latin square design with within and between subjects

[kali3]

Hi everyone,

I have ran an experiment that uses an incomplete latin square design and I am not sure how to run it with SPSS.
So I've got 3 tasks, called R, I and P, but participants only completed two of them (which gave me 6 possible combinations: IR, RI, IP, PI, RP, PR).
They then repeated the tasks in the same order, so I have a repeated measure between time 1 and time 2.

How can I set up my data to analyse it correctly with SPSS?

Thank you so much for your help!

Cheers

[Bruce Weaver]

Here's an old manual from the SPSS Training Dept with some advice on analyzing Latin square designs (see chapter 10).

  http://www.planta.cn/forum/files_planta/anova_for_spss_125.pdf

But nowadays, it probably makes more sense to use the MIXED procedure.  If you go that route, you'll need a LONG data file with one row per measurement of the DV, with variables coding ID and all of the factors in your model.  

kali3 wrote

Hi everyone,

I have ran an experiment that uses an incomplete latin square design and I am not sure how to run it with SPSS.
So I've got 3 tasks, called R, I and P, but participants only completed two of them (which gave me 6 possible combinations: IR, RI, IP, PI, RP, PR).
They then repeated the tasks in the same order, so I have a repeated measure between time 1 and time 2.

How can I set up my data to analyse it correctly with SPSS?

Thank you so much for your help!

Cheers

[kali3]

Hi Bruce and thanks for your quick response. I am looking at doing Mixed models but as my design is fairly complex, I can't find all the information I need to make the right decision when selecting the different options in SPSS. Would you have a reference that uses nested models with mixed design that I could use as an example?

Thanks

[Bruce Weaver]

No, sorry, I don't have a reference for that, and don't find anything obvious with a quick Google search.  But for more old-school ANOVA approaches, you might find something useful here:

http://www.iasri.res.in/ebook/EB_SMAR/

HTH.

kali3 wrote

Hi Bruce and thanks for your quick response. I am looking at doing Mixed models but as my design is fairly complex, I can't find all the information I need to make the right decision when selecting the different options in SPSS. Would you have a reference that uses nested models with mixed design that I could use as an example?

Thanks