select subset of stimuli on a basis of participants' ratings


Author
Message
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: 13K, Visits: 104K

The approach would be the same, although perhaps you'd want to base things on certain ranges, i.e. put all items with responses between, say, -100 to -90 in a single bin, etc.
Regards,
~Dave


mario
mario
Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)
Group: Forum Members
Posts: 7, Visits: 1

Thanks Dave,


the itemcount property helped a lot!



What I'm still interested in is how to make it when the rating is done with a slider. The solution with an empty itemlist for every possible response works fine with my -5 to +5 scale but is not very applicable for a slider from -100 to +100 with an increment of 1.


Any ideas if there's a way to do that with inquisit?



best, Mario



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: 13K, Visits: 104K

You'll probably want to use the 'itemcount' property. See the documentation for details.


<block myblock>
/ trials = [1-10=mylikert]
</block>

<likert mylikert>
/ ontrialend = [if(likert.mylikert.response=="1")item.r1.item=text.mytext.currentitem]
/ ontrialend = [if(likert.mylikert.response=="2")item.r2.item=text.mytext.currentitem]
/ ontrialend = [if(likert.mylikert.response=="3")item.r3.item=text.mytext.currentitem]
/ ontrialend = [if(likert.mylikert.response=="4")item.r4.item=text.mytext.currentitem]
/ ontrialend = [if(likert.mylikert.response=="5")item.r5.item=text.mytext.currentitem]
/ stimulusframes = [1=mytext,debug]
/ anchorwidth = (4%)
</likert>

<text mytext>
/ items = myitems
/ position = (50%, 40%)
</text>

<item myitems>
/ 1 = "A"
/ 2 = "B"
/ 3 = "C"
/ 4 = "D"
/ 5 = "E"
/ 6 = "F"
/ 7 = "G"
/ 8 = "H"
/ 9 = "I"
/ 10 = "J"
</item>

<text debug>
/ items = ("R1=<%item.r1.itemcount%>|R2=<%item.r2.itemcount%>|R3=<%item.r3.itemcount%>|R4=<%item.r4.itemcount%>|R5=<%item.r5.itemcount%>")
/ position = (50%, 30%)
</text>

<item r1>
</item>

<item r2>
</item>

<item r3>
</item>

<item r4>
</item>

<item r5>
</item>


Regards,


~Dave


mario
mario
Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)Partner Member (801 reputation)
Group: Forum Members
Posts: 7, Visits: 1

Hi Community,


I'm interested if there are any solutions for the problem in this old post by now. Any ideas how to implement safeguards in case of bins with less than 3 items?


I was thinking about putting something like this in the block that is presenting the selected items:


/ onblockbegin = [if (item.pret.min5.item.1 == "") item.pret.min5.item= item.pret.min4.item.1]
/ onblockbegin = [if (item.pret.min5.item.2 == "") item.pret.min5.item= item.pret.min4.item.2]
/ onblockbegin = [if (item.pret.min5.item.3 == "") item.pret.min5.item= item.pret.min4.item.3]


But that seems not to be a very smart solution and would be a lot of code to apply to all possible cases...


I'm especially interested in a solution that would even work with a scale of higher resolution like a slider with values from -100 to +100.


Thanks for any ideas.


Mario


Robert Balas
Robert Balas
Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)
Group: Forum Members
Posts: 38, Visits: 85
Dave,

I'm sorry that I did not respond to this. Just to much work - hope that never happens again.

the thing is that I resigned from having participant based selection of stimuli to the study. It seemed to me too much of a task to solve as for the inquisit beginner.

I'll will return to the issue shortly though. I'll keep you posted on this....

All the best,
Robert
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: 13K, Visits: 104K
Hey Robert,

any progress with this? I might be doing something similar fairly soon, so if you could share your experiences (or even your script if that's possible) that would be very helpful.

Thanks,
~Dave

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: 13K, Visits: 104K
Cool - please keep me posted about your progress. This is an interesting, little problem and I'd really like to see the result once it's finished.

Cheers,
~Dave

Robert Balas
Robert Balas
Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)
Group: Forum Members
Posts: 38, Visits: 85
Dave,

Thanks - that works fine now.

Cheers,
Robert
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: 13K, Visits: 104K
Robert,

you'll have to use *keyboard scancodes* rather than the literal key values (i.e. letters) in your '/ ontrialend' attributes. If you write

===pretest trial
<trial pic_pretest>
/ stimulusframes = [1=noreplace(CS_US_pretest); 61=skala_pretest]
/ responseframe = 61
/ response = timeout(3000)
/ validresponse = ("a", "s", "d", "f", "g", "h", "j", "k", "l", ";", "'")
/ ontrialend = [if (trial.pic_pretest.response == "30") item.pret.min5.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "31") item.pret.min4.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "32") item.pret.min3.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "33") item.pret.min2.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "34") item.pret.min1.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "35") item.pret.0.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "36") item.pret.1.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "37") item.pret.2.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "38") item.pret.3.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "51") item.pret.4.item = text.CS_US_pretest.currentitem]
/ ontrialend = [if (trial.pic_pretest.response == "43") item.pret.5.item = text.CS_US_pretest.currentitem]
/ trialcode = "pretest_bodzcow"
</trial>

the script should work as intended (please make sure you double-check these scancodes I inserted).

Best,
~Dave

Robert Balas
Robert Balas
Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)Guru (5.5K reputation)
Group: Forum Members
Posts: 38, Visits: 85
Dear Dave,

I've tried to follow your suggestions, but I guess I must have done sth badly, cause it doesn't seem to work at all.

Could you have a quick look on the attached script? I did the pretest of stimuli (pictures) rated by pressing corresponding keyboard. what I wanted to have are the stimuli sorted into item sets based on participants responses. I've implemented page that is supposed to report how many items there are in particular item sets. Very simple, but doesn't do what it is supposed to.

Am I missing sth in the script?

All the best,
Robert
Attachments
EC_part_chooses.zip (397 views, 770 bytes)
GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search