Definition

It is a design wherein the allocation of

treatments is done by randomizing the
treatments completely over the experimental
units (eu’s) without any restriction imposed on
the units and there is only one criterion for
data classification.
 Advantages

1. Complete flexibility is allowed; any number

of treatments and replicates may be used.
2. Relatively easy statistical analysis, even with
variable replicates and variable experimental
errors for different treatments.
 CRD is commonly used when:

1. Most basic experimental design

2. Design where allocation of the treatments is
done by randomizing the treatment
completely over the entire set of eu’s
without any restriction imposed on the units
 CRD is commonly used when:
3. The treatment level assigned is the only
criterion for data classification
4. Used when the eu’s are sufficiently
homogeneous
5. There is effective local control is assured (as
those in laboratories or greenhouse).
 Requirements
1. n (n>t) homogeneous eu’s t treatments to
compare
2. For balanced design, n = r x t where r is the
number of replications per treatment
3. For unbalanced design, n = r1 + r2 +…. + rn where
ri is the number of replicates of the ith treatment.
 Steps

1. Determine the number of replications per

treatment.
2. Label the eu’s from 1 to n.
3. Generate n random numbers.
4. Rank these random numbers.
 Steps
5. The first r1 ranks will correspond to the eu’s which
will receive the 1st level of the treatment.
6. The next r2 ranks will correspond to the eu’s
which will receive the 2nd level of the treatment.
7. Continue the process until all units have been
assigned to the last level of the treatment.
 Randomization and Layout

Suppose there are t=3 treatments, T1, T2, T3, which

are replicated r1 = 2; r2 = 3; and r3 = 4 times,
respectively; hence the numbers of eu’s is n = r1 + r2 +
r3 =9. The randomization (using random number
generator key on calculator) may be as follows:
 Randomization and Layout
1. Label the eu’s consecutively from 1 to n = 9.
2. Obtain a sequence of n = 9 random numbers. Rank
the numbers in increasing order. Using the
sequence of ranks as a randomization of the eu’s,
assign the first r1 = 2 eu’s to T1, the next r2 = 3eu’s
to T2 and the last r3 = 4 eu’s to T3.
 Randomization and Layout
Random no:
0.739 0.218 0.781 0.396 0.527 0.324 0.463 0.064 0.942
7 2 8 4 6 3 5 1 9

T1 T2 T3
 Data Presentation
Treatment Observations Replication Total Mean
T1 Y11 Y12 … Y11 r1 Y1 Y1
T2 Y21 Y22 … Y11 r2 Y2 Y2
: : : … : : : :
Tt Yt1 Yt1 … Y11 rt Yt Yt
 Example of CRD Data Presentation

Three methods of soil analysis, M1, M2, M3, were

tried by a research institute. Twelve uniform soil
samples were taken from a certain farm and M1 was
randomly assigned to 5 of the soil samples, M2 to 3
samples and M3 to 4 samples. The researcher was
interested in the time to compete the sol analysis.
 Example of CRD Data Presentation
Method Time to complete analysis (hours) Reps Total Mean
M1 2.4 3.8 2.9 4.6 3.1 5 16.8 3.36
M2 4.8 1.6 0.2 0.0 0.0 3 6.6 2.20
M3 7.2 5.3 2.9 3.5 0.0 4 18.9 4.725
12 42.3 3.428

At the level of significance 𝜶=5%, test the hypothesis that

there is no difference in the time to complete the soil
analysis for the three methods.
 Example of CRD Data Presentation
Testing the significance of treatment effects via
ANOVA F-test:
1. State the statistical hypotheses
Ho: μ1 = μ2 = μ3 → All method means are equal.
Ha: μ1 = μ2 ≠ μ3, → At least two method means are
different.
 Example of CRD Data Presentation

2. Formulate test statistic and get its critical value at

the level of significance 𝜶
MSTr
Test statistic: Fcrital (Fcrit) =
MSE
Critical value: Ftabular (Ftab ) = Fɑ [(t-1), (n-t)]
 Example of CRD Data Presentation

