|
Sensory Computer Systems Forums Welcome to the Sensory Computer Systems Online Forums!
|
View previous topic :: View next topic |
Author |
Message |
Jane
Joined: Dec 21 2005 Posts: 5
|
Posted: Fri May 26, 2006 10:28 am Post subject: Many samples, shown in random Pairs ? |
|
|
Many samples, shown in random Pairs ?
We want to do a Split Plate sample study, where our samples are presented in pairs, random and balanced pairs. We want each panelist to see all possible pairs. We have 10 samples. One pair per day per panelist. We know it going to be a long and important study, we just want to make sure we do it right. Can you help? Any comments or hints are appreciated.
Thanks, Jane |
|
Back to top |
|
|
Karen
Joined: May 26 2006 Posts: 3
|
Posted: Fri May 26, 2006 2:01 pm Post subject: |
|
|
Hi Jane -
Your study has 10 samples, presented in sample pairs to the panelists. This is a 10C2 Experimental Design. You said you'd like every panelist to see every possible pair, which is 45 pairs.
Setup your SIMS 2000 Experimental Design as a 10C2 R45. This is 10 Samples, Presenting 2, 45 Reps. You'll need an Experimental Plan which has exactly 45 Blocks. Good news, SIMS 2000 already has a built in Experimental Plan that should fit perfectly. COCHRAN & COX, PLAN 11.14
Number of Samples............: 10
Number of Reps in Plan.......: 1
Number of Blocks.............: 45
Number Presented.............: 2
45 possible pairs...:
Steps:
1. Create your ballot/questionnaire. Standard layout, questions seen "with all samples".
2. Create your SIMS 2000 Experimental Design with 10 samples, 45 Reps, n panelists/respondents.
Also Select: Block Presentation: As In Plan
Sample Presentation: Randomized
See Sample Information Tab, keep it simple by selecting Same Blinding Codes for Reps and Same Blinding Codes Panelists. You'll have 10 unique Blinding Codes.
(Optional, maybe you want to present your ballot in Profile Mode. Profile Mode allows responses in any sample/attribute order. See Picture #11 & #12: http://www.SensoryComputerSystems.com/SIMSNetmeet/ )
3. Create your SIMS 2000 Test Definition.
Please review Rotation Plan report for acceptability.
4. Create your SIMS 2000 Test Data Result Master.
All of your panelists will see all 45 pairs, and will only be doing one pair of samples per day. Chances are that your panelists will be doing the study in varying progress, some ahead of others by a number of days.
Therefore, Execute the Test with:
Select 'Allow Continuous Presentation Over Reps'
(which will automatically set 'Force All Panelists to Exit After Every 2 Samples')
Pretty good looking Experimental Design. Balanced and near perfect rotation of presentation orders. Of course you may opt for less randomization when creating your design. Just be sure to validate your rotation plan report to your expectations.
Code: |
--> ROTATION PLAN - ACCEPTABLY BALANCED ! <--
(tolerance=1)
Pos 1 Pos 2 Totals
------- ------- -------
Sample1 13 14 27
Sample2 13 14 27
Sample3 14 13 27
Sample4 14 13 27
Sample5 14 13 27
Sample6 13 14 27
Sample7 14 13 27
Sample8 14 13 27
Sample9 13 14 27
Sample10 13 14 27 |
Please pre-test everything prior to your Live test.
Supportively yours,
Karen |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|