Chapter 2

Randomized Complete Block Design

(RCBD)

“It doesn't matter how beautiful your theory is, it

doesn't matter how smart you are. If it doesn't agree
with experiment, it's wrong.”
― Richard P. Feynman

Dr. Muhammad Atif

Assistant Professor
Department of Statistics
University Of Peshawar
Outlines

 Introduction to RCBD
 Randomization Procedure of RCBD
 RCBD Model
 Estimation of Model Parameters
 Expected Mean Square Column
 Analysis of RCBD
 Estimation of missing value
 Relative efficiency
Dr. Muhammad Atif
Recall the CRD
 A completely randomized design (CRD) is one
where the treatments are assigned completely at
random so that each experimental unit has the same
chance of receiving any one treatment.
 Homogeneity of experimental units
assumptions: The CRD assumes that all the
experimental units are homogenous and the
treatments are applied to them completely at
random without imposing any restriction on
randomization process.
 Variability in experimental units
Example
ExpDes-
4

T2 T1 T3
T1
moisture gradient

T1 T2 T4
T5

T5 T3 T1
T2

T3 T5 T2
T4

T4 T4 T5
T3

Treatment effects confounded with moisture effect!

 Variability in experimental units
Example
ExpDes-
5

- Treatment effects are

moisture gradient
confounded with moisture T2 T1 T1 T3
effect!
T1 T2 T5 T4

T5 T3 T1
T2
Lack of control: With CRD,
there is no control over T3 T5 T4 T2
external factors that may
influence the experimental T4 T4 T3 T5
outcomes. This can make it
difficult to isolate the effects of
the independent variable on
the dependent variable.
Variability in experimental units
 Hence, CRD is appropriate only for experiments
with homogenous experimental units where
environmental effects are relatively easy to control.

 In such situation where there is large variations in

the experimental units, the Randomized Complete
Block Design is more appropriate.
 Blocking
Solution of controlling external variability
ExpDes-
7

T1 T3 T5 T4 T2
moisture gradient

T5 T2 T3 T1 T4

T3 T2 T4 T5 T1

T4 T5 T1 T2 T3

ck effect now removes moisture effect, fair comparisons among treatme

 Randomized Block Design
(RBD)

- Any experimental design in which the randomization

of treatments is done by imposing one restriction i.e.
the Block should be complete

- Complete Block: A block is complete if it contain the

complete set of treatments

- Block: Group of experimental units assumed to be

internally homogeneous
 Randomized Block Design
(RBD)
- Blocks of units are created to control known
sources of variation in among experimental units.

 There are two classifications or factors in an RBD:

block “effects” and treatment “effects”.
Randomization Process
RCBD Model
 Blood Pressure Example
30 - 12

The data below represent blood pressure

measurements from an experiment
involving 4 age groups, each with 3
persons. The 3 persons within each age
group were randomly assigned to drugs A,
B, and C, with one person per drug.
This was done to keep the drug
assignments balanced within age groups.
30 - 13

For this experiment, the major interest

is in comparing the three drugs in a
manner that provides balance for or
controls for possible age effects.
That is, we are interested primarily in
hypotheses concerning means for the
drugs, but we do have a secondary
interest in means for the age groups.
 Blood Pressure Data
30 - 14

Drug
Age Group A B C Total
1 70 72 80 222
2 76 84 82 242
3 82 86 84 252
4 90 92 88 270
Total 318 334 334 986
 Hypotheses
30 - 15

There are two hypotheses of interest. The one of most

importance examines differences among the Drugs
(Treatments), which can be expressed as:
H0: A = B = C vs H1: A  B  C.

The hypothesis that the Age Group (Block) means are

equal can also be tested. This hypothesis can be
written as:
H0: 1 = 2 = 3 = 4 vs H1: 1  2  3  4
30 - 16

We can examine these hypotheses through the

use of an ANOVA procedure appropriate for
this Randomized Blocks design.

For this design, the blocks—age groups in this

case—are selected in order to provide
meaningful balance or control for this factor and
the random assignment to treatment groups is
then done within each block.
 ANOVA for Testing Hypotheses
30 - 17

ANOVA
Source df SS MS F
Treatments t-1 SS(T) SS(T)/df(T) MS(T)/MS(R)
Blocks b-1 SS(B) SS(B)/df(B) MS(B)/MS(R)
Residual (t-1)(b-1) SS(R) SS(R)/df(R)
Total n-1 SS(Total)
30 - 18

For  = 0.05, we reject the null hypothesis:

H0: A = B = C ,
if
F(Treatments) = MS(T)/MS(R) > F0.95[df(T), df (R)].

We reject the null hypothesis about block differences,

H0: 1 = 2 = 3 = 4 ,
if
F(Blocks) = MS(B)/MS(R) > F0.95[df(B), df (R)].
 Drug Age (Age
2 2
Age Group A B C Total Mean Sum /n effect effect) *n
30 - 19
1 70 72 80 222 74.00 16,428.00 -8.17 200.08
2 76 84 82 242 80.67 19,521.33 -1.50 6.75
3 82 86 84 252 84.00 21,168.00 1.83 10.08
4 90 92 88 270 90.00 24,300.00 7.83 184.08
Total 318 334 334 986 82.17 81,016.33 0.00 401.00

Mean 79.50 83.50 83.50 82.17

2
Sum /n 25,281.00 27,889.00 27,889.00 64,813.07

Drug effect -2.67 1.33 1.33 0.00

2
(Drug effect) *n 28.44 7.11 7.11 42.67

Observations squared
Drug
Age Group A B C Total
1 4,900 5,184 6,400 16,484
2 5,776 7,056 6,724 19,556
3 6,724 7,396 7,056 21,176
4 8,100 8,464 7,744 24,308
Total 25,500 28,100 27,924 81,524
 The SS calculations are:
30 - 20

