Basics of ANOVA
Principles and Practical Applications
By-
Dr. Sanjay Kumar Gupta
Rungta College of Pharmaceutical Sciences and Research
Bhilai (CG)
Outline
Definitions and Terminologies
Introduction
Hypothesis and Type of Errors
What is ANOVA?
Basic Principles of ANOVA
Assumptions in ANOVA
ANOVA Techniques
One-Way ANOVA
Two-Way ANOVA
Practical Applications
2
Definitions and Terminologies
Measures of Central Tendencies (Mean, Mode and Median)
Population Mean and Sample Mean
Variance (High, Low, Within Group, Between Group)
Standard Deviation (SD) and Standard Error of Mean (SEM)
Variables (Dependent and Independent)
Null hypothesis and Alternate Hypothesis
z-test and t-test
Type-I and Type II error
ANOVA
3
Introduction
ANOVA is used to compare means among 3 or more groups to determine
whether significant differences between them.
Itdetermine whether the observed differences among sample means are due
to real differences or just random chance.
The significance of the difference in means of the two samples can be judged
through either z-test or t-test.
Z-test is applied to find out the degree of reliability of statistics in case of a
large sample.
Z-test is based on the normal probability distribution and is used for judging
the significance of several statistical measures, particularly the mean.
4
t-test
is used to test the null hypothesis that the population means of two
groups are same.
t-test
commonly used with small sample sizes, testing the difference between
samples when the variances of two normal distributions are not known.
When there are more than two means, it is possible to compare each mean
with each other using t-tests.
However, multiple t-tests can lead to increase Type I error rate.
ANOVA is used to test differences among several means for significance
without increasing the Type I error rate.
5
Null Hypothesis (H₀):
It represents the default assumption that there is no effect, no difference,
or no relationship between variables.
It assumes that any observed differences are due to random chance.
H₀: µ₁ = µ₂ = µ₃ = ... = µₖ
Alternate Hypothesis (H1):
The alternative hypothesis represents the researcher’s claim or prediction.
It asserts a specific effect, difference, or relationship between variables.
6
In statistical hypothesis testing, a Type I error occurs when
you reject the null hypothesis (H₀) when it is actually true.
In other words, it’s a false positive—you conclude that there is
an effect or difference when there isn’t one in reality.
7
An example to illustrate Type I error:
Suppose a pharma company tests a new drug to reduce BP.
• Null Hypothesis (H₀): The new drug does not affect BP (i.e., it’s equivalent to a placebo).
• Alternative Hypothesis (H₁): The new drug reduces BP significantly.
After study, statistical test shows a significant difference in BP between the drug-treated
and the placebo group. The researchers reject the H0 and conclude that the drug is
effective.
However, if H0 is true (i.e. the drug has no effect), this rejection would be a Type I error.
In other words, they falsely conclude that the drug works when it doesn’t.
In summary, a Type I error is when you mistakenly reject a true H0.
It’s essential to control the risk of Type I errors by choosing an appropriate significance
level and interpreting the results cautiously.
8
What is ANOVA?
ANOVA is a procedure for testing the difference among
different groups of data for homogeneity.
The essence of ANOVA is that the total amount of variations in
a set of data is broken down into two types:
The amount that can attributed to chance, and
The amount that can be attributed to specific causes.
9
Basic Principle of ANOVA
ANOVA tests for differences among the means of the populations by
examining the amount of variation within each of these samples,
relative to the amount of variation between the samples.
10
Assumptions in ANOVA
The experimental errors of data are normally distributed.
Equal variance between treatments, i.e. Homogeneity of variances.
Independence of samples i.e. Each sample is randomly selected and
independent.
11
Types of ANOVA
One-way ANOVA: Tests the impact of one independent variable (factor) on
a dependent variable, comparing the means of more than two groups.
Two-way ANOVA: Examines the influence of two independent variables
and their interaction on a dependent variable.
Multifactor ANOVA: Extends beyond two-way ANOVA, allowing for the
study of multiple factors.
Repeated Measures ANOVA: Used when the same subjects are measured
multiple times under different conditions.
12
The ANOVA Table
• Source of Variation: Divides the total variation in the data into two parts:
• Between-group variation: Due to the differences between the group means.
• Within-group variation: Due to the variability within each group (random error).
Degrees of Freedom (df):
Between-group: dfb=k−1 (where k is the number of groups)
Within-group: dfw=N−k (where N is the total number of observations)
Mean Squares:
Mean Square Between (MSB)
Mean Square Within (MSW)
F-statistic:
F = MSB / MSW
Used to test the significance of the overall ANOVA model.
P-value: If the p-value is less than the significance level (typically 0.05), we reject the null
13
hypothesis and conclude that there is a significant difference between the group means.
One-way ANOVA
Example: A researcher is interested in testing whether the average test scores of students
differ between three different teaching methods.
Steps:
1. State the Hypotheses
H₀: The mean scores for all teaching methods are equal
H₁: At least one teaching method has a different mean score.
2. Calculate SSB, SSW, MSB, MSW, and F-statistic
3. Compare the F-statistic to the critical F-value or use the p-value.
14
15
16
17
Two way ANOVA
It is an extension of one-way ANOVA.
There are two independent variables, and hence named Two way ANOVA.
18
19
20
24
25
ANOVA Limitations
Sensitive to Assumptions: Violations of the normality or homogeneity
of variance assumptions can lead to misleading results.
Doesn’t Indicate Which Groups Differ: While ANOVA tells us if there’s a
difference, it doesn’t specify where the difference lies; post-hoc tests
are required for that.
Limited to Means Comparison: ANOVA only tests for differences in
means and doesn’t provide information about other aspects of the
distributions.
29
Post-Hoc Tests
When the ANOVA test is significant, post-hoc tests are used to identify which
specific groups differ from each other.
Common Post-Hoc Tests:
Tukey’s HSD (Honestly Significant Difference): Used for pairwise
comparisons while controlling for type I errors.
Bonferroni Correction: Adjusts the significance level when conducting
multiple comparisons.
Scheffé Test: More conservative but flexible, allowing for the comparison of
complex group contrasts.
30
Practical Applications of ANOVA
Clinical
Trials/ preclinical studies: Testing the effectiveness of different
treatments on patients/ animals/ subjects.
Agriculture: Comparing crop yields under different fertilizers.
Marketing: Analyzing customer preferences based on different advertising
strategies.
Education: Comparing student performance across different teaching
methods or curricula.
31
Effect on Colon weight/length ratio
140 ###
Colon weight/length ratio (mg/cm)
**
120
***
***,$$
100
80
60
Normal Control SS HPFE-200 HPFE-400
Treatment
Figure 01. Effect on Colon weight/length ratio of different treated groups. Data are analysed using
One Way Anova followed by Tuckey’s Multiple Comparison Test.
###
P<0.001 Vs Normal; **P<0.01 and ***P<0.001 Vs Control; $$P<0.01 Vs HPFE200.
*** *** *** *
100 ***,#
***
*
*** ***
***
***,# *** ***
90 *
*** ***
***,#
80
***
***
***,#
70
***
***,#
60 ***
***
Negative Control
50 *** Positive Control
***
40 *** **
Silveron Gel
30 *** Combined hydrogel
20 *
*
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Days
Figure 02. Effect of different hydrogels on day wise % wound contraction in rats. Data are
analysed using Two-way ANOVA followed by Tuckey’s Multiple Comparison Test.
*
P<0.05, **P<0.01 and ***P<0.001 Vs Negative Control. #P<0.05 Vs Silveron Gel
Thank You
34
