Adding 'weight' to trials in blocks


Author
Message
SWQ27
SWQ27
Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)
Group: Forum Members
Posts: 9, Visits: 1

Hey there,


In a block, I usually want my trials to be selected by the noreplace selectionmode. However, usually I want some types of trial to be presented more often than others. For example, in a 36-trial block I may want 20 trials to be of type1 and the remaining 16 to be type 2's.


Now, how do I add this weight of trials to the trials attribute? I can't simply type in


/trials = [1-36=noreplace(type1, type2)]


Because this way the 20/16 distribution isn't respected. I also can't use


/trials = [1-20=type1; 21-36=type2]


Because this pattern makes it way too easy for participants to guess the next trial and it might evoke habituation of their responses.


My current solution is to use the first attribute and to type out all the trials. Thus, between the parenthesis I enter type1 twenty times and type2 sixteen times, with copy paste of course.


However, there is more than one block and sometimes a block contains way more trials. This makes my own solution quite inefficient. Does anyone know how to enter two (or more) types of trials and tell Inquisit how often they should be presented in a block?


Best,


SWQ


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

At the block level, Inquisit sets up its selection pools according to trial proportions. As you noted


/ trials = [1-36=noreplace(type1, type2)]


will result in 18 type1 trials and 18 type2 trials in random order, the respective trial ratio being 1:1.  Now, in your example, you want to have 20 type1 trials and only 16 type2 trials. By way of some simple math we can see that the desired trial ratio is 20:16=1.25. Luckily 20 and 16 have a common denominator (4), and thus we get the same ratio via 5:4=1.25. Consequently,


/ trials = [1-36=noreplace(type1, type1, type1, type1, type1, type2, type2, type2, type2)]


will give you the desired result.


Best wishes from a fellow Inquisit user,


~Dave


SWQ27
SWQ27
Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)Expert (1K reputation)
Group: Forum Members
Posts: 9, Visits: 1

Wauw, that's quite a smart/mathematical solution.


Thank you so much for your help. This still means I'll have to enter by hand but it's doable now and I don't mind :)


SWQ27


GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search