Experimental designs:

Basic terminology:

Design:- Whenever an agriculture experiment is done by using certain

scientific (statistical) procedure then it is called design.

OR Experimental design are various types of plot arrangement which are

used to test a set of treatments to draw a valid conclusions about a
particular problems.

Treatment:- The objective under comparison is called treatment. The

variable whose effect we want to measure is known as treatment.

Experimental material:- The material which is used in experiment is

known as experimental material . For example, a field, soil seeds etc.

Experimental units:- Small plot of the block to which the treatment can
apply are called experimental unit.

Experiment:- Experiment is a scientifically planned method. The

experiment is conducted to draw a valid conclusion about a particular
problem. The conclusion is based on statistical observation.

Experimental error:- The variation due to environmental factor or due to

uncontrolled factor is called experimental error.

Or the variations among the experimental units having the same treatment
and are treated alike is known as experimental error.

Uniformity trail:- A uniformity trail consists in dividing the whole field

into so many small units of equal size. Uniformity trail are helps as know
the nature of soil fertility in agricultural/ experimental field.

Sampling unit :- The object that is measured in an experiment is called the

sampling unit. This may be different from the experimental unit.
Why need experimental design in breeding?
Definition of Exp.design :- The choice of treatment, the method assigning
treatments to experimental units and arrangement of experimental units in
various pattern to suit the requirements of particular problems are
commonly known as design of experiment. Or Experimental design is the
process of planning and study to meet a specific objectives.
Objectives:- 1. To increase precision of experiment 2. To reduce
experimental error 3. In screening off various treatments 4. In partitioning
of variation into different components 5. Used in proper interpretation of
scientific results and drawing valid conclusions 6. In reducing the soil
heterogeneity 7. In assessment of variance and covariance 8. Shows the
direction of better results 9. Includes the plan for analysis and reporting of
the results
Principles of Experimental Design: There are three basic principles : •
Replication:- Repetition of the treatment under investigation Or To
provide an estimate of experimental error.
Randomization:- The allocation of the treatment to the different
experimental units by a random process is known as randomization. •
Local control:- The principal of making use of greater homogeneity in
groups of experimental units for reducing experimental error.

TYPES OF EXPERIMENTAL DESIGNS

1. Completely Randomized Design
2. Randomized Complete Block Design
3. Latin Square Designs
4. Split Plot Design
5. Augmented Design
6. Lattice Designs (Alpha Lattice Design)

1. Completely Randomized Design (CRD): 

Developed by Proff. R. A. Fisher.
The design which is used when the experimental material is limited and
experimental units homogenous.
It is single factor design.
The principle of local control is not allowed/ adopted in this design.
This design is specially used for pot culture experiments.
No. of plots = No. of replications + No. of treatments.
The design based on two principle of Experimental designs :-1. Replication;
2. Randomization.
Characteristics of CRD CRD is applicable only when the experimental
material is homogenous (Example: Homogenous soil condition in the
field). Usually in the field, the soil will not be Homogenous. Thus, CRD
is not a preferable method in field experiments.  CRD is generally
applicable to the lab experimental conditions. In labs, the environmental
conditions can be easily controlled. The concept of ‘Local-control’ is not
used in CRD.
Advantages of CRD CRD is easy to understand and calculate the
variance. Any no. of replication are used. The number of replications
can vary from treatment to treatment.  CRD has high flexibility and thus
any number of treatments can be used.  Simple statistical analysis is
required in the analysis of CRD. CRD provides maximum number of
degree of freedom.
Disadvantages of CRD CRD can be applied only to homogenous
experiments. The principle of ‘Local-control’ is not used in CRD.

Randomized Complete Block Design (RCBD):- 

RCBD is used when the experimental material is not homogenous and
fertility gradient is moving one direction. 
Developed by Proff. R. A. Fisher in 1924. 
The no. of equal plot in each block is equal to the no. of treatment. 
The design is based on of three principles of experimental designs
(replication ,randomization and local control). 
In case of field experiment the experimental material is divided into a no.
of equal blocks. 
RBD is the most commonly used experimental design in agriculture.

BLOCK 1 BLOCK 2 BLOCK 3 Here T1, T2, T3, T4,and T5 are treatment
Advantages of RBD 
RBD is more efficient and accurate when compared to CRD. 
When material is heterogeneous and no. of treatment more then 20.
Chance of error in RBD is comparatively less. Statistical analysis is
relatively simple and easy. 
Statistical analysis simple when one value is missing. 
Errors of any treatment can be isolated.
Disadvantages of RCBD 
RBD is not advised for very large number of treatments. 
If the heterogeneity of the plot is very high, RBD cannot be applied. When
the number of treatments is very large then the size of each block will be
increased so that there may be heterogeneous blocks within.

Latin Square Design (LSD): 

The experimental design which simultaneously controls the fertility
variation in two directions is called Latin square design (LSD). 
It is two factor design( row may present to level of one factor and column
present the another factor. 
The experimental material is divided into rows and columns. 
Each having the same number of experimental units which is equal to the
number of treatments. 
The treatments are allocated to the rows and the columns such that each
treatment occurs once and only once in the each row and in the each
column. In other words, Latin square designs are adopted for eliminating
the variation of two factors which are generally called rows and columns.

Here , No. of rows = No. of column =No. of treatment No. of replication =

no. of treatment Structure of Analysis of Variance No. of Replication = No.
of Treatment No. of Rows = No. of Columns = No. of Treatment Note :-
Design used when treatment are 5 to 8 or at the most 12.

