luvongo
05-19-2011, 01:12 PM
Hi everyone,
I really hope I can get some assistance ASAP because I've been struggling with a project for nearly a week and getting little to no help from the people who should be helping me (i.e. my professor, lab partner, etc.). So here I am.
Basically, as a class assignment, we ran a mirror drawing experiment where we had our subjects complete the task in four different permutations: using their dominant hand and dominant eye, their dominant hand and their non-dominant eye, their non-dominant hand and their dominant eye, and their non-dominant hand and their non-dominant eye. We were measuring the possible main effects and/or interactions of eye dominance and condition (i.e. any of the four permutations) on both time to complete the task and error rate. (So a 2 x 2 design.)
My professor initially said that I don't have to do a MANOVA because it's probably too complicated (my DVs might not be moderately correlated enough), and that I should just do a one-way ANOVA with each IV separately and then show the estimated marginal means graph. So I did.
The one-way ANOVA showed a significant effect of condition on time but not error rate, but no significant effect on eye dominance on either time or error rate. However, when I ran the estimated marginal means graph in SPSS, it appeared to do its own two-way ANOVA, and while it confirmed the results of the first one-way ANOVA, it added that there was a significant interaction of eye dominance and condition on time, and the post-hoc tests showed different results. (Not contradictory results, mind you, just more significant mean differences between conditions than the one-way.)
I made my estimated marginal means graph two ways, first with eye dominance on the X-axis and then with condition on the X-axis. My professor said that the latter was wrong. The former does look like it would be the better graph to use for the purpose of this project. However, I'm still very confused about what exactly it is that I'm looking at. I'm having trouble putting my exact results into English words, and because the univariate analysis I used to make the graph is not exactly a one-way ANOVA, I'm confused as to how to report it (i.e. what degrees of freedom, etc.).
If someone could get back to me here, or send me a private message, I can show you my output in a word document so you can tell me if it makes sense to you and walk me through it. The paper is due tomorrow, and I've exhausted all of my other available help resources, so I'd be super-appreciative!
Thanks in advance! :)
I really hope I can get some assistance ASAP because I've been struggling with a project for nearly a week and getting little to no help from the people who should be helping me (i.e. my professor, lab partner, etc.). So here I am.
Basically, as a class assignment, we ran a mirror drawing experiment where we had our subjects complete the task in four different permutations: using their dominant hand and dominant eye, their dominant hand and their non-dominant eye, their non-dominant hand and their dominant eye, and their non-dominant hand and their non-dominant eye. We were measuring the possible main effects and/or interactions of eye dominance and condition (i.e. any of the four permutations) on both time to complete the task and error rate. (So a 2 x 2 design.)
My professor initially said that I don't have to do a MANOVA because it's probably too complicated (my DVs might not be moderately correlated enough), and that I should just do a one-way ANOVA with each IV separately and then show the estimated marginal means graph. So I did.
The one-way ANOVA showed a significant effect of condition on time but not error rate, but no significant effect on eye dominance on either time or error rate. However, when I ran the estimated marginal means graph in SPSS, it appeared to do its own two-way ANOVA, and while it confirmed the results of the first one-way ANOVA, it added that there was a significant interaction of eye dominance and condition on time, and the post-hoc tests showed different results. (Not contradictory results, mind you, just more significant mean differences between conditions than the one-way.)
I made my estimated marginal means graph two ways, first with eye dominance on the X-axis and then with condition on the X-axis. My professor said that the latter was wrong. The former does look like it would be the better graph to use for the purpose of this project. However, I'm still very confused about what exactly it is that I'm looking at. I'm having trouble putting my exact results into English words, and because the univariate analysis I used to make the graph is not exactly a one-way ANOVA, I'm confused as to how to report it (i.e. what degrees of freedom, etc.).
If someone could get back to me here, or send me a private message, I can show you my output in a word document so you can tell me if it makes sense to you and walk me through it. The paper is due tomorrow, and I've exhausted all of my other available help resources, so I'd be super-appreciative!
Thanks in advance! :)