EXPERIMENTAL

DESIGNS

FRANCISCO S. ANTONIO
Associate Professor
CAGAYAN STATE UNIVERSITY
 Experimental Designs

set of rules, plans and course of
action taken in the conduct of
experiment
 Attributes of
ExperimentalDesigns

 ensure cost effective collection of appropriate data

 provide appropriate and valid analysis of data

 provide reliable conclusion leading to reliable
inference
 Common Terms Used in
Experimental Design

 *Treatments – procedures or conditions whose effects are
to be measured or compared
 *Experimental unit- a group or experimental materials or
individuals to which a single treatment is applied once
 *Response variable – a characteristic used to measure the
treatment effects
 *Sampling unit – a portion of the experimental on which
the response variable is observed and measured
 *Experimental error- measure of variations among
experimental units treated alike
 Common Terms Used
in Experimental
 Design
 Sources of experimental errors
 a. inherent variability of the experimental materials
 b. errors in experimentation
 c. errors in observations and measurements
 d. combined effects of all extraneous uncontrollable
factors
 Common Terms Used
in Experimental
 Design
 *Sampling error –measure of variations among
sampling units within experimental unit
 *Precision –repeatability of measurements which is
measured by variance
 Ways of increasing precision
 a. increase the number of samples or replications
 b. skillful grouping of experimental materials
 c. proper selection of treatments
 Common Terms Used
in Experimental
 Design
 *Accuracy – unbiasness or the closeness of the
average values of the measurements to the true value

 *Layout – final arrangement of treatments over the

whole experimental areas
Principles of Experimental
Design

 1. Replication – repetition of treatments application on a
number of eu’s
 Functions of replication
 a. to provide an estimate of the experimental error
 b. to increase the precision of estimates
 c. to increase the scope of the experiment

 Factors affecting the number of replications

 a. degree of precision required
 b. uniformity of experimental units
 c. experimental design
 d. time allotment to the experiment
 e. cost and availability of resources
 Principles of Experimental
Design

 2. Randomization – allocation of treatments to the

experimental units by means of a chance device such that

every treatment has equal chance of being assigned to any
experimental unit
 Functions of randomization
 a. provide a random sample of observation
 b. to satisfy the assumption of independence of observations
 c. to eliminate systematic bias in assigning the treatments
 Principles of
Experimental
 Design
 3. Local control or error control - process or technique to minimize experimental
error
 Common techniques for local control

 a. proper use of experimental design

 b. proper choice and shape of experimental units

 c. use of concomitant variable

 d. refinement of experimental technique

 - uniform application of treatments and management

 - more control over external influences

 -devising unbiased measurement echniques

 -preventing gross errors

 EXPERIMENTAL DESIGNS

 1. Completely Randomized Design(CRD)
 2. Randomized Complete Block Design (RCBD)
 3. Latin Square Design(LSD)
 4. Factorial Experiments
 5. Split Plot Designs
 CRD-One Way Classification
Design

 a design where allocation of treatments is done over
the experimental units without restrictions
 When to Use?
 1. eu’s are homogenous

 2. effective local control (greenhouse,

laboratory experiments)
CRD-One Way Classification
Design
 Randomization and Lay-out 
 a. Label the eu’s from 1 to n (e.g. number of
treatments= 3, n=9, number of replications for
T1=2, T2=3, T3=4)
 b. Obtain a sequence of n=9 random numbers
 Random no: .831 .118 .881 .341 .520 .361 .462 .002 .999
 Rank (eu no.) 7 2 8 4 6 3 5 1 9
T1 T2
T3

 c. Rank the numbers in increasing order.

CRD-One Way Classification
Design

 d. Using the sequence of ranks as randomization of the
eu’s, assign the treatments
 Random numbers:

 Layout of the randomization procedure

1
T3 2
T1 3
T3
4
T2 5
T3 6
T2
7
T1 8
T2 9
T3
Data presentation (One
Way Classification)

Treatment Observations Reps Total Mean
T1              
T2              
T3              
               

Linear Model:
Yij = μ + ti + eij
 Testing Significance of
Treatments via ANOVA F test

 1. State the statistical hypothesis

 Ho: μ1 = μ2= μn : all treatments means are equal

 Ha: μ1 =/= μn : at least two treatment means are

different

 2. Formulate the test statistic and get the critical

value at the level of significance α
 
 3. State the decision rule
 Reject Ho if Fc> Ftab
 4. Construct the ANOVA table
Sources Degree of Sum of Mean Fc Ftab α=.05 Ftab α=.05
of freedom Squares Square
variation
Treatment t-1 TrSS MSTr MSTr/MS    
E
Error n-t ESS MSE      
Total n-1 TSS        

 5. State the decision and make the conclusion

 Assumptions Underlying the
Analysis of Variance (ANOVA)

 1. homogeneity of variance of the experimental
errors
 2. normal distribution of the experimental errors
 3. independence of the experimental errors
 4. additivity of the treatment effects and the
environmental effects
 Remedies for Violations of
the Assumptions of ANOVA

 a. logarithmic transformation - appropriate when standard
deviation is proportional to its mean with greatly skewed
distribution and data with multiplicative effect
 - appropriate for count data

 b. square root transformation – appropriate for data where variance

is proportional to its mean as in data with small whole numbers like
counts of rare events and frequency count data

 c. arcsine transformation – appropriate for proportion or percentage

data
 Multiple Comparisons
Among Means

 1. Pairwise Mean Comparisons

 2. Group Comparisons
 3. Trend Comparisons
 Statistical Hypothesis for
Pairwise Comparision

 Ho: u1-un=0 , the treatment means for u1 and un are
not different
 Ha: u1- un =/= 0, the treatment means u1 and un
are different
 Commonly Used Tests for
Pairwise Comparisons

 1. Least Significant Difference (LSD) test

 2. Duncan’s Multiple Range test (DMRT)

 3. Student-Newman-Keul (SNK) test
 4. Honest Significant Different (HSD) test

 5. Scheffe’s (S) test

 Features of LSD test

 1. applicable when F-test in ANOVA is significant

 2. it may not be used for more than four (4) treatments to

test
 3. can be used for preplanned comparisons irregardless of the
number of treatments
 4. experiment has equal or unequal replications
 5. it is the least conservative among pairwise comparisons test
 6. it uses the Student t table
 Features of DMRT

 1. more sensitive than the F-test in the ANOVA; can be used
even if F-test is not significant
 2. used to compare differences between all possible pairs of
means
 3. it is sequential test requiring series of values of the test statistic
each corresponding to a specific set of pair comparisons
 4. used when the experiment has equal replications

 5. uses the new studentized range table

 Features of SNK Test

 1. used only when ANOVA F test is significant

 2. requires equal replications

 3. it accounts for the number of treatments in the
experiment
 4. it is a sequential test

 5. it uses the studentized ranged table

 Features of Scheffe’s S
Test

 1. it is based on an F-tabular values

 2. can be used to test any mean comparisons

 3. can be used for treatment with equal and unequal
reps
 4. the most conservative of all the pairwise tests
GROUP COMPARISON

 1. Linear comparison

 2. Orthogonal contrasts
 3. Set of orthogonal contrasts
TREND COMPARISON

 determine quantitative functional relationship
between the treatments and response variable by
trend comparisons which fit the orthogonal
polynomial trends such as linear, quadratic, cubic ,
quartic, and other higher degree of polynomials
 Randomized
Complete Block
Design (RCBD)

 Attributes of RCBD

 a. precision is increased by proper grouping or
blocking of the experimental units as a form of local
control
 b. eu’s are grouped into blocks such that differences
between units among different blocks are greater
than differences between the units within each block
 Attributes of RCBD

 c. blocking is done so that blocks cut across or ar
perpendicular to the direction of the eu’s gradient
 d. with differences among the blocks, variability is
removed from the experimental error improving the
precision of the experiment
 Randomization and
Layout

 Conditions : no. of treatments = 4

no. of replication = 3

total number of eu’s = 12

 a. Group the eu’s into r = 3 blocks

 b. divide each block into t= 4 eu’s

 c. for each block, allocate the treatments into the eu’s at random and independently of

other blocks as follows;

 1. label the eu’s consecultively from 1 to t=4

 2. obtain a sequence of t=4 numbers using draw lots

 3. using the sequence of draws as the treatment numbers and the drawn numbers as

the eu numbers assign the treatments to the respective eu’s.

