Skip to content

Two Different Interaction Terms for Mixed Factorial ANOVA in R?

2 messages · Angela Radulescu, Richard M. Heiberger

#
I have the following R output from a mixed factorial ANOVA:

Error: subj
                 Df Sum Sq Mean Sq F value  Pr(>F)   group
1   11.3   11.26   0.449 0.50811
singleType        1    0.7    0.66   0.026 0.87220   group:singleType
1  237.5  237.53   9.484 0.00461 **Residuals        28  701.3   25.04
                 ---Signif. codes:  0 ?***? 0.001 ?**? 0.01 ?*? 0.05
?.? 0.1 ? ? 1
Error: subj:singleType
                 Df Sum Sq Mean Sq F value Pr(>F)
singleType        1 1566.4  1566.4 411.445 <2e-16 ***group:singleType
1    3.4     3.4   0.893  0.352    Residuals        30  114.2     3.8
                 ---Signif. codes:  0 ?***? 0.001 ?**? 0.01 ?*? 0.05
?.? 0.1 ? ? 1
Error: Within
            Df Sum Sq Mean Sq F value Pr(>F)Residuals 3286   5747   1.749

Here "group" is the between factor (2 levels) and "singleType" the within
factor (2 levels). I'm not sure why there are two group:singleType
interaction terms. Any help would be much appreciated!

For reference, here is the original aov call:

anova.p = aov(data = dat.colSingle, rt ~ (group*singleType) +
Error(subj/singleType))
summary(anova.p)

Thank you!
#
Angela,

My guess is that your data are not balanced.
That could be due to a typo in one of the factors, or it could be that
you actually have different numbers of observations at some of the
factor levels.

Rich

On Sun, May 17, 2015 at 12:05 PM, Angela Radulescu
<angela.radulescu at gmail.com> wrote: