Republic of the Philippines

ZAMBOANGA STATE COLLEGE OF MARINE SCIENCES ANDTECHNOLOGY

Fort Pilar, Zamboanga City
Tel No.: (062) 991-0643 Telefax: (062) 991-0777 website: http://www.zscmst.edu.ph

Module for
PM 212B(Research and Statistics)

Prepared by

MARLON R. OMATANG, MAEd

Associate Professor 3

GRADUATE SCHOOL

Page 1 of 17 PM212 B (Research and Statistics)

 Module 4:
Parametric Tests : F- test

CONTENT :

 F-test (One - Way ANOVA)

 F-test (Two - Way ANOVA)

Objectives:
At the end of the lesson, the students are expected to:
1. Explain and illustrate when, why, and how to use F-tests (One-Way and Two-
Way ANOVA).
2. Formulate the hypothesis of the problems on F-tests (One-Way and Two- Way
ANOVA).
3. Analyze and solve problems on F-tests (One-Way and Two- Way ANOVA).
4. Use SPSS and Excel to conduct statistical analysis
5. Demonstrate their ability to interpret statistical outputs for decisions

Introduction

In the previous Modules , we learned how to perform inference for a population

mean from one sample and also how to compare population means from two samples
(The Independent and Paired Sample t-test).

In this Module, the F-test or Analysis of Variance (ANOVA) will be introduced.

ANOVA is another parametric test that analyzes variances and used to determine if the
means from more than two populations are the same. In other words, we have a
quantitative response variable and a categorical explanatory variable with more than two
levels.

Page 2 of 17 PM212 B (Research and Statistics)

 Lecture Proper: 1. F-test (One-Way ANOVA)

F- test is another parametric test used to compare the means of two or more groups of
independent samples. Also known as analysis of Variance (ANOVA).

A. Why do we use the F-test?

 The F-test is used to find out if there is a significant difference between and among the
means of the two or more independent groups.

B. When do we use the F-test?

 The F-test is used when there is a normal distribution and level of measurement is
expressed in interval or ratio data.

C. How do we use the F-test?

 To get the F- computed value,

The following are the Steps for the Computation:

𝐺𝑇 2 (∑ 𝑋1 + ∑ 𝑋2 + ∑ 𝑋3 + … + ∑ 𝑋𝑛 )2
(1) CF = =
𝑁 𝑛1 + 𝑛2 + 𝑛3 + ... + 𝑛𝑟

(2) TSS = ∑ 𝑋12 + ∑ 𝑋22 + ∑ 𝑋32 +. . . + ∑ 𝑋𝑛2 - CF

(∑ 𝑋1 )2 (∑ 𝑋2 )2 (∑ 𝑋3 )2 (∑ 𝑋4 )2
(3) BSS = + + +...+ − 𝐶𝐹
𝑛1 𝑛2 𝑛3 𝑛𝑟

(4) WSS = TSS - BSS

Where:
CF = Correction Factor(the Square of the Grand Total divided by total samples)
GT= Grand Total
TSS = Total Sum of Squares (the Total sum of squares minus Correction Factor)
BSS = Between Sum of Squares (Between sum of squares divided by its corresponding
samples minus Correction Factor
WSS = Within Sum of Squares (TSS minus BSS)

Page 3 of 17 PM212 B (Research and Statistics)

 ANOVA Table
Sources of Degrees of Sum of Mean F- Value Decision
Variation Freedom squares Squares Computed Tabular/Critical
(df) (SS) (MS)
Between k-1 BSS 𝐵𝑆𝑆
Groups 𝑑𝑓 𝑀𝑆𝐵
𝐹=
Within (N-1)-(k-1) WSS 𝑊𝑆𝑆 𝑀𝑆𝑊
Groups 𝑑𝑓
Total N-1 TSS

Page 4 of 17 PM212 B (Research and Statistics)

 A. Steps in Testing the Hypothesis:
1. Problem
2. Hypotheses
3. Level of Significance
4. Statistics
5. Decision Rule
6. Conclusion

Example #1.

A professor is administering a 10-item quiz in Statistics to his students in the graduate school of 4
programs. He wants to find out if there is a significant difference in the average scores of their students in
the 4 programs. Use  = 0.05 level of significance. The following data are recorded.