For block 1

draw sequence : 1 2 3 4 = treatment no.

draw numbers: 2 4 1 3 = eu number

For block 2 draw sequence : 1 2 3 4 = treatment no.

draw numbers: 4 1 3 2 = eu number

For block 3 draw sequence : 1 2 3 4 = treatment no.

draw numbers: 3 1 2 4 = eu number

Corresponding Layout

1
T3   1
T2   1
T2
2
T1   2
T4   2
T3
3
T4   3
T3   3
T1
4
T2   4
T1   4
T4

Block 1 Block 2 Block 3

 Data Presentation

Treatment Block Trmt Mean
Total
1 2 3    
T1          
T2          
T3          
T4          
Total          

RCBD Linear Model

 
Yij = μ + ti + bj + eij
 
 RCBD ANOVA Table
Sources

Degree of Sum of Mean Fc Ftab Ftab
of freedom Squares Square α=.05 α=.05
variation

Block r-1 RSS MSR MSR/MS    

E
Treatmen t-1 TrSS MSTr MSTr/MS    
t E
Error (r-1) (t-1) ESS MSE      
tr-1 TSS        
Total
LATIN SQUARE
DESIGN (LSD)

 Features of LSD

 a. two way classification design that can handle
simultaneously two known sources of variation among
experimental units
 b. two grouping or blocking are noted as column and row

 c. rows and columns refers to criteria of classification; they

maybe kinds of treatment or factors simultaneously
occurring assuming no interaction among themselves
 d. applicable when the number of treatments is 4-8
Common Examples of Studies
Done with LSD

 a. digestibility of feeds using fistulated animals
 b. field trials in experimental areas with two fertility
gradients
 c. insecticide field trials where the insect migration
has a predictable direction that is perpendicular to the
dominant fertility gradient of the experimental field
 Common Examples of
Studies Done with LSD

 d. greenhouse trials in which experimental pots are arranged in straight
line perpendicular to the glass or screen walls such that the difference
among rows of pots and distance from glass wall(or screen wall) are
expected two major sources of varaiability among the experimental
pots

 e. lab trials with replication over time such that differences among
experimental units conducted at the same time and among those
conducted over time constitute the two known sources of variability
 Features of Blocking,
Randomization and Layout in
LSD

 a. number of treatments = number of columns =
number of rows
 b. treatments are then arranged in blocks in two ways;
by rows and columns such that each treatment occurs
only once in each column and row
 c. each row is a complete block, each column is
likewise a complete block
LSD Randomization
and Layout

ILLUSTRATION of LSD

 An experiment is to be conducted to
measure lead content of 4 seaweeds . It is
suspected that the source location (S1, S2, S3,
S4) and the chemists (C1, C2, C3, C4) that perform
the chemical analysis contribute to the variability
of the results. It is assumed that the source
location, the chemists and kinds of seaweeds have
no interaction.
 Treatments= T1, T2, T3, T4
 Row classification =S1, S2, S3, S4
 Column Classification=C1, C2, C3, C4
Steps in Randomization

1. Randomize the assignment of the source location
to the rows on the plan
  Column Column 2 Column 3 Column 4
1
Row 1 A B C D
Row 2 B C D A
Row 3 C D A B
Row 4 D A B C
 Steps in Randomizaton

2. Randomize the assignment of the source
location to the rows on the plan
Draw 1 2 3 4= Course location
sequence:
Drawn 3 1 4 2= Plan rows
number :
  Colum Column Column Column 4
n1 2 3
S3 A B C D
S1 B C D A
S4 C D A B
S2 D A B C
Steps in Randomization
 of the chemist to
3. Randomize the assignment
the columns in the plan
Draw 1 2 3 4= chemists
sequence:
Drawn 3 1 4 2= Plan columns
number :

  C3 C2 C4 C1
S3 A B C D
S1 B C D A
S4 C D A B
S2 D A B C
Steps in Randomization

4. Randomize the assignment of treatments to the
treatments on the plan
Draw 1 2 3 4 seaweeds
sequence:
Drawn B(4) 1(D) 4(C) 2(A) Plan columns
number :

