Randomized Complete Block Design (RCBD)
Download
Report
Transcript Randomized Complete Block Design (RCBD)
Randomized Complete Block
Design (RCBD)
Block--a nuisance factor included in an
experiment to account for variation among
eu’s
Presumably, eu’s are homogenous within
a block
Treatments are randomly assigned to eu’s
within each block
RCBD
The model and hypotheses
Yij i j ij ij ,
2
ij ij iid N (0, )
H o : i 0
RCBD
Blocks can be modeled as both fixed and
random effects (Soil example)
– Block: Soil type (fixed or random?)
– Treatment: Nitrogen x Watering Regimen
– Response: IR/R reflection
RCBD
There is some controversy as to whether
fixed block effects should be tested
– F test is considered at best approximate
Additivity of the block and factor effects
– Error includes lack-of-fit
– Practical considerations
Both block and factor could have a
factorial structure
Missing values in RCBD’s
Missing values result in a loss of
orthogonality (generally)
A single missing value can be imputed
– The missing cell (yi*j*=x) can be estimated by
profile least squares
x
ay'i*. by'. j* y'..
a 1b 1
Imputation
The error df should be reduced by one,
since x was estimated
SAS can compute the F statistic, but the pvalue will have to be computed separately
The method is efficient only when a couple
cells are missing
Imputation
The usual Type III analysis is available,
but be careful of interpretation
Little and Rubin use MLE and simulationbased approaches
PROC MI in SAS v9 implements Little and
Rubin approaches
Power analysis
Power calculations change little
– b replaces n in formulas
For Ho : L 0, use
bL2
2 ci2
For H o : 0, use
b
– The error df is (a-1)(b-1)
2
2
i