Small Sample test, (t, F, Z Test and Chi Square Test)
1. Introduction to Small Sample Tests
Meaning of Small Sample Test
Small sample tests are statistical techniques used when the sample size is less than 30 (n<30). Since small samples are more likely to deviate from normality, special tests like the t-Test, F-Test, Z-Test, and Chi-Square Test are used to assess population characteristics and relationships.
Importance of Small Sample Tests
✔ Used when data collection is expensive, time-consuming, or limited.
✔ Helps in medical trials, market research, quality control, and social sciences.
✔ Provides statistically valid inferences even with fewer data points.
2. Types of Small Sample Tests
| Test | Used For | Example |
|---|---|---|
| t-Test | Comparing two means when population variance is unknown | Checking if a new drug lowers blood pressure compared to an existing drug. |
| F-Test | Comparing two variances | Testing if sales revenue variance is higher in two different cities. |
| Z-Test | Comparing means or proportions when variance is known | Checking if students’ average test score differs from 70%. |
| Chi-Square Test | Testing association between categorical variables | Checking if gender influences buying preference for a product. |
3. t-Test (Student's t-Test)
(i) Meaning of t-Test
A t-Test is used when:
✔ The sample size is small (n<30).
✔ The population standard deviation is unknown.
✔ Data follows a normal distribution (or approximately normal).
(ii) Types of t-Test
1. One-Sample t-Test
- Used when comparing the sample mean with a known population mean.
where:
- X̄ = Sample mean.
- μ = Population mean.
- s = Sample standard deviation.
- n = Sample size.
✔ Example: A nutritionist tests if the average calorie intake of students is 2000 kcal based on a sample of 25 students with X̄ =1950 , s = 150.
2. Two-Sample t-Test
- Used when comparing the means of two independent groups.
✔ Example: A company tests if male and female employees have different job satisfaction levels.
3. Paired t-Test
- Used for before-and-after studies where measurements are taken from the same subjects.
✔ Example: Measuring the effectiveness of an exercise program by comparing weight before and after training.
4. F-Test (Variance Ratio Test)
(i) Meaning of F-Test
An F-Test compares the variances of two populations to determine if they are significantly different.
(ii) Test Statistic Formula
where:
(iii) Example of F-Test
A company compares sales revenue variance in two different stores. If the variances differ significantly, they may need different pricing strategies.
✔ Decision Rule: If Fcalculated > Fcritical from the F-table, reject H0.
✔ Application: Used in ANOVA (Analysis of Variance) to compare multiple means.
5. Z-Test for Small Samples
(i) When to Use Z-Test for Small Samples?
- When sample size is small (n < 30) but population variance is known.
- Used when data is normally distributed.
✔ Example: A researcher checks if the average IQ score of a group is 100 based on a sample of 20 students, assuming known variance.
✔ Formula:
6. Chi-Square Test (χ2 Test)
(i) Meaning of Chi-Square Test
The Chi-Square Test measures the association between two categorical variables. It helps determine whether the distribution of observed frequencies differs from expected frequencies.
(ii) Formula for Chi-Square Statistic
where:
- O = Observed frequency.
- E = Expected frequency.
(iii) Types of Chi-Square Test
1. Chi-Square Test for Independence
Used when testing whether two categorical variables are related.
✔ Example: A supermarket tests if customer gender influences product preference (Male vs. Female vs. Product Type).
2. Chi-Square Goodness-of-Fit Test
Used when testing whether an observed distribution fits an expected distribution.
✔ Example: A company checks if the sales of different product sizes (Small, Medium, Large) match the predicted sales ratio of 20%, 50%, and 30% respectively.
✔ Formula: Same as Chi-Square formula above.
7. Comparison of Small Sample Tests
| Test | Purpose | Example |
|---|---|---|
| t-Test | Compares means of one or two groups | Testing if a new teaching method improves test scores |
| F-Test | Compares variances of two groups | Checking if two factories have different product defect variability |
| Z-Test | Compares means when variance is known | Testing if machine-produced screws have an average length of 10 cm |
| Chi-Square Test | Compares categorical variables | Checking if customer satisfaction varies across age groups |
8. Applications of Small Sample Tests
(i) Business & Marketing
✔ t-Test: Evaluating customer satisfaction before and after launching a new product.
✔ Chi-Square Test: Analyzing whether product preferences differ across age groups.
(ii) Healthcare & Medicine
✔ t-Test: Comparing the effectiveness of two medical treatments.
✔ F-Test: Checking variance in patient recovery times between hospitals.
(iii) Finance & Economics
✔ t-Test: Determining if average salaries differ between two departments.
✔ Chi-Square Test: Analyzing customer default rates across different loan types.
9. Conclusion
Small sample tests are essential when working with limited data, ensuring statistical accuracy and valid conclusions. The t-Test is used for comparing means, the F-Test compares variances, the Z-Test compares means when variance is known, and the Chi-Square Test checks associations between categorical variables.