  C3 C2 C4 C1
S3 T4 T1 T3 T2
S1 T1 T3 T2 T4
S4 T3 T2 T4 T1
S2 T2 T4 T1 T3
 Final Layout
 of LSD
5. Rearrange the rows and columns in sequence
(LAYOUT)

  C1 C2 C3 C4
S1 T2 T1 T4 T3
S2 T4 T3 T1 T2
S3 T1 T2 T3 T4
S4 T3 T4 T2 T1
 LSD Data Presentation
Row
Class’n
 Column Classification Row Total

  1 2 3 4  
1          
2          
3          
4          
Col Total          
           
Treatment T1 T2 T3 T4 Total
Total          
 ANOVA TABLE
Source of Degree of  Sum of Mean of Fc Fα=.05 Fα=.01
Variation freedom Squares Squares
Rows r-1 RSS MSR MSR/M    
SE
Columns c-1 CSS MSC MSC/M    
SE
Treatments t-1 TrSS MSTr MSTr/    
MSE
Error (r-1)(c-2) ESS MSE      
TOTAL t2-1 TSS        
 FACTORIAL
EXPERIMENTS

 FACTORIAL
EXPERIMENTS

an experiment where the treatments consist of all
treatment combinations of two or more factors
 Common Terms Used in Factorial
Experiments

*Factor - a kind of treatment whose effect is to
be measured in an experiment
*Levels – the different amounts or categories of a
factor
*Treatment Combinations – treatments formed by
combining the different levels of two or more
factors
 KINDS OF TREATMENT EFFECTS

Simple effect –difference between two levels of
the factor at a given level of another factor

Main effect –average of the simple effects

TWO FACTOR FACTORIAL IN RCBD


 an experiment involving two factors where
the treatment combinations are randomized over
the eu’s using RCBD
 ILLUSTRATION

An experiment to assess simultaneously effect of
Factor A three plantind densities (A1, A2, A3)
and Factor B the kind of fertilizer with 4 levels
(B1, B2, B3, B4) using RCBD with three (3)
replications
 Treatment Combinations= 3 x 4 = 12 treatment
combinations
 Illustration of
Factorial 
Experiment
Treatment Combinations= 3 x 4 = 12 treatment
combinations
T1-A1B1 T1-A2B1 T1-A1B1
T1-A1B2 T1-A2B2 T1-A1B2
T1-A1B3 T1-A2B3 T1-A1B3
T1-A1B4 T1-A3B4 T1-A1B4
 RANDOMIZATION AND LAYOUT
OF 2-FACTOR in RCBD
For block 1
Order of 1 2 3 4

5 6 7 8 9 10 11 12
Draws:
Trt 5 9 2 11 7 1 4 6 10 12 3 8
combinat’n:

For block 2
Order of 1 2 3 4 5 6 7 8 9 10 11 12
Draws:
Trt 5 9 2 11 7 1 4 6 10 12 3 8
combinat’n:

 For block
Order of 1 32 3 4 5 6 7 8 9 10 11 12
Draws:
Trt 8 5 10 9 4 2 3 12 3 11 1 7
combinat’n:
Layout
1 2 3
  1 2 3
  1 2 3

 

A2B1 A3B A1B   A1B3 A1B A2B3   A 2 B A 1B A3 B2
  4
1 2 1 4
4 5 6   4 5 6   4 5 6

A3B3 A2B3 A1B1   A3B1 A2B1 A3B2   AB AB AB

3 1 1 4 1 2
7 8 9
A3B 7 8 9
A1B 7 8 9
A1 B
A1B4 A2B2 2
A1B2 A3B3 4
A2B2 A3B4 3
10 11 12
A2 10 11 12
A2 10 11 12
A2
A3B4 A1B3 B4 A2B4 A3B4 B2 A3B3 A1B1 B3
Block 1 Block 2 Block 3
 DATA PRESENTATION
Factor A Factor B Block Total Mean
1 2 3

A1 B1
B2
 
 
 
    
   
 
 