Programs
A B C D
9 2 3 8
8 2 4 3
8 4 9 4
7 3 3 7
6 4 5 5
10 5 6 5
9 6 7 4

Solution:

1. Problem: Is there a significant difference in the average scores of their students in the 4
programs ?

2. Hypotheses :
Null Hypothesis (Ho): There is no significant difference in the average scores of their
students in the 4 programs.

Alternative Hypothesis (Ha): There is a significant difference in the average scores of

their students in the 4 programs. (two – tailed test)

3. Level of Significance:
 = 0.05
df = k -1 = 4 – 1 = 3 and (N-1)-(k-1) = (28-1) – (4-1) = 27-3 = 24

Tabular value = 3.01 (see at the Appendix G)

4. Statistics: F-Test (One-Way ANOVA)

Page 5 of 17 PM212 B (Research and Statistics)

 A A2 B B2 C C2 D D2
9 81 2 4 3 9 8 64
8 64 2 4 4 16 3 9
8 64 4 16 9 81 4 16
7 49 3 9 3 9 7 49
6 36 4 16 5 25 5 25
10 100 5 25 6 36 5 25
9 81 6 36 7 49 4 16
57 475 26 110 37 225 36 204

𝐺𝑇 2 (∑ 𝑋1 + ∑ 𝑋2 + ∑ 𝑋3 + ∑ 𝑋4 )2 (57+26+37+ 36)2 1562 24,336

(1) CF = = = = = = 869.14
𝑁 𝑛1 + 𝑛2 + 𝑛3 + 𝑛4 7+7+7+7 28 28

(2) TSS = ∑ 𝑋12 + ∑ 𝑋22 + ∑ 𝑋32 + ∑ 𝑋42 - CF = 475 + 110 + 225 + 204 – 869.14

= 144.86

(∑ 𝑋1 )2 (∑ 𝑋2 )2 (∑ 𝑋3 )2 (∑ 𝑋4 )2
(3) BSS = + + +...+ − 𝐶𝐹
𝑛1 𝑛2 𝑛3 𝑛𝑟

(57)2 (26)2 (37)2 (36)2

= + + + − 𝐶𝐹 = 941.42 – 869.14 = 72.28
7 7 7 7

(4) WSS = TSS - BSS = 144.86 – 72.28 = 72.58

ANOVA Table
Sources of Degrees of Sum of Mean F- Value Decision
Variation Freedom Squares Squares Computed Tabular/Critical
(df) (SS) (MS)
Between 3 72.28 24.09 Reject Ho
Groups 7.98 3.01
Within 24 72.58 3.02 Significant
Groups
Total 27 TSS

5. Decision Rule:
If the absolute computed F-value is greater than or equal to the absolute F-tabular
value, Reject the Null Hypothesis (HO).

Page 6 of 17 PM212 B (Research and Statistics)

 6. Conclusion :
Since, the F - computed value of 7.98 is greater than the F - tabular value of 3.01 at
0.05 level of significance with 3 and 24 degrees of freedom, then the Null Hypothesis is
rejected which means that there is a significant difference in the average scores of their
students in the 4 programs.

Excel Output

Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
A 7 57 8.1428571 1.80952381
B 7 26 3.7142857 2.23809524
C 7 37 5.2857143 4.9047619
D 7 36 5.1428571 3.14285714

ANOVA
Source of
Variation SS df MS F P-value F crit
Between
Groups 72.28571429 3 24.095238
Within Groups 72.57142857 24 3.0238095 7.96850394 0.000738 3.008787
Total 144.8571429 27

1. Decision Rule:
If the absolute computed F-value is greater than or equal to the absolute F-tabular
value, Reject the Null Hypothesis (HO).

2. Conclusion :
Since, the F - computed value of 7.98 is greater than the F - tabular value of 3.01
at 0.05 level of significance with 3 and 24 degrees of freedom, then the Null Hypothesis
is rejected which means that there is a significant difference in the average scores of their
students in the 4 programs. Implying that students in program A perform better than in
program B, C, & D in the Statistics quiz.

