[R] Returning the p-value of a factor analysis

[Prof Brian Ripley]

On Thu, 8 May 2003, Wayne Jones wrote:

> Does anyone know how to explicitly refer to the p-value of thet test that
> the chosen number of factors is significant in a factor analysis.
> It's not in the list of values for the factanal command output yet it is
> printed out with the results.

Right, so look at the code to compute it in print.factanal and write a 
small function to extract the value you want from the object.
Something like

Pval.factanal <- function(x)
{
    p <- nrow(x$loadings); factors <- x$factors
    if(!is.na(x$n.obs) && x$dof > 0) {
        dof <- x$dof
        stat <- (x$n.obs - 1 - (2 * p + 5)/6 -
                 (2 * factors)/3) * x$criteria["objective"]
        pchisq(stat, dof, lower.tail = FALSE)
    } else NA
}

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595