Thread Tools  
johnh
Junior Member
 
Join Date: Mar 2008
johnh is on a distinguished road Posts: 19
Randomise blocks in v5?, Posted March 8th, 2014 #1
Can you randomise the order of blocks in Superlab 5?
Reply With Quote
MonikaT
Senior Member
 
Join Date: Mar 2008
MonikaT is on a distinguished road Posts: 1,279
Posted March 11th, 2014 #2
You can use Participant Groups to accomplish this behavior, which is found under the Experiment menu within SuperLab. For example, you can have 3 groups that all see 100% of the blocks, but each group sees the blocks in a different order.

When creating the groups you need to manually shuffle the blocks into the desired order. You can do this for as many groups as you wish. Attached is a demo experiment that has 3 groups, and each group runs through the blocks in a different order.
Attached Files
File Type: sl5 participant_groups_reordering_blocks.sl5 (4.7 KB, 7 views)
Reply With Quote
johnh
Junior Member
 
Join Date: Mar 2008
johnh is on a distinguished road Posts: 19
Posted March 11th, 2014 #3
Thanks. That helps a bit but I have four blocks that I want to randomise for every participant.

Based on your example I would need to create the 24 different combinations and then run one of those at random for each participant. I assume that's the best Superlab can offer which isn't great. Especially as I actually have 4 repetitions of blocks within the same experiment that all need to be randomised!

I'm thinking if I use Superlab I will have to use the 24 combinations method and run it four times per participant.

Unless I'm missing something?

Thanks again.
Reply With Quote
MonikaT
Senior Member
 
Join Date: Mar 2008
MonikaT is on a distinguished road Posts: 1,279
Posted March 12th, 2014 #4
It would be helpful if you can post your experiment so I can get a better idea how it's set-up. Please post your experiment as an Experiment Package. This can be found under the File menu within SuperLab.
Reply With Quote
johnh
Junior Member
 
Join Date: Mar 2008
johnh is on a distinguished road Posts: 19
Posted March 12th, 2014 #5
I've not set the experiment up yet. I'm just looking at feasibility. It looks like it canít be done 'properly' using Superlab so I'll probably use E-Prime. That's a shame given we've recently bought 30 SuperLab licenses!

The concept is simple. There are 16 blocks of trials. The blocks are grouped into 4: blocks 1-4, blocks 5-8, blocks 9-12 and blocks 13-16. The order of blocks within each group has be randomised.

For the first group of 4 blocks there are 24 different combinations:

{1,2,3,4} {1,2,4,3} {1,3,2,4} {1,3,4,2} {1,4,2,3} {1,4,3,2} {2,1,3,4} {2,1,4,3} {2,3,1,4} {2,3,4,1} {2,4,1,3} {2,4,3,1} {3,1,2,4} {3,1,4,2} {3,2,1,4} {3,2,4,1} {3,4,1,2} {3,4,2,1} {4,1,2,3} {4,1,3,2} {4,2,1,3} {4,2,3,1} {4,3,1,2} {4,3,2,1}

Likewise there are 24 combinations for the second group, 24 for the thirs, and 24 for the final group. That makes 96 different ways to run the experiment and would require 96 different Participant Groups to be set up.

In addition we would need to allocate participants to one of the variations at random which would have to be done outside of Superlab.

This is all a bit messy and time consuming.

Superlab really needs a randomise blocks feature!


Sorry if that sounds a bit negative. I do appreciate your help Monika.
Reply With Quote
MonikaT
Senior Member
 
Join Date: Mar 2008
MonikaT is on a distinguished road Posts: 1,279
Posted March 13th, 2014 #6
Your feedback has been noted. We will keep it in mind for future releases.
Reply With Quote
Thread Tools
Posting Rules
Forum Jump:
© Copyright Cedrus Corporation, 1991-2014.
All times are GMT -7. The time now is 08:14 PM.