Page 7 of 17 PM212 B (Research and Statistics)

SPSS Output

ANOVA
Sum of
Squares df Mean Square F Sig.
Between
72.286 3 24.095 7.969 .001
Groups
Within Groups 72.571 24 3.024
Total 144.857 27

Descriptives

95% Confidence
Std. Interval for Mean
Deviatio Lower Upper
N Mean n Std. Error Bound Bound Minimum Maximum
A 7 8.1429 1.34519 .50843 6.8988 9.3869 6.00 10.00
B 7 3.7143 1.49603 .56544 2.3307 5.0979 2.00 6.00
C 7 5.2857 2.21467 .83707 3.2375 7.3339 3.00 9.00
D 7 5.1429 1.77281 .67006 3.5033 6.7824 3.00 8.00
Total 28 5.5714 2.31626 .43773 4.6733 6.4696 2.00 10.00
Multiple Comparisons

Mean 95% Confidence Interval

(I) (J)
GROUP GROUP Difference (I-J) Std. Error Sig. Lower Bound Upper Bound
A B 4.42857* .92949 .001 1.6360 7.2211
*
C 2.85714 .92949 .043 .0646 5.6497
D 3.00000* .92949 .032 .2075 5.7925
B A -4.42857* .92949 .001 -7.2211 -1.6360
C -1.57143 .92949 .431 -4.3640 1.2211
D -1.42857 .92949 .513 -4.2211 1.3640
C A -2.85714* .92949 .043 -5.6497 -.0646
B 1.57143 .92949 .431 -1.2211 4.3640
D .14286 .92949 .999 -2.6497 2.9354
*
D A -3.00000 .92949 .032 -5.7925 -.2075
B 1.42857 .92949 .513 -1.3640 4.2211
C -.14286 .92949 .999 -2.9354 2.6497
*. The mean difference is significant at the 0.05 level.

Page 8 of 17 PM212 B (Research and Statistics)

 1. Decision Rule:
If the absolute computed F-value is greater than or equal to the absolute F-tabular
value, Reject the Null Hypothesis (HO).

2. Conclusion :
Since, the F - computed value of 7.98 is greater than the F - tabular value of 3.01
at 0.05 level of significance with 3 and 24 degrees of freedom, then the Null Hypothesis
is rejected which means that there is a significant difference in the average scores of their
students in the 4 programs. Implying that students in program A perform better than in
program B, C, & D in the Statistics quiz.

To have more understanding, click the link:

1. https://www.youtube.com/watch?v=nmHFFFpOVZs&t=17s
2. https://www.youtube.com/watch?v=rS3k8ONVN-o
3. https://www.youtube.com/watch?v=jYn5Jv7Gh4s

Page 9 of 17 PM212 B (Research and Statistics)

 Lecture Proper: 2. F-test (Two-Way ANOVA)

A. What is F-test two – way ANOVA

A two-way ANOVA test is a statistical test used to determine the effect of two nominal
predictor variables on a continuous outcome variable. It is used to tests the effect of two
independent variables on a dependent variable and it is a test that analyzes the effect of the
independent variables on the expected outcome along with their relationship to the outcome
itself.

B. How do you use the F-test two way ANOVA with interaction effect?

Sources of df Sum of Mean F - Value Interpretation

Variation Squares Square Computed Tabular
(SS) (MS)
Between
𝑆𝑆𝑐 𝐴
Groups (c-1) SSC F-Table
𝑐−1 𝐷
𝑆𝑆𝑟 𝐵
Rows (r-1) SSr F-Table
𝑟−1 𝐷
𝑆𝑆𝑐𝑟 𝐶
Interaction (c-1)(r-1) SSCr F-Table
(𝑐 − 1)(𝑟 − 1) 𝐷
Within 𝑆𝑆𝑤
Group k(n-1) SSW
K(n − 1)
Total N-1 SST

Example # 2

Thirty (30) MPM students were randomly assigned to one of the three Professors in
graduate school and to one of the two methods of teaching then achievement was measured on a
test administered at the end of the semester. Use the two-way ANOVA with interaction effect at
0.05 level of significance to test the following hypothesis:
Factor TEACHER FACTOR
Method A B C
40 38 45
41 39 48
Method of
38 40 48
Teaching X
38 40 50
40 41 50
TOTAL
43 41 40
42 39 41
Method of
43 38 42
Teaching Y
46 38 42
50 40 45
TOTAL

Page 10 of 17 PM212 B (Research and Statistics)

Solution:

1. Problem:
(1) Is there a significant difference in the performance of the three groups of
MPM students under three different professors?
(2) Is there a significant difference in the performance of the three groups of
MPM students under two different methods of teaching?
(3) Is there an Interaction effect between teacher and method of teaching factors ?

