# IA Inferential Tests Help

## Recommended Posts

I can reassure you that, out of all the IAs I've done, Psych IA is the worst...hahahaha. Anyway, I'm doing my experiment on the levels of processing. I have three groups, and I'm comparing how many words each group recalled. I'm almost done, but I'm stuck with the inferential tests. My book states that I have to choose one out of Chi-squared, Mann Whitney-U, and Wilcoxon (I think). None of those 3 seem to suit my data because they all require two groups. So, I'm confused. I think my data is ordinal. I'm not really sure about that, either. Help would be much appreciated

##### Share on other sites

If your design was independent measures (one group did only one condition) then since your data are interval/ratio (e.g. time, weight, height) you would use multiple independent t-tests or an ANOVA. Since ANOVA is pretty advanced I recommend you do the multiple t-tests. The tests you mentioned are for ordinal (rankings like 1-7) and nominal data and are not relevant to your data set. You might want to limit your groups to just a deep and superficial level of processing and ignore the third group.

##### Share on other sites

If your design was independent measures (one group did only one condition) then since your data are interval/ratio (e.g. time, weight, height) you would use multiple independent t-tests or an ANOVA. Since ANOVA is pretty advanced I recommend you do the multiple t-tests. The tests you mentioned are for ordinal (rankings like 1-7) and nominal data and are not relevant to your data set. You might want to limit your groups to just a deep and superficial level of processing and ignore the third group.

I am doing my experiment on the serial position curve. There are two groups, where one group is presented with a list of 20 words faster than the other group. Would I be doing the multiple t-test, too? I am done with my report, just have this one part left.

##### Share on other sites

I am doing my experiment on the serial position curve. There are two groups, where one group is presented with a list of 20 words faster than the other group. Would I be doing the multiple t-test, too? I am done with my report, just have this one part left.

Yes you would also use the independent groups t-test, you can use the link provided in the previous post .

##### Share on other sites

I am doing my experiment on the serial position curve. There are two groups, where one group is presented with a list of 20 words faster than the other group. Would I be doing the multiple t-test, too? I am done with my report, just have this one part left.

Yes you would also use the independent groups t-test, you can use the link provided in the previous post .

Oh, thank you But one more question! Do I have to do it manually (and put the calculations in the appendix), or can I just use the calculator on the net (with no calculations in the appendix)?

##### Share on other sites

Oh, thank you But one more question! Do I have to do it manually (and put the calculations in the appendix), or can I just use the calculator on the net (with no calculations in the appendix)?

I would put the calculations in to be safe, the syllabus says that the appendix should contain the raw data and calculations. I don't know how detailed these calculations need to be though. Try looking at some sample IAs to get an idea, there are some on my blog and on this forum.

## Create an account

Register a new account