COMPLETELY RANDOM DESIGN (CRD)

Description of the Design

-Simplest design to use.

-Design can be used when experimental units are essentially homogeneous.

-Because of the homogeneity requirement, it may be difficult to use this design for field
experiments.

-The CRD is best suited for experiments with a small number of treatments.


Randomization Procedure

-Treatments are assigned to experimental units completely at random.

-Every experimental unit has the same probability of receiving any treatment.

-Randomization is performed using a random number table, computer, program, etc.

Example of Randomization

-Given you have 4 treatments (A, B, C, and D) and 5 replicates, how many experimental
units would you have?


1
D
2
D
3
B
4
C
5
D
6
C
7
A
8
A
9
B
10
D
11
C
12
B
13
A
14
B
15
C
16
B
17
C
18
D
19
A
20
A

-Note that there is no blocking of experimental units into replicates.

-Every experimental unit has the same probability of receiving any treatment.
1
Advantages of a CRD

1. Very flexible design (i.e. number of treatments and replicates is only limited by
the available number of experimental units).

2. Statistical analysis is simple compared to other designs.

3. Loss of information due to missing data is small compared to other designs due to
the larger number of degrees of freedom for the error source of variation.

Disadvantages

1. If experimental units are not homogeneous and you fail to minimize this variation
using blocking, there may be a loss of precision.

2. Usually the least efficient design unless experimental units are homogeneous.

3. Not suited for a large number of treatments.

Fixed vs. Random Effects

-The choice of labeling a factor as a fixed or random effect will affect how you will make
the F-test.

-This will become more important later in the course when we discuss interactions.

Fixed Effect

-All treatments of interest are included in your experiment.

-You cannot make inferences to a larger experiment.

Example 1: An experiment is conducted at Fargo and Grand Forks, ND. If location is
considered a fixed effect, you cannot make inferences toward a larger area (e.g. the
central Red River Valley).

Example 2: An experiment is conducted using four rates (e.g. X, X, 1.5 X, 2 X) of a
herbicide to determine its efficacy to control weeds. If rate is considered a fixed effect,
you cannot make inferences about what may have occurred at any rates not used in the
experiment (e.g. x, 1.25 X, etc.).

Random Effect

-Treatments are a sample of the population to which you can make inferences.

-You can make inferences toward a larger population using the information from the
2
analyses.

Example 1: An experiment is conducted at Fargo and Grand Forks, ND. If location is
considered a random effect, you can make inferences toward a larger area (e.g. you could
use the results to state what might be expected to occur in the central Red River Valley).

Example 2: An experiment is conducted using four rates (e.g. X, X, 1.5 X, 2 X) of an
herbicide to determine its efficacy to control weeds. If rate is considered a random effect,
you can make inferences about what may have occurred at rates not used in the
experiment (e.g. x, 1.25 X, etc.).

Analysis of the Fixed Effects Model

Notation
Statistical notation can be confusing, but use of the Y-dot notation can help
simplify things.

The dot in the Y-dot notation implies summation across over the subscript it
replaces.

For example,

. experiment in the ns observatio of number total N where mean, Experiment
s treatment of number a where total, Experiment
mean Treatment
treatment a in ns observatio of number n where total, Treatment
.. ..
1
1
. .
1
.
= = =
= = =
= =
= = =

=
=
=
N y y
y y
n y y
y y
n
j
ij
a
i
..
i i
n
j
ij i


Linear Additive Model for the CRD


ij i ij
Y + + =

where: Y
ij
is the j
th
observation of the i
th
treatment,
is the population mean,

i
is the treatment effect of the i
th
treatment, and
is the random error.
ij

3
-Using this model we can estimate
i
or for any observation if we are given Y
ij

ij
and .

Example

Treatment 1 Treatment 2 Treatment 3
4 9 8
5 10 11
6 11 8
Y
i.
15 30 27 Y
..
= 72
. i Y
5 10 9
.. Y = 8
.. . Y Yi
-3 2 1

-We can now write the linear model for each observation ( ).
ij
Y

-Write in for each observation.


