psych (version 1.0-25)

factor.pa: Principal Axis Factor Analysis

Description

Among the many ways to do factor analysis, one of the most conventional is principal axes. An eigen value decomposition of a correlation matrix is done and then the communalities for each variable are estimated by the first n factors. These communalities are entered onto the diagonal and the procedure is repeated until the sum(diag(r)) does not vary. For well behaved matrices, maximum likelihood factor analysis (factanal) is probably preferred.

Usage

factor.pa(r, nfactors=1, residuals = FALSE, rotate = "varimax", min.err = 0.001, digits = 2, max.iter = 50)

Arguments

r

A correlation matrix or a raw data matrix. If raw data, the correlation matrix will be found using pairwise deletion.

nfactors

Number of factors to extract, default is 1

residuals

Should the residual matrix be shown

rotate

"none", "varimax", "promax"

min.err

Iterate until the change in communalities is less than min.err

digits

How many digits of output should be returned

max.iter

Maximum number of iterations for convergence

Value

• If it is a LIST, use
• values
  Eigen values of the final solution
• loadings
  An item by factor loading matrix. Suitable for use in other programs (e.g., GPA rotation or factor2cluster.
• fit
  How well does the factor model reproduce the correlation matrix. (See VSS, ICLUST, and principal for this fit statistic.
• communality
  The history of the communality estimates. Probably only useful for teaching what happens in the process of iterative fitting.

Details

Factor analysis is an attempt to approximate a correlation or covariance matrix with one of lesser rank. The basic model is that $nRn \approx nFk kFn'$ where k is much less than n. There are many ways to do factor analysis, and maximum likelihood procedures are probably the most preferred (see factanal).

Principal axes factor analysis has a long history in exploratory analysis and is a straightforward procedure. Successive eigen value decompositions are done on a correlation matrix with the diagonal replaced with diag (FF') until sum(diag(FF')) does not change (very much). The current limit of max.iter =50 seems to work for most problems, but the Holzinger-Harmon 24 variable problem needs about 203 iterations to converge for a 5 factor solution.

Principal axes may be used in cases when maximum likelihood solutions fail to converge.

References

Gorsuch, Richard, (1983) Factor Analysis. Lawrence Erlebaum Associates.

See Also

principal, VSS, ICLUST

Examples

#using the Harmon 24 mental tests, compare a principal factor with a principal components solution
pc <- principal(Harman74.cor$cov,4,rotate=TRUE)
pa <- factor.pa(Harman74.cor$cov,4,rotate="varimax")
round(factor.congruence(pc,pa),2)

#then compare with a maximum likelihood solution using factanal
mle <- factanal(x,4,covmat=Harman74.cor$cov)
round(factor.congruence(mle,pa),2)
#note that the order of factors and the sign of some of factors differ