0

I'm trying to run a factor analysis on a set of 80 dichotomous variables (1440 cases) using the hector function from the polycor package and the instructions I found here: http://researchsupport.unt.edu/class/Jon/Benchmarks/BinaryFA_L_JDS_Sep2014.pdf

Sadly, after I select just the variables interest from the rest of my dataset and run the factor analysis on them, I seem to consistently get the following error and warnings

Error in optim(0, f, control = control, hessian = TRUE, method = "BFGS") : 
  non-finite finite-difference value [1]
In addition: Warning messages:
1: In log(P) : NaNs produced
2: In log(P) : NaNs produced

This is with the command/when I hit the step described in the above PDF: 

testMat <- hetcor(data)$cor

No idea what this means or how to proceed... Your thoughts are appreciated. Thank you!

D. K.
  • 73
  • 7
  • You could try another `method`. This usually happens because the data and model don't work well together and the optimization fails. – Roman Luštrik Jul 21 '16 at 06:46
  • I did have a feeling that may be the case. Any ideas on what other methods might be worth trying? – D. K. Jul 21 '16 at 06:50
  • First try to identify variables which contribute very little to your model but take up degrees of freedom. – Roman Luštrik Jul 21 '16 at 07:02
  • If it matters, the 80 binary variables at hand are dummy variables to indicate whether a word is used by the participant (row) in a separate free-response column. The 80 is already reduced on various criteria from an initial 1800, so I'm not sure what else I can do :\ – D. K. Jul 21 '16 at 12:51

0 Answers0