2. Hypotheses
(1) Ho: There is no significant difference in the performance of the three groups
of MPM students under three different professors.

H1: There is a significant difference in the performance of the three groups of

MPM students under three different professors.

(2) Ho: There is no significant difference in the performance of the three groups
of MPM students under two different methods of teaching.

H1: There is a significant difference in the performance of the three groups of

MPM students under two different methods of teaching.

(3) Ho: Interaction effects are not present.

H1: Interaction effects are present.

3. Level of Significance

 = 0.05
df total = N -1 = 30 – 1 = 29
df within = k(n – 1) = 6(5 – 1) = 24
df column = c–1=3–1=2
df row = r–1=2–1=1
df column x row = (c – 1)(r – 1) = (3 – 1)(2 – 1) = (2)(1) = 2

Page 11 of 17 PM212 B (Research and Statistics)

 4. Statistics: F-Test (Two-Way ANOVA)

TEACHER FACTOR
Factor
Method A B C A2 B2 C2
Method 40 38 45 1,600 1,444 2,025
of 41 39 48 1,681 1,521 2,304
Teaching 38 40 48 1,444 1,600 2,304
X 38 40 50 1,444 1,600 2,500
40 41 50 1,600 1,681 2,500
TOTAL 197 198 241 636 7,769 7,846 11,633 27,248
Method 43 41 40 1,849 1,681 1,600
of 42 39 41 1,764 1,521 1,681
Teaching 43 38 42 1,849 1,444 1,764
Y 46 38 42 2,116 1,444 1,764
50 40 45 2,500 1,600 2,025
TOTAL 224 196 210 630 10,078 7,690 8,834 26,602
Grand 421 394 451 1,266 17,847 15,536 20,467 53,850
Total

𝐺𝑇 2 1,2662 1,602,756
(1) CF = = = = 53,425.20
𝑁 30 30

(2) SST = ∑ 𝐴2 + ∑ 𝐵2 + ∑ 𝐶 2 - CF

= 17,847 + 15,536 + 20,467 - 53,425.20

= 424.80

𝐴2𝑋 𝐴2𝑦 2
𝐵𝑋 𝐵𝑌2 2
𝐶𝑋 𝐶𝑌2
(3) SSw = (∑ 𝐴2 + ∑ 𝐵2 + ∑ 𝐶 2 ) − (∑ +∑ + ∑ +∑ +∑ +∑ )
𝑛 𝑛 𝑛 𝑛 𝑛 𝑛

(197)2 (224)2 (210)2

= 17,847 + 15,536 + 20,467 − ( + + ...+ )
5 5 5

= 53,850 - 53,757.20 = 92.80

(∑ 𝐴)2 + (∑ 𝐵)2 + (∑ 𝐶)2

(4) SSC = – CF
10

4212 + 3942 + 4512

= - 53,425.20 = 53,587.80 - 53,425.20 = 162.60
10

Page 12 of 17 PM212 B (Research and Statistics)

 (∑ 𝑋)2 + (∑ 𝑌)2
(5) SSr = – CF
15

= 53,426.40 - 53,425.20 = 1.20

(6) SSc r = SST - SSw – SSc - SSr = 424.80 - 92.80 –162.60 - 1.20 = 168.20

Two-Way ANOVA
Sources of df Sum of Mean F - Value Interpretation
Variation Squares Square Computed Tabular
(SS) (MS)
Teacher’s
162.60
Factor 2 81.30 21.024 3.40 Significant

Methods of Not
1 1.20 1.20 0.310 4.26
Teaching Significant
Factors*Met
hods 2 168.20 84.10 21.75 3.40 Significant
Interaction
Within
Group 24 92.80 3.867

Total 29 424.80

5. Decision Rule:

If the computed F value is greater than the F critical/tabular value, reject Null
Hypothesis, Ho.

6. Conclusion:

With the computed F value (column/teacher’s Factor) of 21.024 compared to the

F tabular value of 3.40 at 0.05 level of significance with 2 and 24 degrees of freedom,
the null hypothesis is rejected in favor of the alternative hypothesis which means that
There is a significant difference in the performance of the three groups of MPM students
under three different Professors. It implies that Professor C is better than professor A and
B.

