Steps to Solve One-Way ANOVA

1. State the Hypotheses

o Null Hypothesis (H₀): The means of all groups are equal.

o Alterna�ve Hypothesis (H₁): At least one group has a different mean.

2. Organize the Data

o Create a table with groups (categories) and their respec�ve observa�ons.

o Calculate the mean of each group (𝑋𝑋̄_group).

o Calculate the overall mean (𝑋𝑋̄_overall).

3. Compute the Between-Group Sum of Squares (SSB)

o Formula:

o Where:

 ni= number of samples in group i

 Xi = mean of group i

 Xoverall= overall mean

4. Compute the Within-Group Sum of Squares (SSW)

o Formula:

o For each observa�on, subtract the group mean and square the result.

o Sum up these squared differences for all groups.

5. Find the Degrees of Freedom (df)

o For Between-Groups (df_B): k−1(where k is the number of groups)

o For Within-Groups (df_W): N−k (where N is total observa�ons)

6. Compute Mean Squares (MS)

o Mean Square Between (MSB):

 o Mean Square Within (MSW):

7. Compute the F-Sta�s�c

MSB
o Formula: F=MSW

8. Compare with Cri�cal F-Value

o Use an F-table to find the cri�cal F-value based on df_B and df_W at a given
significance level (e.g., 0.05).

o If F > Cri�cal F-Value, reject H0 (significant difference exists).

o If F ≤ Cri�cal F-Value, fail to reject H0(no significant difference).

9. State the Conclusion

o If we reject H0→ At least one group mean is significantly different.

o If we fail to reject H0→ No sufficient evidence to claim a difference.

Ques 1. A teacher wants to compare the performance of three different teaching methods.
She selects 3 students per method and records their test scores.

Method A Method B Method C

4 3 6
5 4 7
3 5 5
At a 5% significance level (α = 0.05), determine whether there is a significant difference in
students’ performance across these teaching methods.

Step 1: Compute Group Means and Overall Mean

Method A Method B Method C Mean Calcula�on

4 3 6
5 4 7
3 5 5
(4+5+3)/3 = 4 (3+4+5)/3 = 4 (6+7+5)/3 = 6 Group Mean (𝑋𝑋̄)
 (4+4+6)/3 = 4.67 Overall Mean (𝑋𝑋̄overall)
Step 2: Compute SSB (Between-Group Sum of Squares)

Step 3: Compute SSW (Within-Group Sum of Squares)

Method A Mean (X - 4)² Method B Mean B (X - 4)² Method C Mean C (X - 6)²

A
4 4 (4-4)²=0 3 4 (3-4)²=1 6 6 (6-6)²=0
5 4 (5-4)²=1 4 4 (4-4)²=0 7 6 (7-6)²=1
3 4 (3-4)²=1 5 4 (5-4)²=1 5 6 (5-6)²=1
Total 2 2 2
SSW=2+2+2=6

Step 4: Compute Degrees of Freedom

• Degrees of freedom for SSB: dfB= k−1 = 3−1 = 2

• Degrees of freedom for SSW: dfW= N−k = 9−3 = 6

Step 5: Compute Mean Squares

Step 6: Compute the F-Sta�s�c

Step 7: Compare with Cri�cal F-Value

At α = 0.05, with df1 = 2 and df2 = 6, the F-table value is 5.14.

Since F = 4.005 < 5.14, we fail to reject the null hypothesis.

Ques 2. Suppose a researcher wants to compare the effec�veness of three different diets on
weight loss. Each diet is tested on 3 individuals, and the observed weight loss (in kg) is given
as follows:

Diet A Diet B Diet C

2 3 5
3 4 6
1 5 4
At a 5% significance level (α = 0.05), determine whether there is a significant difference in the
mean weight loss among the three diets.

Solu�on:
Since the F-sta�s�c is greater than the cri�cal value, we conclude that there is a significant
difference in the mean weight loss among the three diets.
 ANOVA through python code

Python Code for One-Way ANOVA using inbuilt func�ons with p-value

import numpy as np

import scipy.stats as stats

# Sample data: Three groups with different observa�ons

group_A = [5, 7, 9, 6, 8]

group_B = [10, 12, 14, 11, 13]

group_C = [15, 17, 19, 16, 18]

# Perform One-Way ANOVA

f_sta�s�c, p_value = stats.f_oneway(group_A, group_B, group_C)

# Display results

print(f"F-Sta�s�c: {f_sta�s�c:.4f}")

print(f"P-Value: {p_value:.4f}")

# Interpreta�on

alpha = 0.05 # Significance level

if p_value < alpha:

print("Reject the null hypothesis: At least one group mean is significantly different.")

else:

print("Fail to reject the null hypothesis: No significant difference among group means.")

What is the p-value in ANOVA?

The p-value in ANOVA tells us the probability of observing the given F-sta�s�c (or a more
extreme value) under Ho if the null hypothesis (H₀) is true.

How to Interpret the p-value in ANOVA?

• If p-value ≤ α (e.g., 0.05): Reject H₀ → At least one group mean is significantly different.
 • If p-value > α: Fail to reject H₀ → No significant difference among group means.

How is the p-value related to the F-sta�s�c?

• The F-sta�s�c is calculated as the ra�o of between-group variance to within-group

variance.

• The p-value is derived from the F-distribu�on and tells us whether the observed F-value
is large enough to reject H₀.

• A high F-value generally results in a low p-value, indica�ng strong evidence against H₀.

Example:

If you run ANOVA and get:

• F = 4.25, p = 0.012

o Since p < 0.05, we reject H₀, meaning at least one group has a significantly
different mean.

• F = 1.02, p = 0.40

o Since p > 0.05, we fail to reject H₀, meaning there is no strong evidence of a
difference among groups.

Python Code for One-Way ANOVA using inbuilt func�ons with p-value and F-value

import numpy as np

import scipy.stats as stats

# Step 1: Define sample data for three groups

group_A = [5, 7, 9, 6, 8]

group_B = [10, 12, 14, 11, 13]

group_C = [15, 17, 19, 16, 18]

# Step 2: Compute One-Way ANOVA

F_sta�s�c, p_value = stats.f_oneway(group_A, group_B, group_C)

# Step 3: Compute degrees of freedom

df_between = 3 - 1 # Number of groups - 1

df_within = (len(group_A) + len(group_B) + len(group_C)) - 3 # Total samples - Number of
groups

# Step 4: Get cri�cal F-value from F-table (5% significance level)

alpha = 0.05

cri�cal_F = stats.f.ppf(1 - alpha, df_between, df_within)

# Step 5: Display results

print(f"F-Sta�s�c: {F_sta�s�c:.4f}")

print(f"P-Value: {p_value:.4f}")

print(f"Cri�cal F-Value: {cri�cal_F:.4f}")

# Step 6: Decision Making

if F_sta�s�c > cri�cal_F:

print("Reject the null hypothesis: At least one group mean is significantly different.")

else:

print("Fail to reject the null hypothesis: No significant difference among group means.")

# Alterna�ve way using p-value

if p_value < alpha:

print("Conclusion using p-value: Reject H₀.")

else:

print("Conclusion using p-value: Fail to reject H₀.")