🧪 ANOVA (F-Test)

Last Updated: Jan 2026

ANOVA (Analysis of Variance) is a statistical test used to determine whether there is a significant difference between the means of three or more groups.

Instead of comparing means one by one, ANOVA compares variance between groups vs variance within groups.

🗣 Hinglish Tip: ANOVA = 3 ya zyada groups ke mean ek saath compare karna

Why ANOVA is Needed?

If we compare:

Group A vs B
Group B vs C
Group A vs C

using multiple t-tests ✘ → error probability badh jaati hai

ANOVA solves this using one single test $\checkmark$

When to Use ANOVA?

Use ANOVA when:

Comparing 3 or more groups
Data is numerical
Samples are independent
Data is approximately normal
Variances are roughly equal

Types of ANOVA

One-Way ANOVA → One factor (most common)
Two-Way ANOVA → Two factors
Repeated Measures ANOVA

👉 In this tutorial, we cover One-Way ANOVA

ANOVA Notation (Math Standard)

Group means → $\bar{x}_1,\,\bar{x}_2,\,\bar{x}_3$
Overall mean → $\bar{x}$
Number of groups → $k$
Total observations → $N$
F statistic → $F$

ANOVA Core Idea

 $F = Variance Between Groups / Variance Within Groups$

If:

F is large → group means differ significantly
F is small → group means are similar

Example

Three teaching methods are used, and student's scores are recorded.

Group	Scores
A	50, 55, 60
B	65, 70, 75
C	80, 85, 90

Test whether the mean scores differ significantly at 5% significance level.

Step 1: State the Hypotheses

 $H_0: \mu_1 = \mu_2 = \mu_3 \quad \text{(All group means are equal)} \\[0.5em] H_1: \text{At least one group mean is different}$

Step 2: Calculate Group Means

Group	Scores	Group Mean $\bar{x}_i$
A	50, 55, 60	55
B	65, 70, 75	70
C	80, 85, 90	85

Step 3: Calculate Overall Mean

 $\bar{x} = \frac{50+55+60+65+70+75+80+85+90}{9} \\[0.5em] \bar{x} = {630}/{9} = 70$

Step 4: Calculate Sum of Squares Between Groups (SSB)

Formula:

 $SSB = \sum n(\bar{x}_i - \bar{x})^2$

Group	$\bar{x}_i$	$(x - \bar{x}_i)^2$	$n(\bar{x}_i - \bar{x})^2$
A	55	225	675
B	70	0	0
C	85	225	675

 $SSB = 1350$

Step 5: Calculate Sum of Squares Within Groups (SSW)

Formula:

 $SSW = \sum (x - \bar{x}_i)^2$

Group	SSW
A	50
B	50
C	50

SSW = 150

Step 6: Degrees of Freedom

 $df_{\text{between}} = k - 1 = 3 - 1 = 2 \\[0.5em] df_{\text{within}} = n - k = 9 - 3 = 6$

Step 7: Mean Squares

 $MSB = \frac{SSB}{df_{\text{between}}} = \frac{1350}{2} = 675 \\[0.5em] MSW = \frac{SSW}{df_{\text{within}}} = \frac{150}{6} = 25$

Step 8: Calculate F-Statistic

 $F = MSB / MSW \\[0.5em] F = 675 / 25 \\[0.5em] F = 27$

Step 9: Critical Value

At:

- $\alpha = 0.05$ - $df_1 = 2$ - $df_2 = 6$

From F-table:

 $F_{0.05,(2,6)} \approx 5.14$

Step 10: Decision

Calculated F = 27
Critical F = 5.14

Since:

 $27 > 5.14$

👉 Reject $H_0$

Step 11: Conclusion

There is significant evidence that at least one teaching method has a different mean score.

🗣 Hinglish Tip: F value bahut bada → groups alag-alag behave kar rahe hain

Report An Issue

← Previous Next →