Treatment 1 Treatment 2 Treatment 3
4 = 8 9 = 8 8 = 8
5 = 8 10 = 8 11 = 8
6 = 8 11 = 8 8 = 8
Y
i.
15 30 27 Y
..
= 72
. i Y
5 10 9
.. Y = 8
.. . Y Yi
-3 2 1


-Write in the respective
i
for each observation where ..
.
Y Y
i i
=


Treatment 1 Treatment 2 Treatment 3
4 = 8 3 9 = 8 + 2 8 = 8 + 1
5 = 8 3 10 = 8 + 2 11 = 8 + 1
6 = 8 3 11 = 8 + 2 8 = 8 + 1
Y
i.
15 30 27 Y
..
= 72
. i Y
5 10 9
.. Y = 8
.. . Y Yi
-3 2 1


4
-Write in the for each observation.
ij

Treatment 1 Treatment 2 Treatment 3
4 = 8 3 - 1 9 = 8 + 2 - 1 8 = 8 + 1 - 1
5 = 8 3 + 0 10 = 8 + 2 + 0 11 = 8 + 1 + 2
6 = 8 3 + 1 11 = 8 + 2 + 1 8 = 8 + 1 - 1
Y
i.
15 30 27 Y
..
= 72
. i Y
5 10 9
.. Y = 8
.. . Y Yi
-3 2 1

-Note for each treatment .

= 0
ij

-If you are asked to solve for
3
, what is the answer?

-If you are asked to solve for
23
, what is the answer?

-Question: If you are given just the treatment totals ( s), how would you fill in the
. i
Y
values for each of the observations such that the Error SS = 0.

Answer: Remember that the Experimental Error is the failure of observations
treated alike to be the same. Therefore, if all treatments have the same value in
each replicate, the Experimental Error SS =0.

Example

Given the following information, fill in the values for all s such that the Experimental
Error SS = 0.
ij
Y

Treatment 1 Treatment 2 Treatment 3



Y
i.
15 30 27 Y
..
= 72
. i Y
5 10 9
.. Y = 8

5
Answer

Treatment 1 Treatment 2 Treatment 3
5 10 9
5 10 9
5 10 9
Y
i.
15 30 27 Y
..
= 72
. i Y
5 10 9
.. Y = 8


Note in the previous two examples that . 0

=
i
This is true for all situations.
Given
mean). experiment the equals
s treatment of number by the divided means treatment the of sum the (i.e.,
) i' (i, s treatment of pair one least at for :
... :
1
'
2 1 0

= = =

=
t
H
H
t
i
i
i i A
t


). ( 0 that implies definition This
.. .
1
i
Y Y
i
t
i
= =

=


The hypothesis written above can be rewritten in terms of the treatment effects
i
as:

i. least at for 0 :
0 ... :
2 1 0

= = = =
i A
a
H
H

Thus, when we are testing the null hypothesis that all treatments means are the same,
we are testing at the same time the null hypothesis that all treatment effects,
i,
are
zero.

Partitioning the Total Sum of Squares
Remember that:
) (
) (
.
.. .
..
i ij ij
i i
Y Y
Y Y
Y
=
=
=

Thus, can be rewritten as:
ij i ij
Y + + = ) ( ) (
. .. . .. i ij i ij
Y Y Y Y Y Y + + = .

The Analysis of Variance is derived from the partitioning of the corrected Total Sum
of Squares.

6
) )( ( 2 ) ( ) (
)] ( ) [( ) (
) ( Squares of Sum Total
.
1
.. .
t
1 i
2
1
..
t
1 i
t
1 i
2
.. .
2
.
1
.. .
a
1 i 1
2
..
t
1 i
1
2
..
t
1 i
i ij
r
j
i
r
j
i ij i
i ij
n
j
i
r
j
ij
r
j
ij
Y Y Y Y Y Y Y Y r
Y Y Y Y Y Y
and
Y Y
+ + =
+ =
=



= = = = =
= = = =
= =


The last term of the equation equals zero because

= =
=
0 ) (
1
. ij
r
j
i ij
Y Y .
Thus,
2
1
..
t
1 i
t
1 i
2
.. .
1
2
..
t
1 i
) ( ) ( ) (

= = = = =
+ =
r
j
i ij i
r
j
ij
Y Y Y Y r Y Y , which is

Total Sum of Squares = Treatment Sum of Squares + Error Sum of Squares


ANOVA for Any Number of Treatments with Equal Replication

Given the following data:

Treatment
Replicate A B C
1 23 42 47
2 36 26 43
3 31 47 43
4 33 34 39
. i
Y 123 149 172 Y
...
=444

2
ij
Y
3,875 5,805 7,428

Step 1. Write the hypotheses to be tested.

3 2 1
3 2 1
3 2 1
3 2 1
:
:





=
=
= =
or
or
H
H
A
o

H
o
: All three means are equal.

H
A
: At least one of the means is different from
the other means.


