Hi There. I am doing sensory research with 8 samples of rice at one measurement moment. To avoid fatigue/sensory overload, we want our test persons only to try four of those, randomly allocated ofcours. I was thinking of a balanced incomplete block design. However, I am not quite sure how to construct such.
Does anyone have some tips?
Thank you so much.
Very few researchers design their own BIBD's. That's a very specialised area of Combinitorial Mathematics. Most people select an appropriate design from a published catalogue. Or use a program to make a design for them which is not necessarily the most efficient.
I usually go to the website called DesignTheory.org, linked below, to find an appropriate t-design. There, you might find the following BIBD, with 8 items, presented in 14 blocks of 4, each item presented 7 times and each item compared with each other item 3 times.
Block A B C D
1 0 1 2 3
2 0 1 2 4
3 0 1 5 6
4 0 2 5 7
5 0 3 4 5
6 0 3 6 7
7 0 4 6 7
8 1 2 6 7
9 1 3 4 6
10 1 3 5 7
11 1 4 5 7
12 2 3 4 7
13 2 3 5 6
14 2 4 5 6
So you would randomly allocate respondents to one of the 14 blocks, or, better still, assuming that respondents come in to your lab at a more or less random order, you would systematically allocate respondents to the next one of the 14 blocks to ensure that all blocks were about evenly included.
http://www.maths.qmul.ac.uk/~leonard/designtheory.org/database/t-designs/
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS