slyly
04-03-2006, 11:37 AM
Hi,
I need to analyze a linguistics experiment we did. It was a 2x2 mixed design (one factor: word order, second factor: stress placement). 32 subjects took part and each judged 5 sentence of each combination on how good it was on a level on 1-6, where 1 is the worst and 6 is the best.
My first question is if people agree that as a linguistic acceptability task, this experiment needs a nonparametric test.
More importantly, however, I'm confused as to what test I should use. Based on what I've read, it seems like the Friedman's ANOVA test may be correct for this, but I'm not confident that this is the correct choice.
Any advice or suggestions would be most appreciated.
Betsy
I need to analyze a linguistics experiment we did. It was a 2x2 mixed design (one factor: word order, second factor: stress placement). 32 subjects took part and each judged 5 sentence of each combination on how good it was on a level on 1-6, where 1 is the worst and 6 is the best.
My first question is if people agree that as a linguistic acceptability task, this experiment needs a nonparametric test.
More importantly, however, I'm confused as to what test I should use. Based on what I've read, it seems like the Friedman's ANOVA test may be correct for this, but I'm not confident that this is the correct choice.
Any advice or suggestions would be most appreciated.
Betsy