CODEWITHSUNDEEP
rpcasurveymissing-data
user2747473
user2747473

Reputation: 21

R Missing Values

I am new to R and trying to run a factor analysis (method = pca. rotation = varimax and number of factors is 6). I've a survey data that I loaded to R, and then selected a subset of cases, and variables. I did the following:

arab<-read.table("C:\\Users\\admin\\Desktop\\spsapaper\\arabb.csv",header=T,sep=",")

abjor<-subset(arab,COUNTRY=="1")

items<- c("Q2011", "Q2012", "Q4012", "Q4013", "Q5022", "Q5051", "Q5052", "Q5056", "Q5055", "Q6131", "Q6133", "Q6132", "Q2464", "Q2462", "Q2013", "Q2473", "Q2476", "Q2475", "Q2014", "Q5022", "Q2323")


install.packages("psych")

library(psych)

So far everything ran smoothly, and I used the fix() function to check the data in the editor and it worked. Then I ran:

fit <- factanal(abjor, 6, rotation="varimax")

and it gave me:

 Error in cov.wt(z) : 'x' must contain finite values only

I tried to fix the problem by deleting missing values via:

abjorr<- na.omit(abjor)

and it gave me

Error in factanal(abjorr, 6, rotation = "varimax") : 
  factor analysis applies only to numerical variables

The fix() produced an empty data editor with no values.

Then, I tried, nabjor<- v[ !is.na( abjor ) ]

and ran fix() and it produced a data editor with all NA values.

I'm trying to handle missing values to run a factor analysis, but it seems that I'm missing something?

Upvotes: 2

Views: 3025

Answers (1)

John Paul
John Paul

Reputation: 12654

How about:

abjor2<-complete.cases(abjor)

The complete.cases function will give you only those rows that do not have any NA values.

Upvotes: 1

Related Questions