HYPOTHESIS-TESTING-2
1. Hypothesis Testing Overview
Definition: Hypothesis testing is a statistical method that determines if a claim about a population parameter is valid based on sample data.
Purpose: To assess the validity of our beliefs or assumptions about a population.
2. Methods for Hypothesis Testing
2.1 Traditional Method
2.2 P-value Method
2.3 Confidence Interval Method
3. Formulation of Hypotheses
3.1 Null Hypothesis (H0)
Represents the initial assumption; states no significant effect or difference exists between variables.
Example: H0: The new drug has no effect on blood pressure.
3.2 Alternative Hypothesis (HA or H1)
Asserts that a difference or effect does exist.
Example: H1: The new drug reduces blood pressure.
4. Significance Level (α)
Alpha (α): A threshold set to determine how strong evidence is needed before rejecting the null hypothesis.
5. Steps in Hypothesis Testing (Traditional Method)
State Hypotheses: Identify H0 and H1.
Find Critical Values: Determine cutoff points based on α.
Compute Test Statistic: Measure evidence against H0.
Decision: Reject or not reject H0 based on comparison of test statistic to critical values.
Summarize Results: Provide findings.
6. Z-test for a Mean (Traditional Method)
6.1 When to Use
Appropriate for large sample sizes (n ≥ 30) and when population variance is known.
6.2 Z-test Formula
Compute test statistic using the formula:Z = (X̄ - μ) / (σ / √n)Where:
X̄ = sample mean
μ = population mean
σ = population standard deviation
n = sample size
6.3 Example Problem
Claim: Mean age of doctors in a hospital system is greater than 46 years.
Decision: Reject H0 if Z > 1.645 at α=0.05.
7. Z-test for a Proportion (Traditional Method)
7.1 Requirements
Assumptions include a random sample and binomial experiment conditions.
7.2 Example Problem
Claim: 17% of young people are obese; conduct test to validate.
8. T-test for a Mean (Traditional Method)
8.1 When to Use
Used for small sample sizes (n < 30) and unknown population standard deviation.
8.2 Example Problem
Claim: Average number of infections per week in a hospital is 16.3.
9. Chi-Square Test for Variance (Traditional Method)
9.1 When to Use
Tests claims about variance or standard deviation.
9.2 Example Problem
Claim: Variance in IQ of patients differs from a known population variance.
10. P-value Method in Hypothesis Testing
10.1 Definition
P-value: Probability of obtaining a sample statistic at least as extreme as the one observed, given that H0 is true.
10.2 Decision Rule
If P-value ≤ α, reject H0.
If P-value > α, do not reject H0.
10.3 Example Problem
Claim regarding tuition costs at a college.
11. Confidence Intervals Method
11.1 Relationship to Hypothesis Testing
CI provides a range of values likely to include the population parameter based on sample data.
If the CI does not contain the hypothesized mean, H0 can be rejected.
11.2 Example Problem
Claim about the weight of sugar bags; find the CI to assess validity.