With just two treatments, it is immaterial whether
you call a factor nominal (unordered) or ordinal.
Otherwise put, I understand that Jonckheere-Terpstra
is equivalent to Mann-Whitney for two treatments.
However, you don't say and I can't guess how
you intend to take the differences in players into
account.
Nick
[email protected]
[email protected]
> We ran an experiment with 24 subjects for each treatment. They had to
> choose how to invest their initial endowment. They had the
> opportunity to
> invest this sum both in a public good and in a private good. We ran
> different treatments where the game was repeated for 8
> periods. If I want
> to compare the average contribution to the public good in two
> different treatments, is the Mann-Whitney test suitable (we have
> different players
> in each treatment)? Should the Jonckheere test provide a better
> performance?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