With regards to the F value (row/Methods of Teaching) of 0.310 it is lesser than

the F tabular value of 4.26 at 0.05 level of significance with 1 and 24 degrees of
freedom. Hence the null hypothesis of no significant differences in the performance of
the three groups of MPM students under two different methods of teaching is confirmed.

However, the F value (interaction) of 21.75 is greater than the F tabular value of
3.40 at 0.05 level of significance with 2 and 24 degrees of freedom. Thus the research
hypothesis is confirmed which means that the interaction effect is present between the
professors and their methods of teaching . MPM Students under professor C have better

Page 13 of 17 PM212 B (Research and Statistics)

 performance under method of teaching X while MPM students under professor A have
better performance under method Y.

EXCEL Output
Anova: Two-Factor With Replication
SUMMARY A B C Total
Method X
Count 5 5 5 15
Sum 197 198 241 636
Average 39.4 39.6 48.2 42
Variance 1.8 1.3 4.2 20
Method Y
Count 5 5 5 15
Sum 224 196 210 630
Average 44.8 39.2 42 42
Variance 10.7 1.7 3.5 10
Total
Count 10 10 10
Sum 421 394 451
Average 42.1 39.4 45.1
Variance 13.6555556 1.377778 14.1

ANOVA
Source of Variation SS df MS F P-value F crit
Sample(Methods) 1.2 1 1.2 0.3 0.58263 4.25968
Columns
(Teachers) 162.6 2 81.3 21 0.00001 3.40283
Interaction(T/M) 168.2 2 84.1 22 0.00000 3.40283
Within 92.8 24 3.866667
Total 424.8 29

SPSS Output

Descriptive Statistics
Dependent Variable: data
Std.
Factor Methods Mean Deviation N
A X 39.4000 1.34164 5
Y 44.8000 3.27109 5
Total 42.1000 3.69534 10

Page 14 of 17 PM212 B (Research and Statistics)

 B X 39.6000 1.14018 5
Y 39.2000 1.30384 5
Total 39.4000 1.17379 10
C X 48.2000 2.04939 5
Y 42.0000 1.87083 5
Total 45.1000 3.75500 10
Total X 42.4000 4.48490 15
Y 42.0000 3.18479 15
Total 42.2000 3.82731 30

Tests of Between-Subjects Effects

Dependent Variable:data
Sum of Mean
Source Squares df Square F Sig.
Corrected
332.000a 5 66.400 17.172 .000
Model
Intercept 53425.200 1 53425.200 1.382E4 .000
Factor 162.600 2 81.300 21.026 .000
Methods 1.200 1 1.200 .310 .583
Factor *
168.200 2 84.100 21.750 .000
Methods
Error 92.800 24 3.867
Total 53850.000 30
Corrected Total 424.800 29
a. R Squared = .782 (Adjusted R Squared = .736)

Books:
1. Angeles, Ma. Felisa and et al., Simplified Approach To Statistics, 2005
2. Bluman, Allan G., Elementary Statistics Step-by-Step Approach, 2008
3. Berenson, Mark L. and et el., Business Statistics Concepts and Applications, Eight Edition,
2002

Links :

Page 15 of 17 PM212 B (Research and Statistics)

 1) https://www.statisticshowto.com/probability-and-statistics/hypothesis-
testing/anova/#OneWayANOVA2.
2) https://www.analyticsvidhya.com/blog/2018/01/anova-analysis-of-variance/
3) https://statistics.laerd.com/statistical-guides/one-way-anova-statistical-guide.php
4) https://www.scribbr.com/statistics/two-way-anova/

To have more understanding, click the link:

1. https://www.youtube.com/watch?v=hPG8IvKMXoc
2. . https://www.youtube.com/watch?v=iOoPHunv8NM
3. https://www.youtube.com/watch?v=0V5scynrVjY
4. https://www.youtube.com/watch?v=hPG8IvKMXoc
5. https://www.youtube.com/watch?v=uHEh1Ta0wHo

Page 16 of 17 PM212 B (Research and Statistics)

Page 17 of 17 PM212 B (Research and Statistics)