 
B3          
B4          
           
A2 B1          
B2          
B3          
B4          
             

A3 B1          
B2          
B3          
B4          
           
 ANOVA F-TEST FOR A 2-FACTORAL
RCBD

 Construct the two- way table of total as follows
Factor Factor B A
A totals
  1 2 3 4  
1          
2          
3          
4          
B totals          
Construct the ANOVA Table 
Source of Degree of Sum of Mean of Fc Fα=.05 Fα=.01
Variation freedom Squares Squares
Block r-1 RSS MSR MSR/MSE P<.05 P<.01
Treatment ab-1 TrSS MSTr MSTr/MSE    
A a-1 ASS MSA MSA/MSE    
B b-1 BSS MSB MSB/MSE    
A xB (a-1) (b-1) ABSS MSAB MSAB/MS    
E
Error (r-1)(ab-1) ESS MSE      
TOTAL Abr-1 TSS        
 NOTES ON INTERPRETATION
OF FACTORIAL EXPERIMENTS

In general, tests on main effects have
straightforward interpretation only when the
interactions are not significant
The presence or absence of main effect tellls
nothing about the presence or absence of
interactions
With a significant interaction, the factors do not
act indepenently of each other; th simple effect
depends upon the level of the other factor of the
interaction term
 NOTES ON INTERPRETATION
OF FACTORIAL EXPERIMENTS

If the interaction is not significant, the factors
act independently of each other, the simple
effects of a factor are the same for all levels of
the other factor and so the simple effects are
equal to the corresponding main effect
Non-independent factors indicate the complexity
of the situation which have been missed had a
single-factor approach to experiment been used
for each factor
  
CRD WITH SUBSAMPLING


 -
 CRD WITH SUBSAMPLING

 more than 1 observation on each experimental unit
is done in CRD with subsampling
 ILLUSTRATION OF crd WITH
SUBSAMPLING

 An experiment is conducted to determine the
sugar content of three varieties of sugarcane. At
maturity, four cnes of each variety are available
for refractometer brix test. From each cane,
three determinations are made: at the base,
middle and top portions.
 DATA PRESENTATION OF CRD
WITH SUBSAMPLING
Treatment Sampling
Unit

Experimental Unit Treatment Treatment
Total Mean
1 2 3 4    

1 1            
2            
3            
Total            
2 1            
2            
3            
Total            
3 1            
2            
3            
Total            
TOTAL              
 ANOVA TABLE
Source of
 Degree Sum of Mean of Fc Fα=.0 Fα=.0
Variation of Squares Squares 5 1
freedom
Treatment t-1 TrSS MSTr      
Experimenta t(r-1) ESS MSE      
l Error

Sampling tr(s-1) DSS MSD      

error

TOTAL Trs-1 TSS        

Split Plot Design

 Uses

 levels of factor A require larger eu’s than those
required of factor B
 greater precision is desired for comparisons among
the levels of factor B than that desired for factor B
 it is known that the larger variations can be expected
among the levels of factor a and those expected on
factors A
 Randomization and Layout
   
Block Block
 Block
1 2 3
F1 F2 F3 F3 F1 F2 F4 F1 F4

F2 F4 F1 F1 F4 F3 F3 F3 F2

F4 F3 F2 F2 F2 F1 F1 F4 F1

F3 F1 F4 F4 F3 F4 F2 F2 F3

T3 T1 T2 T1 T3 T2 T2 T1 T3
 Data Presentation
Type of Kind of Farm Mean
Irrigation
Fertilizer 
1 2 3

F1        

T1 F2
F3
 
 
 
 
 
 
 
 

F4        

         

F1        

T2 F2
F3
 
 
 
 
 
 
 
 

F4        

         

F1        

T3 F2
F3
 
 
 
 
 
 
 
 

F4        

         
 Analysis of Results
Test the significance of treatments via
ANOVA F-test
SV df SS  MS Fc
Mainplot        
Block r-1 RSS MSR MSB/MSEa
A a-1 ASS MSA MSA/MSE
a

Error (a) (a-1)(r-1) EaSS MSEa  

Subplot        
B b-1 BSS MSB MSB/MSEb
AxB (a-1)(b-1) ABSS MSAB MSAB/MS
Eb
Error(b) a(r-1)(b-1) EbSS MSEb  
Strip Plot Design

 Different sizes of
plots

 Row-strip plot for the first factor A (row factor)

 Column- strip plot for the second factor B (column

factor)

