order effects (Simon Makin )

Subject: order effects
From:    Simon Makin  <s.makin(at)DCS.SHEF.AC.UK>
Date:    Mon, 3 Jun 2002 19:52:21 +0100

Hi hearing folks,

I have a quick question about experimental design:

I have three related stimuli sets I want to analyse together, but due to
not
really being able to shut subjects up in a booth for hours at a time I
was
planning to seperate them into three seperate sessions. Now - the kind
of
stimuli I'm using are known to be subject to training effects, so is it
enough to counterbalance by assigning each subject to a different,
random,
order of sessions (1,2,3 - 1,3,2 - 2,1,3 - 2,3,1 etc..), or do I have to
be
careful to assign the same number of subjects to each of the possible
six
orders? In which case, what do I do if I can't get a multiple of 6
subjects?

Thanks in advance for any advice...

Cheers,

Si.