7
Step 2. Calculate the Correction Factor.

0 . 428 , 16
3 * 4
444
2 2
..
= = =
rt
Y
CF

Step 3. Calculate the Total SS


0 . 680
428 , 16 108 , 17
) 39 ... 31 36 23 (
2 2 2 2
2
=
=
+ + + + =
=

CF
CF Y TotalSS
ij

Step 4. Calculate the Treatment SS (TRT SS)


5 . 300
0 . 16428 5 . 16728
428 , 16
4
172
4
149
4
123
2 2 2
2
.
=
=

+ + =
=

CF
r
Y
TRTSS
i



Step 5. Calculate the Error SS

Error SS = Total SS Treatment SS

= 680 300.5

= 379.5
8
Step 6. Complete the ANOVA table


Sources of variation Df SS MS F
Treatment t-1 = 2 300.5 150.25 3.563
NS
Error t(r-1) = 9 379.5 42.167
Total rt-1 = 11 680.0


Step 7. Look up Table F-values.

F
0.05;2,9
= 4.26
F
0.01;2,9
= 8.02

Step 8. Make conclusions.










0 3.563 4.26 8.02
-Since F-calc (3.563) < 4.26 we fail to
reject Ho:
3 2 1
= = at the 95%
level of confidence.
-Since F-calc (3.563) < 8.02 we fail to
reject Ho:
3 2 1
= = at the 99%
level of confidence.


Step 9. Calculate Coefficient of Variation (CV).

100 * %
Y
s
CV =

Remember that the Error MS = s
2
.

( )
% 6 . 17
100 * 37 / 494 . 6
100 *
3 * 4
444
167 . 42
%
=
=

= CV

9
ANOVA for Any Number of Treatments with Unequal Replication

Given the following data:

Treatment
Replicate A B C D
1 2.0 1.7 2.0 2.1
2 2.2 1.9 2.4 2.2
3 1.8 1.5 2.7 2.2
4 2.3 2.5 1.9
5 1.7 2.4
. i
Y 10 5.1 12 8.4 Y
..
=35.5

2
ij
Y
20.26 8.75 29.06 17.7

Step 1. Write the hypotheses to be tested.

4 3 2 1
: = = =
o
H
H
A
: At least one of the means is different from one of the other means.

Step 2. Calculate the Correction Factor.

132 . 74
17
5 . 35
2 2
..
= = =
i
r
Y
CF

Step 3. Calculate the Total SS


638 . 1
132 . 74 77 . 75
) 9 . 1 ... 8 . 1 2 . 2 0 . 2 (
2 2 2 2
2
=
=
+ + + + =
=

CF
CF Y TotalSS
ij

10
Step 4. Calculate the Treatment SS (TRT SS)

978 . 0
132 . 74 110 . 75
132 . 74
4
4 . 8
5
12
3
1 . 5
5
10
2 2 2 2
2
.
=
=

+ + + =
=

CF
r
Y
TRTSS
i
i



Step 5. Calculate the Error SS

Error SS = Total SS Treatment SS

= 1.638 0.978

= 0.660


Step 6. Complete the ANOVA table

Sources of variation Df SS MS F
Treatment t-1 = 3 0.978 0.326 6.392
**
Error By subtraction = 13 0.660 0.051
Total Total number of observations -1 = 16 1.638

Step 7. Look up Table F-values.

F
0.05;3,13
= 3.41
F
0.01;3,13
= 5.74

Step 8. Make conclusions.









0 3.41 5.74 6.39
Since F-calc (6.392) > 3.41 we
reject Ho:
4 3 2 1
= = = at
the 95% level of confidence.
Since F-calc (6.392) > 5.74 we
reject Ho:
4 3 2 1
= = = at
the 99% level of confidence.
11


Step 9. Calculate Coefficient of Variation (CV).

100 * %
Y
s
CV =

Remember that the Error MS = s
2
.

( )
% 82 . 10
100 * 088 . 2 / 2259 . 0
100 *
17
5 . 35
051 . 0
%
=
=

= CV


ANOVA with Sampling (Equal Number of Samples Per Experimental Unit)

Linear Model
ijk ij i ijk
Y + + + =

Where: Y
ijk
is the k
th
sample of the j
th
observation of the i
th
treatment,
is the population mean,

i
is the treatment effect of the i
th
treatment,
is the random error, and
ij

is the sampling error.

ijk

ANOVA table

SOV Df F
Treatment t-1 Treatment MS/Experimental Error MS
Experimental error (tr-1) - (t-1)
Sampling Error (trs-1) - (tr-1)
Total trs-1

12
Facts about ANOVA with Sampling

-There are two sources of variation that contribute to the variance appropriate to
comparisons among treatment means.

