Skip to content

Post hoc analysis for ANOVA with repeated measures

3 messages · Frodo Jedi, David Winsemius, Richard M. Heiberger

#
Dear all,
how can I perform a post hoc analysis for ANOVA with repeated measures (in 
presence of a balanced design)?
I am not able to find a good example over internet in R...is there among you 
someone so kind to give
me an hint with a R example please?

For example, the aov result of my analysis says that there is a statistical 
difference between stimuli (there are 7 different stimuli).
...I would like to know which stimuli are involved.
Error: subject
          Df Sum Sq Mean Sq F value Pr(>F)
Residuals 14 227.57  16.255               

Error: subject:stimulus
          Df Sum Sq Mean Sq F value  Pr(>F)  
stimulus   6 11.695 1.94921  2.3009 0.04179 *
Residuals 84 71.162 0.84717                  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 

Error: subject:condition
          Df Sum Sq Mean Sq F value   Pr(>F)   
condition  1 42.076  42.076  12.403 0.003386 **
Residuals 14 47.495   3.393                    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 

Error: subject:stimulus:condition
                   Df Sum Sq Mean Sq F value Pr(>F)
stimulus:condition  6  9.124  1.5206  1.4465 0.2068
Residuals          84 88.305  1.0513
In attachment you find the table I am using, you can access it by means of:
scrd<-  
read.csv(file='/Users/....../tables_for_R/table_quality_gravel.csv',sep=',',header=T)


The data are from an experiment where participants had to evaluate on a seven 
point likert scale
the realism of some stimuli, which are presented both in condition "A" and in 
condition "AH".


Thanks in advance

Best regards
#
My suggestion is that you take the weekend lull in readership to now  
read the Posting Guide which it appears you nave not yet done. Had you  
done so you should have encountered the advice about how to  
successfully attach data files. (None came through.) You would also  
have read the request that before posting that you search the R-help  
archives for previous questions that match yours. (There are several  
using the search terms "post-hoc repeated measures":

http://search.r-project.org/cgi-bin/namazu.cgi?query=post-hoc+repeated+measures&max=100&result=normal&sort=score&idxname=functions&idxname=Rhelp08&idxname=Rhelp10&idxname=Rhelp02