A Latin square is called self conjugate if its arrangement in rows and

columns are the same. Characteristics of LSD 
LSD is a design where the experimental material is divided into ‘m’ rows'
columns and ‘m’ treatments – assigned by randomization method to rows
and columns. 
The randomization in such a way that each treatment occurs only once in
each row and in each column. Application of LSD 
Field experiment Animal experiment Experiments on a long strip of
land
Advantages of LSD Statistical analysis is relatively simple (complicated
than CRD and RBD). 
Statistical analysis is simple if one value is missing. 
Most efficient design when compared to CRD and RBD. 
Standard error less than 1%. Disadvantages of LSD 
LSD is not suitable for agricultural experiments. 
Statistical analysis is complicated when two or more values are missing.
Difficult when treatments are more than ten. 
No. of treatment 5 to 12 no more treatment or less treatment.

Comparison between CRD, RCBD and LSD

CRD Fully homogeneous Any no. of treatment studied Vary from
treatment to treatment RBD Heterogeneous in one direction More then 20
Same for the all treatment but not necessarily LSD Heterogeneous in two
direction 5 to 12 No. of replication = no. of treatment In case Experimental
Material No. of treatments No. of replications

Randomized Complete Block Design: This is one of the most commonly

used designs in agricultural research, particularly in plant breeding
programmes. Its primary distinguishing feature is the presence of blocks
(replications) of equal size, each of which contains all the treatments

Here ; Four treatment and five blocks. An experimental units in each cell.

Charecterstics of RCBD The RCBD is the standard design for agricultural

experiments where similar experimental units are grouped into blocks or
replicates. It is used to control variation in an experiment by accounting
for spatial effects in field or greenhouse. e.g. variation in fertility or
drainage differences in a field Advantages of RCBD Generally more
precise than the CRD. No restriction on the no. of treatment or replicates.
Some treatment may be replicated more items than others Missing plots
are easily estimated Whole treatment or entire replicates may be deleted
from the analysis.  if experimental error is heterogeneous ,valid
comparison can still be made.
Disadvantages of RCBD Error DF is smaller than that for the CRD
(problem with a small no. of treatment). If there is a larger variation
between experimental units within a blocks , a large error term may result (
this may be due too many treatment) If there are missing data , a RCBD
experiment may be less efficient than a CRD.

Split Plot Design (SPD): Experimental plots are split or divided into
main plots, sub plots and ultimate-plots.  In this design several factors are
studied simultaneously with different levels of precision. The factors are
such that some of them require larger plots like irrigation, depth of
ploughing and sowing dates, and others require smaller plots.

Aplication of SPD When all the factors are not of equal importance.
When some of the factors cannot be tasted on small amount of material
and require large one for the purpose Advantages of SPD Useful when all
the factors are not equal importance, i.e., some of them required larger
plots and others require smaller plots. When some of the factor have
small amount of material, they can used as sub plots or ultimate plots in
this design. Disadvantages of SPD The layout and analysis more
complicated as compared to that RBD and LSD. To provides lesser degree
of freedom for the estimation of error variance than RBD.

Augmented Designs:  Developed by Federer (1956). This is an

experimental design which is used to test a large number of germplasm
lines in a limited area and error estimate with help of checks varieties. to
evaluate a large number of germplasm lines This design is commonly
used and other Designs are not appropriate due to large number of entries.
In augmented designs the goal is to compare existing (control) treatments
with new treatments that have an experimental constraint of "limited
replication“.

Experimental lines replicated once.  Checks occur in each block.  Checks

used to estimate block effects.  Checks provide error term.  Need a
mechanism to adjust for field variation  Difficult to maintain
homogeneous blocks when comparing so many genotypes Advantage of
Augmented design To evaluate more genotypes,Test in more
environments . Fewer check plots are required than for designs with
systematic repetition of a single check Provide an estimate of standard
error that can be used for comparisons genotypes, Between new genotypes
and check varieties Observations on new genotypes can be adjusted for
field heterogeneity. Flexible – blocks can be of unequal size

Disadvantage of Augmented Design Considerable resources are spent

on production and processing of control plots Relatively few degrees of
freedom for experimental error, which reduces the power to detect
differences among treatments Unreplicated experiments are inherently
imprecise, no matter how sophisticated the design

Lattice Designs It is an incomplete-block design. Developed by Yates.

Resolved if the blocks are grouped into larger blocks and the large blocks
form a complete-block design. To provide for the elimination of soil
differences by the use of two different groupings of the same experimental
plots.

Alpha lattices Alpha-lattice designs are used when the large number of
genotypes (treatments) and small area. There are no checks varieties for
estimation error. Error checks within the same varieties. Alpha-lattices
are used to reduce the effect of within-complete-block variation. They can
provide risk- and cost-free increases in precision in trails. They can also
provide repeatability, particularly in trials. The design permits removal of
incomplete-block effects from the plot residuals. Maximizes the use of
comparisons between genotypes in the same incomplete-block.

Charecterstics of Alpha lattice a) Resolvable incomplete block design b)

Work of any combination t=s*k c) Useful for field trial d) No. of treatments
is fixed e) more flexibility in choice of s and Here ; t = number of treatments
k = number of units per block (block size) b = total number of blocks in the
experiment r = number of replicates of each treatment s = number of blocks
in each complete replication

Advantages of alpha lattice design This design allows the adjustment of

treatment means fir block effects. This in turn brings benefit from the
small incomplete blocks which help varietal comparisons under mere
homogeneous condition. This design also provides effective control
within replicate variability.