3. State the decision rule

Reject Ho and accept Ha if Fcrit ≥ Ftab; otherwise
failed to reject Ho.
Alternatively, reject Ho and accept Ha if (Fcrit ≥
Ftab) < 𝛼; otherwise failed to reject Ho.
 Example of CRD Data Presentation

4. Construct the ANOVA table outlined as follows:

Source of Degrees of Sum of Mean
Fcrital Ftabular
Variation (SV) Freedom (df) Squares (SS) Square (MS)
MSTr Fɑ [(t-1),
Treatments t–1 TrSS MSTr MSE (n-t)]
Experimental Error n–t ESS MSE
Total n–1 TSS
 (𝑮𝒓𝒂𝒏𝒅 𝑻𝒐𝒕𝒂𝒍)𝟐
𝑪𝒐𝒓𝒓𝒆𝒄𝒕𝒊𝒐𝒏 𝑭𝒂𝒄𝒕𝒐𝒓 𝑪𝑭 =
𝒏
(𝟒𝟐. 𝟑)𝟐
=
𝟏𝟐
𝟏, 𝟕𝟖𝟗. 𝟐𝟗
=
𝟏𝟐
= 𝟏𝟒𝟗. 𝟏𝟎𝟖
 Source of Degrees of Sum of Mean
Fcrital Ftabular
Variation (SV) Freedom (df) Squares (SS) Square (MS)
MSTr Fɑ [(t-1),
Treatments 2 11.163 MSTr MSE (n-t)]
Experimental Error 9 25.339 MSE
Total 11 36.502
 (𝑺𝒖𝒎 𝒐𝒇 𝑴𝒆𝒕𝒉𝒐𝒅 𝟏)𝟐 (𝑺𝒖𝒎 𝒐𝒇 𝑴𝒆𝒕𝒉𝒐𝒅 𝟐)𝟐 (𝑺𝒖𝒎 𝒐𝒇 𝑴𝒆𝒕𝒉𝒐𝒅 𝟑)𝟐
𝑻𝒓𝑺𝑺 = + + − 𝑪𝑭
𝑹𝒆𝒑𝒍𝒊𝒄𝒂𝒕𝒊𝒐𝒏 𝑹𝒆𝒑𝒍𝒊𝒄𝒂𝒕𝒊𝒐𝒏 𝑹𝒆𝒑𝒍𝒊𝒄𝒂𝒕𝒊𝒐𝒏
(𝟏𝟔. 𝟖)𝟐 (𝟔. 𝟔)𝟐 (𝟏𝟖. 𝟗)𝟐
𝑻𝒓𝑺𝑺 = + + − 𝑪𝑭
𝟓 𝟑 𝟒
𝟐𝟖𝟐. 𝟐𝟒 𝟒𝟑. 𝟓𝟔 𝟑𝟓𝟕. 𝟐𝟏
= + + − 𝟏𝟒𝟗. 𝟏𝟎𝟖
𝟓 𝟑 𝟒
= 𝟓𝟔. 𝟒𝟒𝟖 + 𝟏𝟒. 𝟓𝟐 + 𝟖𝟗. 𝟑𝟎𝟑 − 𝟏𝟒𝟗. 𝟏𝟎𝟖
= 𝟏𝟔𝟎. 𝟐𝟕𝟏 − 𝟏𝟒𝟗. 𝟏𝟎𝟖 = 𝟏𝟏. 𝟏𝟔𝟑
𝑻𝑺𝑺 = (𝑶𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏 𝟏)𝟐 +(𝑶𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏 𝟐)𝟐 +(𝑶𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏 𝟑)𝟐 +(𝑶𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏 𝒏)𝟐 −𝑪𝑭

