Large Sample Test

1. Introduction to Large Sample Tests

Meaning of Large Sample Test

A Large Sample Test is a statistical method used when the sample size (n) is greater than 30 (n>30). In such cases, the Central Limit Theorem (CLT) ensures that the sample distribution approximates a normal distribution, regardless of the original population distribution. Large sample tests are widely used in business, finance, healthcare, and social sciences for estimating population parameters and testing hypotheses with greater accuracy.

Importance of Large Sample Tests

✔ More accurate and reliable due to large sample representation.
✔ Reduces the impact of random errors compared to small sample tests.
✔ Used in marketing surveys, industrial quality control, medical trials, and economic research.

2. Assumptions of Large Sample Tests

1. Sample Size is Large → Typically n > 30.
2. Central Limit Theorem Applies → Even if the population is not normally distributed, the sampling distribution of the sample mean will be approximately normal.
3. Random Sampling → The sample must be randomly selected to ensure unbiased estimates.
4. Independent Observations → Data points should not influence each other.
5. Known or Estimable Variance → Population variance (σ²) should be known or estimable from the sample.

3. Types of Large Sample Tests

Large sample tests are primarily based on the Z-Test, which is used for:

✔ Testing population means when variance is known.
✔ Testing population proportions for categorical data.
✔ Comparing two sample means or proportions.

(i) Z-Test for Single Mean

Purpose

To test whether the sample mean is significantly different from the population mean.

Hypotheses

Test Statistic Formula

Where:

• X̄ = Sample mean.
• μ0 = Population mean.
• σ = Population standard deviation.
• n = Sample size.

Example

A company claims that the average salary of its employees is ₹50,000. A researcher takes a random sample of 100 employees and finds the sample mean salary to be ₹48,500 with a standard deviation of ₹5,000. Is the company’s claim valid at 5% significance level?

✔ Given:
n = 100, X̄ =48,500, μ0=50,000, σ=5,000, α=0.05

✔ Decision Rule:

• Critical Z-value at α=0.05 (two-tailed) = ±1.96.
• Since Z=−3 = -3 falls outside ±1.96, we reject H0​.

✔ Conclusion: The average salary is significantly lower than ₹50,000.

(ii) Z-Test for Difference of Means (Two Sample Test)

Purpose

To compare the means of two independent groups.

Hypotheses

H0 : μ1 = μ2 (No significant difference)

H1 : μ1 ≠ μ2(There is a significant difference)

Test Statistic Formula

Where:

• X̄1, X̄2​ = Sample means of two groups.
• σ1,σ2​ = Standard deviations of two populations.
• n1,n2​ = Sample sizes of two groups.

Example

Two factories produce the same product. The first factory has a sample mean production time of 30 minutes (n1 = 50, σ1 = 4), and the second factory has a sample mean of 28 minutes (n2 = 60, σ2=5). Is there a significant difference in production times at 5% significance level?

✔ Decision Rule:

• Critical Z-value at 5% (two-tailed) = ±1.96.
• Since Z=2.42 is greater than 1.96, we reject H0​.

✔ Conclusion: There is a significant difference in production times.

(iii) Z-Test for Proportion

Purpose

Used when testing population proportions (e.g., percentage of people preferring a brand).

Test Statistic Formula

Where:

• p^​ = Sample proportion.
• p0 = Population proportion.
• n = Sample size.

Example

A survey claims that 40% of customers prefer Brand A. A sample of 500 customers finds that 220 prefer Brand A. Is the survey accurate?

✔ Given:
n = 500, p0​ = 0.40, p^ = 220 / 500 = 0.44

✔ Decision Rule:

• Critical Z-value at α=0.05 (two-tailed) = ±1.96.
• Since Z=1.83 is within ±1.96, we fail to reject H0​.

✔ Conclusion: The survey claim is statistically valid.

4. Applications of Large Sample Tests

✔ Business & Marketing → Customer preference analysis, product pricing impact, and survey testing.
✔ Healthcare & Medicine → Drug efficacy comparison, patient recovery time analysis.
✔ Manufacturing & Quality Control → Production efficiency analysis, defect rate testing.
✔ Finance & Economics → Stock market return comparison, inflation rate assessment.

5. Conclusion

Large sample tests provide high accuracy and reliability in hypothesis testing when n>30n > 30n>30. The Z-Test is commonly used for testing means and proportions in diverse fields like business, healthcare, finance, and manufacturing. By carefully setting significance levels and interpreting results correctly, organizations can make data-driven, statistically sound decisions.