Why no changes in for loop

Question

My data looks like below.

  ID       Group timing glucose_level
   <chr>    <dbl>  <int>         <dbl>
 1 black 7      0      0           136
 2 black 1      0      0           116
 3 blue 20      0      0           144
 4 green 18     0      0           114
 5 red 4        0      0           126
 6 red 5        0      0            80
 7 green 17     0      0           111
 8 green 3      0      0           109
 9 red 20       0      0            96
10 black 39     0      0           140

There are some missing values in glucose level. Below are part of glucose level data

[697] 128 157 132 142 141 128  97 120 123 131 132 126 140 103 147 181 217 257 218 234 240 281 273 224 210 227  NA  NA 245
[726] 230 252 270 238 134 173 193 151 128 180 218 218 190 225 214 186 140 237 239 279 246 244 146 196 157 178 140 127 187
[755] 206 177 220 179 167 127 219 223 241 162 235 140 187 154 172 116 139 194 173 150 187 131 176 114 154 180 223 150 219
[784] 130 169 104 136 132 121 175 169 128 110 101 100  92 122 196 203  96 143 129  NA  72 141 143 129 149 132 107  94  76
[813]  80  95  63 198 181  86 122

I wanna use a loop to replace the missing values. Here are my code:

    for(i in 1:length(data)){
    if(is.na(data[i,'glucose_level'])){
      if(data[i,'Group']==0){
        data[i,'glucose_level']=162.7059
      }else if(data[i,'Group']==1){
        data[i,'glucose_level']= 163.1415
      }else{
        data[i,'glucose_level']= 165.9106
      }
}
}

I print out data$glucose_level and find there are still missing values in it.why no changes in my data???

You are checking if the data is NA (missing) before you change any of the values. Since you don't have any NA values, that `if()` statement is always false, you probably meant to find non missing values. use `!is.na()` instead — MrFlick, Oct 06 '20 at 04:21
I don't see any in the data you posted. Make sure to provide a [reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) with sample input and desired output that we can copy/paste into R for testing. — MrFlick, Oct 06 '20 at 04:24
Also you are doing `1:length(data)` which almost certainly not right. You probably meant `1:nrow(data)` — MrFlick, Oct 06 '20 at 04:25

score 1 · Answer 1 · answered Oct 06 '20 at 04:35

You can use nested ifelse or case_when and check for conditions and assign values accordingly.

library(dplyr)

data <- data %>%
          mutate(glucose_level = case_when(!is.na(glucose_level) ~ glucose_level,
                                             Group == 0  ~ 162.7059, 
                                             Group == 1 ~ 163.1415, 
                                             TRUE ~ 165.9106))

score 0 · Answer 2 · answered Oct 07 '20 at 00:03

We can use fcase from data.table

library(data.table)
setDT(data)[, glucose_level := fcase(!is.na(glucose_level), glucose_level,
                                         Group == 0, 162.7059, 
                                         Group == 1,163.1415, 
                                         165.9106)]

Why no changes in for loop

2 Answers2