COMPLETE RANDOMIZED DESIGN (CRD)
Where the treatments are assigned completely at random so that each
treatment units has the same chance of receiving any one treatment.
This is suitable for only the experiment material is homogenous.(ex:
laboratory experiments, green house studies etc.)
Not suitable for heterogenous study. (ex: field experiments)
Advantages :
Simple and easy.
Provides maximum number of degrees of freedom.
Disadvantages :
Less accurate than other designs.
Heterogeneity of experimental material will be increased.
Increased experimental error and reduce precision.
RANDOMIZED BLOCK DESIGN (RBD) OR
RANDOMIZED COMPLETE BLOCK DESIGN (RCBD)
Most widely used experimental designs in agricultural research.
Experimental materials is grouped in to homogenous sub groups. The
sub groups is commonly termed as block. Since each block will consists
the entire set of treatments, a block is equivalent to a replication.(ex:
field experiments)
Advantages :
More efficient than CRD.
The statistical analysis is simple and easy.
Reduced experimental error and increase precision.
Disadvantages :
Not suitable for large number of treatments.
It is a versatile design.
LATIN SQUARE DESIGN (LSD)
A Latin square experiment is assumed to be a three factor experiment.
The factors are rows, columns and treatments.
It is assumed that there is no interaction between rows, columns and
treatments.(rows = columns = treatments)
It is differ from randomized block designs in the experimental units are
grouped in blocks in two different ways, i.e. by rows and columns.
Advantages :
LSD is more efficient than RBD and CRD.
The experimental error is small compared than other deigns.
Analysis is simple even with missing plots.
Disadvantages:
Number of treatments is limited to the number of replicates which
seldom exceeds 12.
If have less than 5 treatments, the df for controlling random variation is
relatively large and the df for error is small.
LAYOUT OF CRD
Treatments = N P K
N K P N
N = 4 level
P K N K
P = 3 level
K = 5 level K N P K
ANOVA MODEL FOR CRD:-
Sources of df SS MS F
variation
Treatments t-1 TSS TMS = TSS/ t-1 TSS/ EMS
Error n-t ESS EMS= ESS/n-t
Total n-1 Total SS
)
LAYOUT OF RBD
Treatments = 4 Block 1 Block 2 Block 3
Replications = 3 B A C
A B D
C C B
D D A
ANOVA FOR RBD :-
Sources of df SS MS F
variation
Replication r–1 RSS RMS RMS/ EMS
Treatment t–1 TSS TMS TMS/EMS
Error (r-1)(t-1) ESS EMS
Total rt - 1 Total SS
LAYOUT OF LSD
A B C D E
B A E C D
C D A E B
D E B A C
E C D B A
ANOVA FOR LSD :-
Source of variation df SS MS F
Rows t-1 RSS RMS RMS/EMS
Columns t-1 CSS CMS CMS/EMS
Treatments t-1 TSS TMS TMS/EMS
Error (t-1)(t-2) ESS EMS
Total t2 - 1 Total SS
Criteria CRD RBD LSD
1. Condition a. The use of amount a. The use of fertility a. The use of fertility
of use: of experimental variation moves in one variation move in
material to be tested direction. two direction.
and nature of the
fertility variation of
the field.
2. Layout: b. The whole field is b. The field is divided b. The field is
divided directly into into homogenous (plots divided into equal
plots is equal size = number of number of rows and
(product of treatments). columns ( number of
replications = rows and columns =
treatments). number of
treatments).
3. c. Differ from c. Differ from c. Number of
Replications: treatment to treatment treatment to treatment replication = number
(4 to 5 replications are (4 to 5 replications are of treatments ( 5 to
sufficient). sufficient). 12 replications are
sufficient).
Criteria CRD RBD LSD
4. d. In this case d. Replication or d. Rows, columns and
Randomization: treatment wise. block-wise. treatments are
reshuffled with the
help of random
number.
5. Local control: e. Not adopted. e. Adopted by e. Adopted by
homogenous blocks. homogenous blocks.
[Link]: f. Easy and simple. f. Analysis is easy but f. Analysis is simple
it becomes complicated but it becomes
when missing plot complicated when
technique is applied. several plot yield are
missing.
7. Components g. Total variation g. Total variation is g. Total variation is
of variation : is divided into two divided into three divided into four
components components (blocks, groups ( rows,
(treatment and treatments and error). columns, treatments
error). and error).