𝑻𝑺𝑺 = [ 𝟐. 𝟒 𝟐 + 𝟑. 𝟖 𝟐+ 𝟐. 𝟗 𝟐+ 𝟒. 𝟔 𝟐+ 𝟑. 𝟏 𝟐+ 𝟒. 𝟖 𝟐 𝟏. 𝟔 𝟐+

𝟎. 𝟐 𝟐 + 𝟕. 𝟐 𝟐
+ 𝟓. 𝟑 𝟐
+ 𝟐. 𝟗 𝟐 + 𝟑. 𝟓 𝟐 ] − 𝑪𝑭

𝑻𝑺𝑺 = [𝟓. 𝟕𝟔 + 𝟏𝟒. 𝟒𝟒 + 𝟖. 𝟒𝟏 + 𝟐𝟏. 𝟏𝟔 + 𝟗. 𝟔𝟏 + 𝟐𝟑. 𝟎𝟒 + 𝟐. 𝟓𝟔 +

𝟎. 𝟎𝟒 + 𝟓𝟏. 𝟖𝟒 + 𝟐𝟖. 𝟎𝟗 + 𝟖. 𝟒𝟏 + 𝟏𝟐. 𝟐𝟓] − 𝟏𝟒𝟗. 𝟏𝟎𝟖

𝑻𝑺𝑺 = 𝟏𝟖𝟓. 𝟔𝟏 − 𝟏𝟒𝟗. 𝟏𝟎𝟖 = 𝟑𝟔. 𝟓𝟎𝟐

 Source of Degrees of Sum of Mean
Fcrital Ftabular
Variation (SV) Freedom (df) Squares (SS) Square (MS)
MSTr Fɑ [(t-1),
Treatments 2 11.163 MSTr MSE (n-t)]
Experimental Error 9 25.339 MSE
Total 11 36.502

𝑬𝑺𝑺 = 𝑻𝑺𝑺 − 𝑻𝒓𝑺𝑺

𝑬𝑺𝑺 = 𝟑𝟔. 𝟓𝟎𝟐 − 𝟏𝟏. 𝟏𝟔𝟑

= 𝟐𝟓. 𝟑𝟑𝟗
 Source of Degrees of Sum of Mean
Fcrital Ftabular
Variation (SV) Freedom (df) Squares (SS) Square (MS)
MSTr Fɑ [(t-1),
Treatments 2 11.163 5.582 MSE (n-t)]
Experimental Error 9 25.339 2.815
Total 11 36.502

𝑻𝒓𝑺𝑺 𝟏𝟏. 𝟏𝟔𝟑

𝑴𝑺𝑻𝒓 = = = 𝟓. 𝟓𝟖𝟐
𝒅𝒇𝒕 𝟐
𝑬𝑺𝑺 𝟐𝟓. 𝟑𝟑𝟗
𝑴𝑺𝑬 = = = 𝟐. 𝟖𝟏𝟓
𝒅𝒇𝒆 𝟗
 Source of Degrees of Sum of Mean
Fcrital Ftabular
Variation (SV) Freedom (df) Squares (SS) Square (MS)
Treatments 2 11.163 5.582 1.983 4.26
Experimental Error 9 25.339 2.815
Total 11 36.502

𝑴𝑺𝑻𝒓 𝟓. 𝟓𝟐𝟖
𝑭𝒄𝒓𝒊𝒕 = = = 𝟏. 𝟗𝟖𝟑
𝑴𝑺𝑬 𝟐. 𝟖𝟏𝟓

𝒅𝒇𝟏 = 𝟐
𝑭𝒕𝒂𝒃 = 𝟒. 𝟐𝟔
𝒅𝒇𝟐 = 𝟗
 Example of CRD Data Presentation

5. State the decision and make the conclusion

Based on ANOVA results shown in the table, the
computed F-crit of 1.983 is lower than the F-tab of 4.26
at 5% level of significance. It means that there is no
significant differences in the average (mean) time to
complete the soil analysis for the three methods.