"Reapply" item properties after random draw


"Reapply" item properties after random draw

Author
Message
fflade
fflade
Expert
Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)

Group: Forum Members
Posts: 11, Visits: 117
Hi Dave,

And again I hope for a share of your wisdom (i.e. programming solutions). And here it goes:
I have a range of item lists with diverse length, lets call them A (2 stimuli, animals), B (3 stimuli, flowers) and C (4 stimuli, insects). Now, I want to do several things with a random subsample of these lists. so far I managed to draw a random subsample S (e.g. 4 out of the 9 stimuli, irrespective of their original category), and display and store that subsample as an item list for further use. Per participant, the whole experiment only runs with items from S. Now, however, I want to "rematch" the stimuli from that subsample with the appropriate questions, e.g. "What animal is this?", "What flower is this?". Unfortunately, the draw is random and the original "animal" etc. "property" of items in S is lost, so I don't know how to match the questions to the appropriate stimuli. Can the items in S be somehow compared to the ones in A, B and C and when there is a match, the item is identified and can, on the basis of this, be paired with the correct question? Thank you for any ideas on this!

Cheers,
Felicitas

Dave
Dave
Supreme Being
Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)

Group: Administrators
Posts: 9.1K, Visits: 39K
I would set up a corresponding (initially emtpy) item list Q and store the question-type associated with the randomly chosen item that you store in S there. I.e., do those things in parallel at the very start and link the ultimate stimulus elements S and Q as you would do normally (cf. the "How to present stimulus pairs" topic in the documentation).

Hope this helps.

fflade
fflade
Expert
Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)Expert (1.2K reputation)

Group: Forum Members
Posts: 11, Visits: 117
Thanks! Now, there is only one thing missing: the corresponding "item" in Q should not be a text, but a certain set of likert scale anchors.

Dave
Dave
Supreme Being
Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)Supreme Being (493K reputation)

Group: Administrators
Posts: 9.1K, Visits: 39K
Hmm, I'm afraid it's not possible to set <likert> anchors or -- more likely -- their labels in this fashion. What you can do, however, is use standard <text> elements *instead* of the <likert>'s /anchors attribute. You'd simply position them as needed and display them via the <likert>'s /stimulustimes or -frames just like your other stimuli.

GO


Reading This Topic


Login
Existing Account
Email Address:


Password:


Social Logins

Select a Forum....






Millisecond Forums


Search