[R] Tukey test on ANCOVA

[Michael Rennie]

Michael Rennie wrote:
>
> Hi Paul,
>
> Short answer is you don't. Instead, you estimate your adjusted means, 
> and then compare adjusted means using one of a couple of different 
> options- Quinn and Keough (2002, Experimental design and data analysis 
> for biologists) suggests then using specific comparisons of the 
> adjusted means using planned contrasts (see page 353), followed by and 
> adjustment of your alpha for the # of comparisons if necessary. Sokhal 
> and Rohlf suggest using another method that requires the use of 
> statistical tables. Either way, the equations can easily be programmed 
> into R. I prefer the Q&K method as it's pretty straightforward, and 
> have coded it in R myself. I can send you some code based on the fly 
> example in Q&K if you're interested.
>
> Mike
>
> r-help-request at r-project.org wrote:
>> Message: 118
>> Date: Thu, 25 Mar 2010 19:48:28 -0400
>> From: Paul Simonin<paul.simonin at uvm.edu>
>> To: R Help Listserve<r-help at r-project.org>
>> Subject: [R] Tukey test on ANCOVA
>> Message-ID:<4BABF64C.4020707 at uvm.edu>
>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>>
>> Greetings!
>>
>>      I have one quick question: How do you do a Tukey test on an ANCOVA?
>>
>> Thanks for any tips!
>> -Paul
>>
>>
>>
>> ------------------------------
>