Mike Scoles wrote:
> Although I have taught statistics for almost 20 years, I still don't understand
> the relevance of directional alternate hypotheses. The critical region of the
> test statistic is determined by the statistic's distribution given that the null
> hypothesis is true. This conditional distribution has little to do with the
> investigator's opinion of which direction it should (or more accurately, could)
> be false.
When I took my first stats class (from a psychologist), it was explained to
me that you'd use a one-tail if you have the directional hypothesis a priori - this
way you'd capitalize on chance only half the time. But this works only when you
acept as "nonsignificant" something that came out in the opposite direction - in
which case you'd were supposed to say to yourself - "Oh Pooh! I guess I'll put this
study in the file-drawer and assume it's non-significant." I wouldn't do that!
Perhaps a better way is to report the actual p value from a two-tailed
test, and in the language of your paper use language that reflects where you are on
the distribution (e.g. clearly significant, marginal, worthy of note, and so forth).
In my undergrad methods/stats class I do mention both - but stress that the
true guardian against false claims is the replication process.
--
---------------------------------------------------------------
John W. Kulig [EMAIL PROTECTED]
Department of Psychology http://oz.plymouth.edu/~kulig
Plymouth State College tel: (603) 535-2468
Plymouth NH USA 03264 fax: (603) 535-2412
---------------------------------------------------------------
"What a man often sees he does not wonder at, although he knows
not why it happens; if something occurs which he has not seen before,
he thinks it is a marvel" - Cicero.