SS(Total) = 702 + 762 +  + 842 + 882 – 9862/12

= 81,524 – 81,016.33
= 507.67

SS(T) = 3182/4 + 3342/4+ 3342/4 – 9862/12

= 81,059 – 81,016.33
= 42.67

SS(B) = 2222/3 + 2422/3 + 2522/3 + 2702/3 – 9862/12

= 81,417.33 – 81,016.33
= 401.00

SS(R) = SS(Total) – SS(T) – SS(B)

= 507.67 – 42.67 – 401
= 64.00
30 - 21

ANOVA
Source df SS MS F
Drug 2 42.67 21.34 1.99
Age Group 3 401.00 133.67 12.47
Residual 6 64.00 10.70
Total 11 507.67
30 - 22

F(Drug) = MS(T)/MS(R) = 21.34/10.7 = 1.99,

is not greater than
F0.95[df(T), df (R)] = F0.95[2, 6] = 5.14,
so H0: A = B = C should not be rejected.

F(Block) = MS(B)/MS(R) = 133.67/10.7 = 12.47

is greater than
F0.95[df(B), df (R)] = F0.95[3, 6] = 4.32
so H0: 1 = 2 = 3 = 4 should be rejected.
 Looking at models
30 - 23

We can consider a model that characteristizes the individual

observations that could be:

yij = μ + τ i + δ j + εij,

where

yij is the data point for the ith treatment and jth block,

τi is the effect for treatment i,

δj is the effect for block j, and

εij is the random effect.

 Simple Repeated Measures Designs
30 - 24

When a measurement is repeated on

each participant so that there are
multiple measurements per person,
then the resulting dependency of
measurements over time on the same
person should be considered
appropriately when analyses are
undertaken.
30 - 25
An example is a study that measured
blood pressure levels at several time
points for persons assigned to one of k
treatment groups.
Recall that we were able to use a paired
t-test for testing hypotheses about
differences between two time points.
If there are more than two time points,
then something else has to be done.
 The simplest solution to this situation
30 - 26

is to use the procedures outlined for

randomized blocks designs discussed
above, whereby each participant is
considered a block.
The following example includes data
for blood pressure measurements over
three different time points for each
person for a total of eight persons.
 Blood pressure measurements at three
time points for eight persons
30 - 27

Subject Baseline Day 3 Day 7 Total

1 70 73 72 215

2 65 71 69 205

3 68 73 74 215

4 73 75 73 221

5 78 80 76 234

6 67 65 71 203

7 72 72 75 219

8 75 81 74 230
Total 568 590 584 1,742
30 - 28

When the repeated measures are on the

same person over time, persons can be
treated as “blocks.” The randomized
block procedure can then be used.
SS(Total ) = (70)2 + (65)2 + … + (74)2 – (1,742)2/24

5682 590 2 584 2 1, 742 2

SS(time)    
8 8 8 24

2152 2052 230 2 1, 742 2

SS(Persons)    ...  
3 3 3 24
 Blood Pressure Subject
2
Subject Baseline Day 3 Day 7 Total Mean Sum /n Effect Effect2*n
1 70 73 72 215 71.67 15,408 -0.92 2.52
2 65 71 69 205 68.33 14,008 -4.25 54.19
3 68 73 74 215 71.67 15,408 -0.92 2.52
4 73 75 73 221 73.67 16,280 1.08 3.52
30 - 29 5 78 80 76 234 78.00 18,252 5.42 88.02
6 67 65 71 203 67.67 13,736 -4.92 72.52
7 72 72 75 219 73.00 15,987 0.42 0.52
8 75 81 74 230 76.67 17,633 4.08 50.02
Total 568 590 584 1,742 72.58 12,6714 0.00 273.83

Mean 71.00 73.75 73.00 72.58 1,7422/24 = 126,440.17

2
Sum /n 40,328 43,512 42,632 126,472
126,822 - 126,440.17 = 381.83
Time effect -1.58 1.17 0.42 0.00
2
(Time effect) *n 20.06 10.89 1.39 32.33 126,472 - 126,440.17 = 32.33

126,714 - 126,440.17 = 273.83

Observations Squared 381.83 - 32.33 - 273.83 = 75.66

Subject Baseline Day 3 Day 7 Total
1 4,900 5,329 5,184 15,413
2 4,225 5,041 4,761 14,027
3 4,624 5,329 5,476 15,429
4 5,329 5,625 5,329 16,283
5 6,084 6,400 5,776 18,260
6 4,489 4,225 5,041 13,755
7 5,184 5,184 5,625 15,993
8 5,625 6,561 5,476 17,662
Total 40,460 43,694 42,668 126,822
30 - 30

SS(Error) = SS(Total) – SS(Treatments) – SS(Persons)

SS(Total) = 126,822 - 126,440.17 = 381.83

SS(Time) = 126,472 - 126,440.17 = 32.33

SS(Persons) = 126,714 - 126,440.17 = 273.83

SS(Error) = 381.83 - 32.33 - 273.83 = 75.66

30 - 31

ANOVA
Source df SS MS F
Time 2 32.33 16.17 2.99
Persons 7 273.83 39.11 7.24
Error 14 75.66 5.4
Total 23 381.83

The null hypothesis for no time differences,

H0: 1 = 2 = 3,
is accepted since
F(Time) = MS(Time) / MS(Error) = 2.99
is less than F0.95(2,14) = 3.74
30 - 32

The null hypothesis for no differences among

persons,

H0: 1 = 2 = 3 = 4 = 5 = 6 = 7 = 8

is rejected since

F(Person) = MS(Person) / MS(Error) = 7.24

is greater than F0.95(7,14) = 2.76