Experimental

Design

Dr. Intesar N.
El-Saeiti
Associate
MSc PROGRAMS IN STATISTICS Professor

Biostatistics (2626)

Dr. Intesar N. El-Saeiti

Associate Professor

University of Bengahzi-Libya

The 5nd Lecture (Analysis of Variance)

Joune 26,2024

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 1 / 18
 Analysis of Variance Experimental
Design

Dr. Intesar N.
One-Way ANOVA (”analysis of variance”) compares the El-Saeiti
Associate
means of two or more independent groups in order to Professor

determine whether there is statistical evidence that the

associated population means are significantly different.
One-Way ANOVA is a parametric test.
This test is also known as:
One-Factor ANOVA
One-Way Analysis of Variance
Between Subjects ANOVA
The variables used in this test are known as:
Dependent variable
Independent variable (also known as
the grouping, or factor) This variable divides cases into
two or more mutually exclusive levels, or groups
Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 2 / 18
 ANOVA Experimental
Design

Dr. Intesar N.
Statistical technique specially designed to test whether El-Saeiti
Associate
the means of more than two quantitative populations Professor

are equal.
Developed by Ronald A. Fisher in 1920s. ANOVA
instead of multiple t-tests
If you are comparing means between more than two
groups, why not just do several two sample t-tests to
compare the mean from one group with the mean from
each of the other groups?
The problem with the multiple t-tests approach is that
as the number of groups increases, the number of two
sample t-tests also increases.
As the number of tests increases the probability of
making a Type I error also increases.
Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 3 / 18
 Data Requirements Experimental
Design

Dr. Intesar N.
Your data must meet the following requirements: El-Saeiti
Associate
Professor
1 Dependent variable that is continuous (i.e., interval or
ratio level)
2 Independent variable that is categorical (i.e., two or
more groups) Cases that have values on both the
dependent and independent variables Independent
samples/groups (i.e., independence of observations)
There is no relationship between the subjects in each
sample. This means that: subjects in the first group
cannot also be in the second group no subject in either
group can influence subjects in the other group no
group can influence the other group Random sample of
data from the population
Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 4 / 18
 Assumptions for ANOVA Experimental
Design

Dr. Intesar N.
El-Saeiti
1 The populations from which the samples are obtained Associate
Professor
must be normally or approximately normally distributed
(Normality).
2 The samples must be independent of each other
(Independence).
3 The variances of the populations must be equal
(Homogeneity of variances).
4 The null hypothesis is that the means are all equal

H0 : µ1 = µ2 = ... = µk

5 The alternative hypothesis is that at least one of the

means is different.
H1 :at least one of them is different.
Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 5 / 18
 Test Statistic Experimental
Design

Dr. Intesar N.
The test statistic for a One-Way ANOVA is denoted as F. El-Saeiti
Associate
For an independent variable with k groups, the F statistic Professor

evaluates whether the group means are significantly

different.
Basic idea is to partition total variation of the data into
two sources
1 Variation between groups.
2 Variation within groups.
If H0 is true the standardized variances are equal to one
another

SST = SSB + SSW

Total Sum = Sum of Squares + Sum of Squares
of Squares between groups within groups
Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 6 / 18
 ANOVA Table Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Source d.f Sum of Mean Sum F-test

of Variation Squares of Squares
SSB
Between k −1 SSB MSB=
K −1
MSB
Groups Fcal =
MSW
SSW
Within N −k SSW MSW=
N −K
Groups
Total N −1 SST

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 7 / 18
 Interpret the results of ANOVA Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor
To test whether or not there is a significant difference
between the means, the p-value is compared to the
significance level (α).

H0 : µ1 = µ2 = ... = µk

H1 : at least one of them is different.

P-value ≤ α : H0 is rejected and H1 is accepted indicating
that at least one of the means is different.
P-value > α : H0 is not rejected indicating that all the
means are equal.

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 8 / 18
 Post-hoc tests Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Analysis of variance answers global questions:

Are there any differences between the means?
More specific questions:
Differ certain pairs or groups of mean values?
Suggestion:
continue with post-hoc tests only, if the p-value of the
ANOVA < 0.05

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 9 / 18
 Analyze > Compare Means > One-Way ANOVA Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 10 / 18
 Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 11 / 18
 Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 12 / 18
 Spss steps Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 13 / 18
 Example Experimental
Design

Dr. Intesar N.
El-Saeiti
In the sample dataset, the variable Sprint is the respondent’s Associate
Professor
time (in seconds) to sprint a given distance, and Smoking is
an indicator about whether or not the respondent smokes (0
= Nonsmoker, 1 = Past smoker, 2 = Current smoker). Let’s
use ANOVA to test if there is a statistically significant
difference in sprint time with respect to smoking status.
Sprint time will serve as the dependent variable, and
smoking status will act as the independent variable.
Before the Test
Just like we did with the paired t test and the independent
samples t test, we’ll want to look at descriptive statistics and
graphs to get picture of the data before we run any
inferential statistics.

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 14 / 18
 Example Experimental
Design

Dr. Intesar N.
El-Saeiti
The sprint times are a continuous measure of time to sprint Associate
Professor
a given distance in seconds. From the Descriptives procedure
(Analyze> Descriptive Statistics > Descriptives), we see
that the times exhibit a range of 4.5 to 9.6 seconds, with a
mean of 6.6 seconds (based on n=374 valid cases). From
the Compare Means procedure (Analyze > Compare
Means > Means), we see these statistics with respect to
the groups of interest:
N Mean Std.Deviation
Nonsmoker 261 6.411 1.252
Past smoker 33 6.835 1.024
Current smoker 59 7.121 1.084
Total 353 6.569 1.234

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 15 / 18
 Running the Procedure Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

1 ClickAnalyze > Compare Means > One-Way

ANOVA.
2 Add the variable Sprint to the Dependent List box, and
add the variable Smoking to the Factor box.
3 Click Options. Check the box for Means plot, then click
Continue.
4 Click OK when finished.

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 16 / 18
 Spss output Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Sum of Squares df Mean Squares F Sig.

Between 26.788 2 13.394 9.209 .000

Groups
Within 509.082 350 1.455
Groups
Total 535.870 352

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 17 / 18
 Discussion and Conclusions Experimental
Design

Dr. Intesar N.
El-Saeiti
Associate
Professor

Conclusions

We conclude that the mean sprint time is significantly

different for at least one of the smoking groups
(F2,350 = 9.209, p < 0.001).
Note that the ANOVA alone does not tell us specifically
which means were different from one another.
To determine that, we would need to follow up with
multiple comparisons (or post-hoc) test

Dr. Intesar N. El-Saeiti Associate Professor (UOB) Experimental Design Joune 26,2024 18 / 18