1. Sampling Error = variation among sampling units treated alike ( ).
2
s

2. Experimental Error = variation among experimental units treated alike

( ).
2 2
E s
r +

-The Experimental Error MS is expected to be larger than the Sampling Error MS.

-If the Experimental Error variance component is not important, the Sampling Error MS
and the Experimental Error MS will be of the same order of magnitude.

-If the Experimental Error variance component is important, the Experimental Error MS
will be much larger than the Sampling Error MS.

Example

Temperature
8
o
12
o
16
o

Pot number Pot number Pot number
Plant 1 2 3 1 2 3 1 2 3
1 3.5 2.5 3.0 5.0 3.5 4.5 5.0 5.5 5.5
2 4.0 4.5 3.0 5.5 3.5 4.0 4.5 6.0 4.5
3 3.0 5.5 2.5 4.0 3.0 4.0 5.0 5.0 6.5
4 4.5 5.0 3.0 3.5 4.0 5.0 4.5 5.0 5.5
Y
ij.
15.0 17.5 11.5 18.0 14.0 17.5 19.0 21.5 22.0
Y
i..
44.0 49.5 62.5
Y

=156.0

Note i = treatment, j = replicate, and k = sample.

Step 1. Calculate correction factor:

676
) 4 )( 3 ( 3
156
2 2
...
= =
rts
Y


13
Step 2. Calculate the Total SS:


( )
5 . 36
0 . 676 5 . 712
5 . 5 ... 0 . 3 0 . 4 5 . 3
2 2 2 2
2
=
=
+ + + + =
=

CF
CF Y TotalSS
ijk

Step 3. Calculate the Treatment SS:


042 . 15
0 . 676 04 . 691
0 . 676
) 4 ( 3
5 . 62
) 4 ( 3 (
5 . 49
) 4 ( 3
44
2 2 2
2
..
=
=

+ + =
=

CF
rs
Y
S TreatmentS
i


Step 4. Calculate the SS Among Experimental Units Total (SSAEUT)


25 . 23
0 . 676 25 . 699
0 . 676
4
0 . 22
...
4
5 . 11
4
5 . 17
4
15
2 2 2 2
2
.
=
=

+ + + + =
=

CF
s
Y
SSAEUT
ij

14

Step 5. Calculate the Experimental Error SS:

Experimental Error SS = SSAEUT SS TRT

= 23.25 15.042

= 8.208

Step 6. Calculate the Sampling Error SS:

Sampling Error SS = Total SS SSAEUT

= 36.5 23.25

= 13.25

Step 7. Complete the ANOVA Table:

SOV Df SS MS F
Treatment t-1 = 2 15.042 7.521 5.498
*
Experimental Error (tr-1) - (t-1) = 6 8.208 1.368
Sampling Error (trs-1) - (tr-1) = 27 13.25
Total trs-1 = 35 36.5


Step 8 Look up Table F-values.

F
0.05;2,6
= 5.14
F
0.01;2,6
= 10.92

Step 8. Make conclusions.










0 5.14 10.92

5.498
Since F-calc (5.498) > 5.14 we
reject Ho: at the 95% level of
confidence.
Since F-calc (5.498) < 10.92 we
fail to reject Ho: at the 99% level
of confidence.
15
ANOVA When the Number of Subsamples are Not Equal

=
ions ofobservat total
Y
Y TotalSS
ijk
#
2
... 2
df = #observations 1

=
. #
2
...
2
..
ofobs total
Y
s r
Y
S TreatmentS
k j
i
df = # treatments 1

=
. #
2
...
2
.
ofobs total
Y
s
Y
SSAEUT
k
ij
df = # Experimental units 1

SS Experimental Error = SSAEUT SS TRT df = SSAEUT df TRT df

SS Sampling Error = Total SS SSAEUT df = Total df SSAEUT df





Assumptions Underlying ANOVA

Experimental errors are random, independently, and normally distributed about a
mean of zero and with a common variance (i.e. treatment variances are
homogenous).

The above assumption can be express as NID(0,
2
).

Departure from this assumption can affect both the level of significance and the
sensitivity of F- or t-tests to real departures from H
o
:

This results in the rejection of Ho when it is true (i.e. a Type I Error) more often
than calls for.

The power of the test also is reduced if the assumption of NID(0,
2
) is violated.

Violation of the assumption NID(0,
2
) with the fixed model is usually of little
consequence because ANOVA is a very robust technique.

Violation of the basic assumptions of ANOVA can be investigated by observing
plots of the residuals.

Residuals will be discussed in more detail when Transformations are discussed later
in the semester.
16