Skip to content

How to extract numbers from ANOVA tables?

3 messages · Brinkmann, Lars, Brian Ripley, John Fox

#
On Sat, 8 Dec 2007, Brinkmann, Lars wrote:

            
The Value section of the relevant help page tells you how the values are 
stored.  If you need more details, reading the appropriate print() method 
will show you where the printed numbers come from (and occasionally they 
are computed in the print method).

For e.g. lm or glm fits coeF(summary(fit)) will extract the 
coefficients matrix.

Had you given an example of what you wanted to extract we could have given 
you an example of how to do it.
PLEASE do!
#
Dear Lars,

In addition to looking at the relevant help pages, a nice thing about R is
that the objects are there for you to examine. For example:
[1] "Df"      "Sum Sq"  "Mean Sq" "F value" "Pr(>F)"
Classes 'anova' and 'data.frame':       11 obs. of  5 variables:
 $ Df     : int  1 1 1 1 1 1 1 1 1 1 ...
 $ Sum Sq : num  817.71  37.59   9.37  16.47  77.48 ...
 $ Mean Sq: num  817.71  37.59   9.37  16.47  77.48 ...
 $ F value: num  116.42   5.35   1.33   2.34  11.03 ...
 $ Pr(>F) : num  5.03e-10 3.09e-02 2.61e-01 1.41e-01 3.24e-03 ...
 - attr(*, "heading")= chr  "Analysis of Variance Table\n" "Response: mpg"

Thus, e.g., av$"Sum Sq" returns the sums of squares:
[1] 817.7129524  37.5939529   9.3709293  16.4674349  77.4757948   3.9493082
 [7]   0.1297687  14.4742372   0.9717105   0.4066688 147.4944300

You could do the same thing with the object returned by summary().

I hope that this helps,
 John

--------------------------------
John Fox, Professor
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox 
--------------------------------