CODEWITHSUNDEEP
r
rnso
rnso

Reputation: 24623

Mysterious source of output in R?

I am using following code using mtcars data and factanal function for factor analysis. The print of fit$loadings give the proportional variance but it does not seem to be there in str(fit$loadings) : 

> fit <- factanal(mtcars, 3, rotation="varimax")
> fit$loadings

Loadings:
     Factor1 Factor2 Factor3
mpg   0.643  -0.478  -0.473 
cyl  -0.618   0.703   0.261 
disp -0.719   0.537   0.323 
hp   -0.291   0.725   0.513 
drat  0.804  -0.241         
wt   -0.778   0.248   0.524 
qsec -0.177  -0.946  -0.151 
vs    0.295  -0.805  -0.204 
am    0.880                 
gear  0.908           0.224 
carb  0.114   0.559   0.719 

               Factor1 Factor2 Factor3
SS loadings      4.380   3.520   1.578
Proportion Var   0.398   0.320   0.143  <<<<<<<<<<<<< I NEED THESE NUMBERS AS A VECTOR
Cumulative Var   0.398   0.718   0.862
> 
> str(fit$loadings)
 loadings [1:11, 1:3] 0.643 -0.618 -0.719 -0.291 0.804 ...
 - attr(*, "dimnames")=List of 2
  ..$ : chr [1:11] "mpg" "cyl" "disp" "hp" ...
  ..$ : chr [1:3] "Factor1" "Factor2" "Factor3"

How can I get Proportional variance vector from fit$loadings? Thanks for your help.

Upvotes: 2

Views: 58

Answers (1)

Julius Vainora
Julius Vainora

Reputation: 48251

Let obj <- fit$loadings. Here is a complete path how to obtain the result.

By writing fit$loadings (or obj) we actually call print(obj). So, after looking at str, you might want to check what does the specific print method do with obj. To know what method we should look for, we check class(obj) and get "loadings".

Then, writing print.loadings does not give anything because the function is hidden. Therefore, since function factanal is in the package stats, we call stats:::print.loadings and get a complete source code of the function. By inspecting it, we see that we can get the desired result as follows.

colSums(obj^2) / nrow(obj)
#   Factor1   Factor2   Factor3 
# 0.3982190 0.3199652 0.1434125

Upvotes: 2

Related Questions