Hi Professor Granaas!
The observations that your student is collecting is indeed a problem.
Because they are correlated (being collected over time), the standard
errors for the regression approach that he is planning to use are
probably too low creating type I error problems. You should look into a
repeated measures type of design. Given what he is trying to do, it is
difficult to suggest what steps he might take exactly.
Hope this helps.
Gary Winkel
City University of New York
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
