Handle missing values for likert scale in R

Question

I have data with 5 columns in a variable called studentData. Each column has 326 rows, except one which has 3 rows missing. Each column is a 5 point likert value, from the set mylevels <- c('Strongly disagree', 'Disagree', 'Neither agree nor disagree', 'Agree', 'Strongly agree')

When I print the number of levels in each column it gives me value as 6 for the second column (studentData$Increased.confidence), because it has 3 missing values which R interprets as another factor for this column.

> sapply(studentData, function(x) { length(levels(x)) } ) # The number of levels in each factor
              ï..Increased.engagement                  Increased.confidence               Improved.writing.skills 
                                    5                                     6                                     5 
   Made.useful.contribution.to.course Should.keep.games.for.future.students 
                                    5                                     5

Because of this I get the error stating that the number of levels should be same for likert function to work. How should I handle those 3 missing values?

> studentLikert <- likert(studentData)
Error in likert(studentData) : 
  All items (columns) must have the same number of levels

There is no NA, just three missing values because the survey respondent choose not to answer it — Vipin Verma, Apr 27 '17 at 09:25
Ok then, consider making your question reproducible, it may clear up things. — Roman Luštrik, Apr 27 '17 at 09:31
How can I upload my xls or csv file for someone to reproduce it? Can you suggest some way? — Vipin Verma, Apr 27 '17 at 09:44
Don't. Just make a mock data.frame which demonstrates the problem. See [here](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example). — Roman Luštrik, Apr 27 '17 at 09:49

Bea · Accepted Answer · 2017-04-27T09:55:10.010

1

Try this: define your columns as factor making sure that you exclude missing values from factor level definition by using exclude=' '

a <- c('A','B','C','','A')
b <- c('A','B','A','C','B')
df <- data.frame(a,b)

mylevels <- c('A', 'B', 'C')
df <- as.data.frame(lapply(df,function(x) {factor(x,levels=mylevels, exclude="")}))

edited Apr 27 '17 at 09:55

answered Apr 27 '17 at 09:45

Bea

1,110
12
20

Handle missing values for likert scale in R

1 Answers1