888  
ST04283      0       44      
ST04238      0       50      
LM04829      1       24    LM23921  
ST91124      0       89
ST29001      0       55

I tried accomplishing this by writing the syntax in three ways, because I wasn't sure if there is a problem with the way the logic was set up. All three produce the same output.

df$SigmaID <- ifelse(test = df$SigmaID != "", yes = df$SigmaID, no = df$StudyID)

df$SigmaID <- ifelse(df$SigmaID == "", df$StudyID, df3$SigmaID)

df %>% mutate(SigmaID = ifelse(Gender == 0, df$StudyID, df$SigmaID)

Output: instead of pulling the values from from the StudyID column, it is populating one to four digit numbers.

StudyID    Gender  Region  SigmaID
LM24008      1       20    LM24008  
LM82993      1       16    LM28888  
ST04283      0       44    5  
ST04238      0       50    4908  
LM04829      1       24    LM23921
ST91124      0       89    209
ST29001      0       55    4092

I have tried recoding the empty spaces to NA and then calling on NA in the logic, but this produced the same output as seen above. I'm wondering if it could have anything to do with variable type or variable attributes and something's off about how it's reading the characters in StudyID. Would appreciate feedback on this issue!

Answer 1

Here is how to do it:

df$SigmaID[df$SigmaID == ""] = df$StudyID[df$SigmaID == ""]

df[df$SigmaID == ""] selects only the rows where SigmaID==""

I also recommend using data.table instead of data.frame . It is faster and has some useful syntax features:

library(data.table)
setDT(df) # setDT converts a data.frame to a data.table
df[SigmaID=="",SigmaId:=StudyID]

Answer 2

Following up on this, As it turns out. default R converts string types into factors. There are a few ways of addressing the issue above.

i <- sapply[df, is.factor]
df[i] <- lapply(df[i], as.character)

Another method:

df <- read.csv("/insert file pathway here", stringAsFactors = FALSE)

This is what I found to be helpful. I'm sure there are additional methods of troubleshooting this as well.