In article <firstname.lastname@example.org>, email@example.com (Meriel Huggard) wrote:
> They are then asked to examine if there is a significant difference > between, say, the third column and the fourth, fifth and sixth columns. > > The solution set I have been given takes each pair of columns in turn > (col 3 and col 4; col 3 and col 5 and col 3 and col 6) and in each case > does a two sample t-test on the means. > > What I would like to know is: > > Is this method better than a one way ANOVA for the relevant columns ? > (And if so, why ?)
In order to answer this question you need to explore what happens with an ANOVA test.
When you get a significant result from an one way ANOVA, what is the strongest statement you can make about differences between the means?
When you answer that question, you can answer the question you pose above.
+=============================================================+ Paul C. Bernhardt, M.S. in Psychology, University of Utah PhD graduate student in Educational Psychology +=============================================================+