randomize and counterbalance 3 levels of 1 IV

SuperLab SuperLab 4 Support
1

Maybe this is a dumb question, but I can’t seem to figure this out:

I have 3 versions of 18 different stimuli (all text vignettes, one response needed for each, same question every time). I want each subject to get 1 version of each stimulus, so in the end they have seen 6 of each level of the IV (18 trials total). Each subject should get all stimuli in a random order.

This can be done, right?

Thanks!

2

A colleague and I went over this and we are having a hard time understanding. Is there anyway you can post your experiment as an Experiment Package so we can get a better idea? The experiment package feature is found under the File menu within SuperLab.

3

i don’t have the experiment made because i don’t know how to do it. the main thing is that any subject can only see one vignette out of a set of three, and we have 18 sets. there is one IV: vignette type, and one DV: a judgment of our relevant dimension.

this is a very typical within subjects repeated measures design. all subjects get exposed to all three levels of the IV - we have 18 vignettes that are each manipulated three ways.

i need to have each subject get 6 of each vignette type, but see only one token from each set, for a total of 18 trials. make sense?

thanks so much!

4

i think hisham was going to try and write a macro to do this at some point, but for now i just created three participant groups and they get assigned to one of three stimulus lists i created.

i encourage you guys to include counterbalancing in the program - i was really surprised to learn SL couldn’t handle a repeated measures design like this!

5

  • Four code values are needed
    • 1st code value “Ok to present condition 1” with values “Yes” and “No”

    • 2nd code value “Ok to present condition 2” with values “Yes” and “No”

    • 3rd code value “Ok to present condition 2” with values “Yes” and “No”

    • 4th code value “Trial Type” with values “Condition 1” and “Condition 2” and “Condition 3”

  • All three trials have the code values tagged

  • You need three macros, one for each condition

  • Each [B]Expression[/B] is set to "Number of Trials Presented in This Block is greater than or equal to 6 trials" and marked with either condition 1, 2 or 3
  • Under the[B] If True[/B] tab, SuperLab will "Change value of the code to "Ok to present condition 1, 2, or 3" to No" in all trials in this block"

Forums Jasper 1249.sl4 (10.7 KB)

6

thanks so much for working on this! i’ve been running the experiment the way i described above, but i will certainly try this out.