CODEWITHSUNDEEP
ranalysispsychfactor-analysis
Tooba
Tooba

Reputation: 23

parallel analysis works with psych:fa.parallel but gives error message with paran package

I am running the parallel analysis with fa.parallel which works but the problem is that it provides or suggests a number of factors lower (3) than what I would expect (5):

fa.parallel(test3[, c(7:28)], fm="ml",sim=TRUE,n.iter = 100)

The answer I get in the R console:

Parallel analysis suggests that the number of factors = 3 and the number of components = 3

And a graph.

But how can I further see eigenvalues?

Secondly, as I tried another way to run the parallel analysis using paran package, it does not compute the parallel analysis instead gives me the error message:

Error in svd(X) : infinite or missing values in 'x'.

I have tried to search for this error message which I did not find in context of parallel analysis but in PCA and it has to do with missing values which I acknowledge I have in my dataset. What should I do? the code used for paran is:

paran(test2[, c(7:28)], iterations = 5000, centile = 0, quietly = FALSE, 
      status = TRUE, all = TRUE, cfa = TRUE, graph = TRUE, color = TRUE, 
      col = c("black", "red", "blue"), lty = c(1, 2, 3), lwd = 1, legend = TRUE, 
      file = "", width = 640, height = 640, grdevice = "png", seed = 0)

Upvotes: 1

Views: 485

Answers (1)

Shane
Shane

Reputation: 1

For whatever reason, you need to manually omit NAs. Try this:

paran(na.omit(test2[, c(7:28)]), iterations = 5000, centile = 0, quietly = FALSE, 
      status = TRUE, all = TRUE, cfa = TRUE, graph = TRUE, color = TRUE, 
      col = c("black", "red", "blue"), lty = c(1, 2, 3), lwd = 1, legend = TRUE, 
      file = "", width = 640, height = 640, grdevice = "png", seed = 0)

Upvotes: 0

Related Questions