Obtain overall variance explained by factor solution using factormat

[Alison El Ayadi]

Hi Statalisters,

I am looking to indicate the % of explained common variance of a measure using the factormat command and I am not having luck figuring out how to specify this request in my code.
Can anybody help?

Thanks,
Alison

[Mike Lacy]

My guess is that you want "proportion of common variance in a *set* of variables that is attributable to the first factor."

If that is the case, the listing at the intersection of "Proportion" and "Factor1" in the standard output of -factormat- should work, presuming you choose some kind of common factor model ("principal factor"), which is the default in Stata.
So, what about this as an example?

sysuse auto
correlate price-gear_ratio
mat R = r(C)
factormat R, n(69)

[Alison El Ayadi]

Thank you so much, Mike. I should have been more explicit in my post. I am actually looking to understand the total proportion variance by all of my identified factors combined. I have used a promax rotation to allow for correlation of factors, thus do not receive a cumulative proportion of variance in my output. Here is the section of code that I would expect it to be in:

Factor analysis/correlation Number of obs = 58
Method: iterated principal factors Retained factors = 4
Rotation: oblique promax (Kaiser on) Number of params = 82

--------------------------------------------------------------------------
Factor | Variance Proportion Rotated factors are correlated
-------------+------------------------------------------------------------
Factor1 | 4.81856 0.4276
Factor2 | 3.89734 0.3459
Factor3 | 3.34365 0.2967
Factor4 | 2.65407 0.2355
--------------------------------------------------------------------------
LR test: independent vs. saturated: chi2(231) = 781.09 Prob>chi2 = 0.0000

Thanks so much,
Alison