Hi there,
in the documentation / forum I could only find the / subjects command alternating between conditions and the random command. Given that we have a relatively small sample size we'd like to predefine a random sequence and then simply allocate participants to the conditions (e.g., 1,2,1,1,1,1,1,2,2,1,2,2,2,2). Is this possible?
Thanks,
Rene
Not automatically. You would indeed have to do this *manually*. I.e., if you're running participants in the lab, simply enter odd and even ids as needed. If on the web, distribute launch page links to participants which include id as query parameter.
Thanks Dave. Just to make sure I understood you correctly. I'd use "/subject = (1 of 2)" for odd and "/subject = (1 of 2)" for even subject IDs. To stick with my example, this would mean that I would run the first seven participants as subjects 1,2,3,5,7,9,11 and then proceed with 4 and 6 for the next two and so forth. Is that correct?
If yes, wouldn't it be easier to run the two conditions as one-conditional experiments? Or are there drawbacks when it comes to combining the data?
Best
No, that would of course be *(2 of 2)* for even numbers.
Yes.
Not sure what that means or why it would be easier?
Thanks for the quick reply. 2 of 2 is what I meant. Copy & paste...
As for the alternative: Instead of programming one experiment with two conditions I could program two experiments with one condition. Instead of doing it the way described above we simply open experiment 1 for condition 1 trials and experiment 2 for condition two trials. This way it would be easier to keep track of the amount of participants (i.e., subject ID always equals N) and it might be less error prone. Or am I overlooking something?
Yes, you can do that if you want to. It may be less error prone in some respects, but more in others (suppose you discover a mistake in the procedure, but forget to fix it in one version; suppose you decide to modify certain aspects for a follow-up study: now you need to make sure to keep changes in sync across several files, i.e. doubling the effort, needless redundancy, etc.).
However, those are really questions of lab and participant managing, and as such they are really quite unrelated to Inquisit. I say: Do whatever fits your lab's workflow.