Hypothesis testing is a statistical method used to make decisions or inferences about a

population based on sample data. It involves testing an assumption (hypothesis) about a

population parameter.

Key Steps in Hypothesis Testing:

1. Formulate the Hypotheses:

o Null Hypothesis (H0): Assumes no effect or no difference.
o Alternative Hypothesis (H1): Assumes an effect or difference exists.
2. Select a Significance Level (α): Usually 0.05, indicating a 5% risk of rejecting the
null hypothesis when it is true.
3. Choose a Test and Compute the Test Statistic: Depending on the data and
hypothesis, use a specific test (e.g., Z-test or T-test).
4. Make a Decision: Compare the test statistic to a critical value or p-value to decide
whether to reject H0

Important Parametric Tests

Parametric tests are statistical tests that rely on assumptions about the data's distribution,
typically assuming that the data follows a normal distribution. These tests are powerful and
widely used when their assumptions are met.

1. Z-Test

 Purpose: Tests the difference between means when the sample size is large, and population
variance is known.
 Applications:
o One-sample Z-test: Compare the sample mean to a known population mean.
o Two-sample Z-test: Compare the means of two independent groups.
o Z-test for proportions: Compare population proportions.
 Example: Testing if the average height of a group of students differs from the national
average when the population variance is known.

2. T-Test

 Purpose: Tests the difference between means when the sample size is small, and the
population variance is unknown.
 Types:
1. One-Sample T-Test: Compare a sample mean to a known population mean.
2. Independent T-Test: Compare means of two independent groups.
3. Paired T-Test: Compare means of the same group at two different times.
 Example: Testing if a new drug improves patient recovery times compared to an existing
drug.
3. Chi-Square Test

 Purpose: Tests the association between categorical variables or checks if observed

frequencies match expected frequencies.
 Applications:
o Chi-square test of independence: Tests if two categorical variables are related.
o Chi-square goodness-of-fit test: Tests if observed data matches an expected
distribution.
 Example: Testing if gender is related to choice of study program.

4.F-Test

 Purpose: Tests if the variances of two populations are equal.

 Applications:
o Variance comparison in two groups.
o Precursor to ANOVA and regression analysis.
 Example: Testing if the variability in scores differs between two classrooms.

Z-Test

The Z-test is a statistical test used to determine if there is a significant difference between
sample data and the population, or between two samples, when:

 The sample size is large (n>30).

 The population standard deviation (σ) is known.

Steps in Z-Test:

1. Formulate Hypotheses:
o Null Hypothesis (H0H_0H0): Assumes no difference or effect.
o Alternative Hypothesis (H1H_1H1): Assumes a difference or effect exists.
2. Calculate the Z-Statistic:

The formula for a z-test in hypothesis testing is

z=(x̄ −μ)/(σ/√n)
 where:

 x̄ : is the sample mean

 μ: is the hypothesized population mean
 σ: is the known population standard deviation
 n: is the sample size

3. Compare to Critical Value or p-Value:

o For α=0.05\alpha = 0.05α=0.05, the critical Z-value is ±1.96 (two-tailed test).

Activity for Z-Test:

Objective: Determine if the average height of students in the class differs significantly from
the national average (e.g., 5.5 feet).

Steps:

1. Measure the height of 40 students in the class (large sample size).

2. Calculate the sample mean (Xˉ) and compare it to the national average (μ=5.5).
3. Assume the population standard deviation (σ=0.3).
4. Use the Z-formula to calculate the Z-statistic.
5. Compare the calculated Z to the critical Z-value (±1.96±) to make a conclusion.

T-Test
The T-test is used when:

 The sample size is small (n≤30).

 The population standard deviation is unknown.

Types of T-Tests:

1. One-Sample T-Test: Compares the sample mean to a known population mean.

2. Independent T-Test: Compares the means of two independent groups.
3. Paired T-Test: Compares means of the same group at two different times.

Activity for T-Test:

Objective: Evaluate if a new teaching method improves student scores.

Steps:
 1. Take a small group of students (e.g., n=15) and measure their test scores before and after
using the new teaching method.
2. Calculate the mean scores for "before" and "after."
3. Compute the standard deviation of the differences between the two sets of scores.
4. Use the paired T-test formula to calculate the T-statistic:

T = ¯d /sd/√n.

5. Compare the T-statistic to the critical T-value

Scenario:
Hypothesis: A company claims that its battery lasts an average of 100 hours. You test 50
batteries and find an average lifespan of 98 hours, with a known population standard
deviation of 4 hours.

Soltion : You perform a Z-test to determine if the average battery life is significantly
less than claimed.

Hypothesis: A teacher wants to know if a new teaching method improves student scores.
She tests 10 students before and after using the method.

Solution:A paired T-test is used to determine if the improvement is significant.

Hypothesis Testing of Means
1. Population Normal, Infinite, Variance Known (Z-Test)

 Use Case: Large or small sample size, hypothesis can be one-sided or two-sided.
 Example: Testing if the average height of students differs from the national average,
given the population variance.

