R help - May 2003 - Returning the p-value of a factor analysis

Hi there, 

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.

Thanks in advance.

Wayne




Dr Wayne R. Jones
Statistician / Research Analyst
KSS Group plc
St James's Buildings
79 Oxford Street
Manchester M1 6SS
Tel: +44(0)161 609 4084
Mob: +44(0)7810 523 713



KSS Ltd
A division of Knowledge Support Systems Group plc
Seventh Floor  St James's Buildings  79 Oxford Street  Manchester  M1 6SS 
England
Company Registration Number 2800886 (Limited) 3449594 (plc)
Tel: +44 (0) 161 228 0040	Fax: +44 (0) 161 236 6305
mailto:kssg@kssg.com		http://www.kssg.com


The information in this Internet email is confidential and may b... {{dropped}}

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

Wayne Jones <JonesW at kssg.com> writes:
> Hi there, 
> 
> 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.
> 
> Thanks in advance.
It's calculated when printed. This is somewhat in bad style (according
to my tastes at least), but we have that happening in a few places. So
you need to redo the computations, which can be found inside
print.factanal. In 1.7.0, this function is hidden in the mva namespace
but you can view it with get("print.factanal",environment(factanal)).


-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907