Scribd
Upload
66 views13 pages

Incomplete Block Design Overview

The document discusses incomplete block design, which is a statistical method where not all treatments can be compared together due to limited plots in each block. It explains the concept of intra and inter block analysis, emphasizing that treatments can still be compared indirectly through common treatments in blocks. Additionally, it introduces balanced lattices and alpha-lattice designs, highlighting their flexibility and application in comparing multiple treatments.

Uploaded by

Tesfaye Abera
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
66 views13 pages

Incomplete Block Design Overview

The document discusses incomplete block design, which is a statistical method where not all treatments can be compared together due to limited plots in each block. It explains the concept of intra and inter block analysis, emphasizing that treatments can still be compared indirectly through common treatments in blocks. Additionally, it introduces balanced lattices and alpha-lattice designs, highlighting their flexibility and application in comparing multiple treatments.

Uploaded by

Tesfaye Abera
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd

Incomplete Block Design

An Introduction
1. General Concept

 Suppose we have 9 treatments but only 3


blocks and 3 plots in a block.

Block I Block II Block III


T1 T4 T7

T2 T5 T8

T3 T6 T9
General (contin.)

 Incomplete block design may be defined as a


design where all treatments can’t appear
together in a block, since the number of plots
per block is less than number of treatments
General (contin.)

 Consider two design structures:


 (I)

1 a b b c c a b
c
2 a a b c c a b
b
3 a a b b c a b
c
General (contin.)

 The design is symmetric, implying that all


pair-wise mean difference, A-B, A-C are all
equally precise.

 However, this design is not incomplete block


but repeated appearance of a treatment in
the same block.
General (contin.)
 This design is not symmetric, but in total
balance, because variance of mean
differences are equal, i.e., var(A-B) = var(b-c)

(II)
1 a a b c
2 a a b c
3 b c
General (contin.)
 In IB treatments are not compared free of block
effects. This takes us to new concepts and ideas.
Intra block analysis
Inter block analysis

The second option gives the IB designs the power to


compare two treatments which have not occurred
together in a block but which occurred with some of
the treatments in block which is common to both.
General (contin.)

 Example Suppose that we have five


treatments to be compared in two blocks of 4
plots each
T1 T2
T2 T3

T3 T4

T4 T5
General (contin.)

 Although T1 have not occurred with T5, they


are compared through T2, T3, and T4 which
occurred with both T1 & T5 in one of the two
blocks

 Effect of T1-T5 is confounded with block


effect. Therefore, we can only compare T1
and T5 by first estimating block effect.
Balanced Lattices

 Every
pair of treatments occurs once in the
same incomplete block

 Consequently,all pairs of treatments are


compared with the same degree of freedom

 See SAS Example ([Link])


Alpha-Lattice

Alpha Lattice are:


 Generalization of the Lattice designs
 Advantage of - design is that it is flexible. It exists
for all combination of variety, rep and plots per block.
 Can have incomplete block of different size
 Checks can also be used
 There are different alpha lattices
 (0,1), (0,1,2). . . . .
Example

 Table 5.3 Alpha design for comparing 28 treatments in blocks of 4 plots



 Replicate 1
 Block I 1 8 15 22
 Block II 2 9 16 23
 Block III 3 10 17 24
 Block IV 4 11 18 25
 Block V 5 12 19 26
 Block VI 6 13 20 27
 Block VII 7 14 21 28
 Replicate 2
 Block VIII 1 9 17 26
 Block IX 2 10 18 27
 Block X 3 11 19 28
 Block XI 4 12 20 22
 Block XII 5 13 21 23
 Block XIII 6 14 15 24
 Block XIV 7 8 16 25
Example (cont.)
 Replicate 3
 Block XV 1 11 21 27
 Block XVI 2 12 15 28
 Block XVII 3 13 16 22
 Block XVIII 4 14 17 23
 Block XIX 5 8 18 24
 Block XXI 6 9 19 25
 Block XXI 7 10 20 26

 See Analysis using SAS ([Link])

You might also like