 Intersection plot for the interaction between the

two factors
 Row and column strips should always be
perpendicular to each other
 Illustration
  Block 1  

Block 2   Block 3
R1       R1       R1      
R2       R2       R2      
R3       R3       R3      
R4       R4       R4      
  C1 C2 C3   C1 C2 C3   C1 C2 C3
 Table Presentation
Factor  Block
A B 1 2 3 4
1 1        
  2        
3        
         
2 1        
  2        
3        
         
3 1        
2        
3        
           
Summarize the means in a two-way table as
follows

Factor Factor B Factor
A B1 B2 B3 A
: means

A1        
A2        
A3        
Factor        
B
means
 Analysis of Variance
(ANOVA)
SV Df

SS MS F
Block r-1 RSS MSR MSR/MSEa
Horizontal a-1 ASS MSA MSA/MSEa
Factor (A)
Error(a) (r-1)(a-1) EaSS MSEa  
Vertical b-1 BSS MSB MSB/MSEb
factor (B)
Error (b) (r-1)(b-1) EbSS MSEb  
AXB (a-1)(b-1) ABSS MSAB MSAB/MSE
c
ErroR© (r-1) (a-1) EcSS MSEc  
(b-1)
Total Rab-1 TSS    
Strip-Split-Plot
Design

 Randomization and Layout
  B1 B2 B3
A2
 
A3
C2
C1
C1
 C1
C2
C1
C1
C2
C2
  C2 C2 C1
A1 C2 C2 C1
  C1 C1 C2

  B2 B1 B3
A3 C1 C2 C2
  C2 C1 C1
A1 C2 C1 C1
  C1 Block1
C2 C2
A2 C2 C1 C2
  C1 C2 C1
Randomization and Layout

  B3 B1 B2
A1 C1 C2 C1
  C2 C1 C2
A3 C1 C2 C2
  C2 C1 C1
A2 C2 C1 C1
  C1 C2 C2

  B1 B3 B2
A2 C1 C1 C2
  C2 C2 C1
A1 C2 C2 C1
  C1 C1 C2
Block3
A3 C1 C2 C1
  C2 C1 C2
Block4
 (ANOVA)
SV df SS MS F
Block
Horizontal
r-1
a-1 RSS
ASS
MSR
MSA
MSR/MSEa
MSA/MSEa
Factor (A)
Error(a) (r-1)(a-1) EaSS MSEa  
Vertical factor b-1 BSS MSB MSB/MSEb
(B)
Error (b) (r-1)(b-1) EbSS MSEb  
AXB (a-1)(b-1) ABSS MSAB MSAB/MSEc
Error (c) (r-1) (a-1)(b-1) EcSS MSEc  
Split plot factor c-1 CSS MSC MSC/MSEd
(C)
AxC (a-1)(c-1) ACSS MSAC MSAC/MSEd
BxC (b-1)(c-1) BCSS MSBC MSBC/ MSEd
AxBxC (a-1)(b-1)(c-1) ABCSS MSABC MSABC/ MSEd
Error (C) ab(r-1)(c-1) EdSS MSEd  
Total rabc-1 TSS    
 Data Presentation
Type of Depth of Fertilizer rate   Block    
Irrigation Plowing 1 2 3 4

Light
Shallow None
Average
  
 
 
 
 
 
 
 
Heavy        
  Medium None
Average
 
 
 
 
 
 
 
 
 
  Heavy        
  Deep None        
  Average        
 
 
Heavy        
 
             
Heavy Shallow None        
Average        
  Heavy        
 
Medium None        
 
  Average        
  Heavy        
  Deep None        
 
Average        
 
Heavy        
 SUMMARY
STATISTICS ASIDE
FROM ANOVA

 
1. Coefficient of variation(%CV) – experimental error
expressed as percentage of the mean
 -measures degree of precision of the experiment
-higher %CV, lower reliability of experimental result
2. Standard error of a treatment mean(se of a mean) –
measure the average error in estimating the true
treatment mean
-measures degree of a precision of the mean as a
estimate of the true treatment mean
 
3. Standard error of the difference between two
treatment means
-measure the average error in estimating the
difference between two treatment means
a. estimates of treatment means and effects
b. estimates of the effect of treatment