Skip to content

Running Omega in R

4 messages · John Smith, Rui Barradas, Peter Dalgaard +1 more

Original
John Smith
# 
Hi,

I opened a question on stack overflow I?m hoping this mailing list can help with.
I have a dataset below (this is made up but produces the same error I am getting)

structure(list(Q1 = c(4, 5, 3, 5, 4, 5, 3, 5, 5, 5, 6,
3, 5, 4, 6, 5, 5, 6, 7, 4, 5, 5, 3, 4, 4, 5, 4, 3, 5, 4, 5, 5,
6, 6, 3, 6, 3, 4, 4, 4, 6, 5, 3, 2, 6, 6, 4, 5, 4, 3, 6, 4, 4,
5, 6, 2, 4, 3, 4, 6, 4, 6, 4, 5, 5, 6, 4, 6, 5, 5, 4, 5, 6, 6,
2, 5, 4, 3, 4, 4, 4, 6, 3, 3, 5, 4, 4, 4, 5, 5, 5, 3, 6, 6, 6,
6, 5, 4, 3, 5), Q2 = c(7, 4, 4, 4, 4, 6, 6, 6, 7, 6, 5,
6, 5, 4, 5, 6, 6, 6, 7, 5, 4, 4, 6, 6, 4, 4, 6, 2, 6, 5, 4, 6,
4, 6, 6, 6, 5, 4, 4, 4, 4, 3, 3, 4, 4, 4, 4, 6, 2, 6, 6, 5, 4,
6, 6, 4, 4, 7, 6, 5, 5, 5, 5, 6, 5, 5, 4, 5, 5, 5, 4, 6, 7, 5,
5, 5, 6, 5, 6, 5, 6, 7, 2, 6, 5, 7, 3, 5, 5, 3, 3, 3, 7, 4, 5,
6, 6, 6, 5, 7), Q3 = c(5, 4, 5, 6, 4, 4, 5, 4, 2, 6, 5,
5, 5, 5, 7, 5, 5, 6, 7, 6, 3, 6, 6, 6, 5, 6, 6, 5, 5, 4, 5, 5,
6, 6, 5, 6, 5, 5, 4, 4, 6, 4, 4, 4, 4, 4, 4, 5, 5, 4, 5, 5, 4,
3, 5, 4, 5, 6, 6, 6, 4, 5, 5, 5, 6, 4, 5, 5, 7, 4, 5, 6, 6, 5,
5, 3, 3, 5, 4, 6, 5, 5, 1, 3, 5, 3, 2, 5, 4, 6, 6, 6, 6, 4, 6,
3, 6, 6, 6, 5), Q4 = c(6, 6, 4, 7, 4, 6, 7, 6, 7, 6, 6,
6, 5, 7, 7, 6, 6, 5, 7, 7, 6, 6, 7, 7, 6, 6, 6, 5, 6, 7, 5, 6,
7, 5, 4, 6, 4, 3, 6, 4, 6, 6, 6, 3, 5, 7, 5, 6, 4, 6, 7, 6, 7,
4, 6, 3, 5, 7, 5, 4, 6, 6, 4, 6, 5, 5, 5, 5, 7, 7, 7, 6, 6, 6,
5, 6, 6, 4, 5, 7, 6, 7, 3, 5, 6, 5, 6, 5, 5, 7, 7, 6, 6, 2, 7,
6, 6, 7, 7, 5)), .Names = c("Q1", "Q2", "Q3",
"Q4"), row.names = c(NA, 100L), class = "data.frame")


When i run Cronbach Alpha with R i get a result

psych::alpha(construct,
         na.rm = TRUE,
         title = 'myscale',
         n.iter = 1000)


When i run Omega using R i get the following error message

"Error in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : I am sorry: missing values (NAs) in the correlation matrix do not allow me to continue.
Please drop those variables and try again. In addition: There were 50 or more warnings (use warnings() to see the first 50)"

psych::omega(m = construct,
      nfactors = 1, fm = "pa", n.iter = 1000, p = 0.05,
      title = "Omega", plot = FALSE, n.obs = 100)


Stackoverflow Link: http://stackoverflow.com/questions/41533231/running-omega-with-psych-library-in-r?noredirect=1#comment70278453_41533231

If I have the wrong mailing list, could you direct me to the appropriate one (I?m aware it might not be an R issue but more of a stats question)



Thank you for your time
Rui Barradas
# 
Hello,

I no nothing about package psych so if you ask whether this is the wrong 
list you can always try

maintainer("psych")
[1] "William Revelle <revelle at northwestern.edu>"

Hope this helps,

Rui Barradas

Em 15-01-2017 12:32, John Smith escreveu:
Peter Dalgaard
# 
...although publishing a package is not the same as issuing a blank cheque for free support, so you may want to dig a little deeper first. 

I'd check the documentation for whether the data are in the right format. In particular, can "m" be a data frame? If it wants a matrix, you probably need to give it one (m=as.matrix(construct) should do it).

-pd

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      David L Carlson
    

    

      
      #
    

  


  

      
I get an additional error message when I run your example so perhaps your version is not the current (1.6.12) one? 

Loading required namespace: GPArotation
Failed with error:  ?there is no package called ?GPArotation??
Error in omegah(m = m, nfactors = nfactors, fm = fm, key = key, flip = flip,  : 
  I am sorry, you need to have the  GPArotation package installed

Then I get a great deal of error output followed by the one you list. The first error is 

"Omega_h for 1 factor is not meaningful, just omega_t" (repeated 174 times). I'm copying the package maintainer as this seems like overkill.

Your problem is addressed in the first line of the manual page for omega:

"McDonald has proposed coefficient omega as an estimate of the general factor saturation of a test. One way to find omega is to do a factor analysis of the original data set, rotate the factors obliquely, do a Schmid Leiman transformation, and then find omega. "

You cannot rotate a single factor so you cannot compute omega.

-------------------------------------
David L Carlson
Department of Anthropology
Texas A&M University
College Station, TX 77840-4352

-----Original Message-----
From: R-help [mailto:r-help-bounces at r-project.org] On Behalf Of peter dalgaard
Sent: Sunday, January 15, 2017 2:27 PM
To: Rui Barradas <ruipbarradas at sapo.pt>
Cc: John Smith <jSmith.Coursera at outlook.com>; r-help at r-project.org
Subject: Re: [R] Running Omega in R

            

      
      
      
    

        
...although publishing a package is not the same as issuing a blank cheque for free support, so you may want to dig a little deeper first. 

I'd check the documentation for whether the data are in the right format. In particular, can "m" be a data frame? If it wants a matrix, you probably need to give it one (m=as.matrix(construct) should do it).

-pd