Subject assignment issue


Author
Message
Keith Leavitt
Keith Leavitt
Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)
Group: Forum Members
Posts: 27, Visits: 1

Hi,


I found the previous post about random assignment via subject number for 4 conditions. However, I still have a question based upon what I found with a previous study:


If I use subjects=1 of 4, 2 of 4, 3 of 4, 4 of 4, can this really lead to close to equal sample sizes? Won't this assign subjects whose number ends in 1, 5 or 9 to condition 1, subjects whose number ends in 2,6, or 0 to condition 2,  ...and 3 or 7 to condition 3, and 4 or 8 to condition 4? This would mean that 30% of subjects would end up in condition 1, 30% in condition 2, but only 20% each in conditions 3 and 4, right?


I'm about to relaunch, and I just want to make sure I get relatively equal sample sizes...



Keith Leavitt
Keith Leavitt
Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)
Group: Forum Members
Posts: 27, Visits: 1

BTW--in the previous study I used sequential subject numbers for assignment...


Dave
Dave
Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)
Group: Administrators
Posts: 12K, Visits: 98K

Won't this assign subjects whose number ends in 1, 5 or 9 to condition 1, subjects whose number ends in 2,6, or 0 to condition 2,  ...and 3 or 7 to condition 3, and 4 or 8 to condition 4? This would mean that 30% of subjects would end up in condition 1, 30% in condition 2, but only 20% each in conditions 3 and 4, right?


Err, no. But see for yourself by running the attached script / uploading it to your web account (script originally posted in this thread).


~Dave


BTW: As far as I know, Inquisit uses a simple modulo function for condition assignment based on subject number. Let's say there are n=4 conditions. Now compute {subjectnumber mod n}. The modulo function returns the remainder of the division subjectnumber/condition. If the result is 0, the subject will receive the nth condition (i.e. 4th condition in this example). If the remainder is 1, the participant will be assigned to condition 1. If the remainder is 2, off you go in condition 2. Lastly if the remainder is 3 -- well, you guessed it...


Attachments
conditioncheck.exp (859 views, 4.00 KB)
Keith Leavitt
Keith Leavitt
Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)Esteemed Member (2.2K reputation)
Group: Forum Members
Posts: 27, Visits: 1

OK; that generally makes sense. Thanks for the help!


GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search