randomization of conditions between groups ONLINE


Author
Message
eleonora_parr
eleonora_parr
Respected Member (490 reputation)Respected Member (490 reputation)Respected Member (490 reputation)Respected Member (490 reputation)Respected Member (490 reputation)Respected Member (490 reputation)Respected Member (490 reputation)Respected Member (490 reputation)Respected Member (490 reputation)
Group: Forum Members
Posts: 40, Visits: 148
Dear inquisiters
I have an experiment with two conditions, which order I have to counterbalance between subjects. I need that both conditions are presented as first the same amount of time in the group that I will test online. To do so, I used this piece of code:
<expt>
/ subjects = (1 of 2)
/ groupassignment = groupnumber
/ blocks = [1=Consent1; 2= demographics; 3= calibration;
            
            4=instructions_imagery;
            5-12= noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            13=endpractice_imagery ;
            
            14=instructions_imagery;
            15-22 = noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            23=endpractice2_imagery;
            
            24-71= list.blockselector_imagery; 72=break; 73-120= list.blockselector_imagery; 121=skit;
            
            122=instructions_pred;
            123-130=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con, practice_5_pain_backward_inc,
                            practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            131=endpractice_pred;
            
            132=instructions_pred;
            133-140=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con, practice_5_pain_backward_inc,
                            practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            141=endpractice2_pred;
            
            142-189=list.blockselector_prediction; 190=break; 191-246=list.blockselector_prediction; 247=endexpt; 248=debrief;]
</expt>

<expt>
/ subjects = (2 of 2)
/ groupassignment = groupnumber
/ blocks = [1=Consent1; 2= demographics; 3= calibration;

            4=instructions_pred;
            5-12=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con, practice_5_pain_backward_inc,
                            practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            13=endpractice_pred;
            
            14=instructions_pred;
            15-22=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con,practice_5_pain_backward_inc,practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            23=endpractice2_pred;
            
            24-71=list.blockselector_prediction;72=break;73-120=list.blockselector_prediction;121=skit;
            
            122=instructions_imagery;
            123-130=noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            131=endpractice_imagery;
            
            132=instructions_imagery;
            133-140=noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            141=endpractice2_imagery;
            142-189=list.blockselector_imagery; 190=break; 191-246=list.blockselector_imagery; 247=endexpt; 248=debrief;]
</expt>        

So that "group 1" would do as first the "Imagery" condition and as second the "Prediction" condition, and "group 2" would be presented with the "prediction" condition as first and the condition "imagery" as second.
 This works perfectly when I try it myself and I type on group "1" or "2", but as the experiment needs to be run online, how do I tell inquisit to assign group 1 and group 2 in an alternate way between subjects, so that I have two equal number of people in both groups?

On the webpage, after having uploaded all the files needed to run the experiement, I usually select these options for the subjectid:

Choose how participant ids should be generated:
QueryParameter
Enter the name of the participant id parameter (default is "subjectid"). TurkPrime users should specify "workerId".
subjectid


What about how the group should be determined? Is it correct to select these options??:

Choose how the group id should be determined:
RandomNoReplace
Enter the total number of groups (if any) in the experiment.
2


Thank you so much!!

Eleonora
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
eleonora_parr - 10/18/2021
Dear inquisiters
I have an experiment with two conditions, which order I have to counterbalance between subjects. I need that both conditions are presented as first the same amount of time in the group that I will test online. To do so, I used this piece of code:
<expt>
/ subjects = (1 of 2)
/ groupassignment = groupnumber
/ blocks = [1=Consent1; 2= demographics; 3= calibration;
            
            4=instructions_imagery;
            5-12= noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            13=endpractice_imagery ;
            
            14=instructions_imagery;
            15-22 = noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            23=endpractice2_imagery;
            
            24-71= list.blockselector_imagery; 72=break; 73-120= list.blockselector_imagery; 121=skit;
            
            122=instructions_pred;
            123-130=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con, practice_5_pain_backward_inc,
                            practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            131=endpractice_pred;
            
            132=instructions_pred;
            133-140=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con, practice_5_pain_backward_inc,
                            practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            141=endpractice2_pred;
            
            142-189=list.blockselector_prediction; 190=break; 191-246=list.blockselector_prediction; 247=endexpt; 248=debrief;]
</expt>

<expt>
/ subjects = (2 of 2)
/ groupassignment = groupnumber
/ blocks = [1=Consent1; 2= demographics; 3= calibration;

            4=instructions_pred;
            5-12=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con, practice_5_pain_backward_inc,
                            practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            13=endpractice_pred;
            
            14=instructions_pred;
            15-22=noreplace(practice_5_safe_backward_con, practice_5_safe_backward_inc, practice_5_pain_backward_con,practice_5_pain_backward_inc,practice_5_safe_backward_con,practice_5_safe_backward_inc,practice_5_pain_backward_con,practice_5_pain_backward_inc);
            23=endpractice2_pred;
            
            24-71=list.blockselector_prediction;72=break;73-120=list.blockselector_prediction;121=skit;
            
            122=instructions_imagery;
            123-130=noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            131=endpractice_imagery;
            
            132=instructions_imagery;
            133-140=noreplace(practice_Congruent_Green, practice_Incongruent_Green, practice_Congruent_Red, practice_Incongruent_Red);
            141=endpractice2_imagery;
            142-189=list.blockselector_imagery; 190=break; 191-246=list.blockselector_imagery; 247=endexpt; 248=debrief;]
</expt>        

So that "group 1" would do as first the "Imagery" condition and as second the "Prediction" condition, and "group 2" would be presented with the "prediction" condition as first and the condition "imagery" as second.
 This works perfectly when I try it myself and I type on group "1" or "2", but as the experiment needs to be run online, how do I tell inquisit to assign group 1 and group 2 in an alternate way between subjects, so that I have two equal number of people in both groups?

On the webpage, after having uploaded all the files needed to run the experiement, I usually select these options for the subjectid:

Choose how participant ids should be generated:
QueryParameter
Enter the name of the participant id parameter (default is "subjectid"). TurkPrime users should specify "workerId".
subjectid


What about how the group should be determined? Is it correct to select these options??:

Choose how the group id should be determined:
RandomNoReplace
Enter the total number of groups (if any) in the experiment.
2


Thank you so much!!

Eleonora

For the group ID, you'll want to choose either sequential generation or random without replacement (RandomNoReplace). In either case, indicate the total number of groups as 2.

Note that you will likely not get an exactly equal distribution of order 1 vs order 2. A group ID is generated and assigned when a potential participant loads the experiment's start page, regardless of whether that person then actually decides to start the experiment at that time or not. Some people will not, i.e. you will see gaps and imbalances.

GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search