2. Population Normal, Finite, Variance Known (Z-Test with Finite Population

Multiplier)

 Use Case: Large or small sample size, hypothesis can be one-sided or two-sided.
 Formula:
 Example: Testing if the mean score of a limited group of students (e.g., from a single
class) differs from the expected average.

3. Population Normal, Infinite, Small Sample, Variance Unknown (T-Test)

 Use Case: Small sample size (n<30), hypothesis can be one-sided or two-sided.
 Example: Testing if a new teaching method improves scores in a small class.

4. Population Normal, Finite, Small Sample, Variance Unknown (T-Test with Finite
Population Multiplier)

 Use Case: Small sample size, finite population, hypothesis can be one-sided or two-
sided.
 Formula:

t=Xˉ−μ0s/n×N−nN−1t = \frac{\bar{X} - \mu_0}{s / \sqrt{n}} \times \sqrt{\frac{N - n}{N -

1}}t=s/nXˉ−μ0×N−1N−n

 Example: Comparing the test scores of a small, finite group (e.g., 20 students in a
class of 25).

5. Non-Normal Population, Large Sample (Z-Test Approximation)

 Use Case: Large sample size (n>30n > 30n>30), variance may be known or unknown,
hypothesis can be one-sided or two-sided.
 Formula:

z=Xˉ−μ0σ/nz = \frac{\bar{X} - \mu_0}{\sigma / \sqrt{n}}z=σ/nXˉ−μ0

 or, if variance is unknown:

z=Xˉ−μ0s/nz = \frac{\bar{X} - \mu_0}{s / \sqrt{n}}z=s/nXˉ−μ0

 Example: Testing if the mean weight of a large group of individuals (not normally
distributed) is equal to a specified value.

Notes:

1. One-Sided vs. Two-Sided Hypothesis:

o One-sided: Tests if the mean is greater or less than μ0\mu_0μ0.
o Two-sided: Tests if the mean is different from μ0\mu_0μ0.

2. Finite Population Multiplier: Used when the population is finite to adjust for
sampling proportion.
3. Degrees of Freedom (d.f.): For T-tests, d.f.=n−1d.f. = n - 1d.f.=n−1.

Example : Hypothesis Testing for Means

Scenario:

A factory produces light bulbs, and the average lifespan of the bulbs is claimed to be 800
hours. A quality control team selects a random sample of 25 bulbs and finds the sample
mean lifespan to be 790 hours with a sample standard deviation of 20 hours. The team
wants to test if the actual mean lifespan is less than the claimed 800 hours, at a significance
level of 5% (α=0.05).

Steps:

1. Define the Hypotheses:

o Null Hypothesis (H0): μ=800 (The mean lifespan is 800 hours).
o Alternative Hypothesis (Ha): μ<800 (The mean lifespan is less than 800 hours).
 2. Determine the Test to Use:

 Population Variance Unknown: Use a T-Test since the sample size (n=25) is small
(<30), and the population variance is not provided.

 Test Statistic Formula:

T = (X‾ - μ0) / (s / √n),

where X‾ is the sample mean, (790)
μ0 represents the population mean,(800)
s is the standard deviation of the sample(20)
n stands for the size of the sample.(25)

3.Calculate the Test Statistic

5. Determine the Degrees of Freedom and Critical Value:

o Degrees of freedom (d.f.d.f.d.f.): n−1=25−1=24
o For a one-tailed test at α=0.05, the critical t-value from the t-table is
approximately -1.711.

6. Decision Rule:

 If the calculated t-value is less than the critical t-value (−2.5<−1.711), reject H0.

7. Conclusion:
Since −2.5<−1.711, the null hypothesis is rejected. There is evidence to suggest that the mean
lifespan of the light bulbs is less than 800 hours.

A t-table (also called the Student's t-distribution table) is a statistical table used to determine
the critical t-values for hypothesis testing when the sample size is small and/or the population
standard deviation is unknown

The t-table provides the critical values of the t-statistic that corresponds to specific degrees
of freedom (d.f.) and a chosen significance level (α\alphaα). These values help determine
whether the calculated t-statistic is significant or not when comparing the observed sample
statistic to the population parameter.

One-Tailed Test and Two-Tailed Test are concepts in hypothesis testing, used to determine the
significance of statistical results.
 A one-tailed test is used when the research hypothesis predicts the direction of the effect or
relationship.

 It tests whether a sample statistic is greater than or less than a hypothesized value.

Example:

A company claims that a new drug increases sleep duration. A one-tailed test would evaluate

Two-Tailed Test

 A two-tailed test is used when the research hypothesis does not predict the direction of the
effect, only that there is an effect.

A company claims a new teaching method changes student performance. A two-tailed test
would evaluate:

 One-Tailed Test: Used when you are testing for an effect in one specific direction (less
or greater).
 Two-Tailed Test: Used when you are testing for a difference in either direction.
 Left-Tailed Test: A specific case of a one-tailed test where you test for less than.
 Right-Tailed Test: A specific case of a one-tailed test where you test for greater than.