Hypothesis Testing: Concepts, Procedures, and Applications

Hypothesis Testing

Introduction to Hypothesis Testing

Hypothesis testing is a fundamental statistical procedure used to evaluate claims about population parameters, such as means or proportions. It involves formulating two competing hypotheses—the null and the alternative—and using sample data to determine which hypothesis is better supported.

• Null Hypothesis (H0): A statement that the population parameter equals a specific value.

• Alternative Hypothesis (Ha): A statement that the population parameter differs from the value stated in H0.

• Hypothesis Test: A procedure to test a claim about a property of the population.

Example: The mean weight of adult males in the US is less than 179 lbs. (H0: μ = 179, Ha: μ < 179)

Steps in Hypothesis Testing

The process of hypothesis testing follows a structured sequence:

1. State the null and alternative hypotheses in symbolic form.

2. Select the appropriate significance level (α).

3. Identify the test statistic and calculate its value.

4. Determine the critical value(s) or P-value.

5. Make a decision: reject or fail to reject the null hypothesis.

6. State the conclusion in simple, non-technical terms.

Symbolic Representation

• Use mathematical symbols such as <, >, =, ≠ to represent hypotheses.

• Example: Claim that most adults would erase their personal data online if they could: H0: p = 0.5, Ha: p > 0.5

Significance Level

Definition and Application

The significance level (α) is the probability cutoff used to determine if the sample data provides significant evidence against the null hypothesis. Common values are 0.10, 0.05, and 0.01.

• Significance level: $\alpha = P(\text{rejecting } H_0 \text{ when } H_0 \text{ is true})$

Test Statistic

Definition and Calculation

The test statistic is a value calculated from sample data that is used to decide whether to reject the null hypothesis. It is typically a z-score, t-score, or other standardized value.

• Formula for z-test (proportion): $z = \frac{\hat{p} - p}{\sqrt{\frac{p(1-p)}{n}}}$

• Example: For a survey of 565 adults, 95% would erase their data online. $z = \frac{0.59 - 0.5}{\sqrt{0.5 \times 0.5 / 565}} = 4.28$

Critical Region and Types of Tests

Critical Region

The critical region is the area of the sampling distribution that corresponds to values of the test statistic that would cause us to reject the null hypothesis.

• Two-tailed test: Critical region is in both tails.

• Left-tailed test: Critical region is in the left tail.

• Right-tailed test: Critical region is in the right tail.

P-value Method

Definition and Use

The P-value is the probability of obtaining a test statistic at least as extreme as the one observed, assuming the null hypothesis is true. It is used to decide whether to reject H0.

• Decision rule: If P-value < α, reject H0. If P-value > α, fail to reject H0.

• Example: For z = 4.28, P-value ≈ 0.0000 (very small, so reject H0).

Critical Value Method

Definition and Use

The critical value is the value of the test statistic that separates the region where we reject H0 from values that do not reject H0. It depends on the significance level and the type of test.

• Example: For α = 0.05 in a right-tailed test, critical value is 1.645. If z = 4.28 > 1.645, reject H0.

Stating Conclusions

Technical and Non-Technical Conclusions

After performing the test, conclusions should be stated in both technical and non-technical terms. Use "fail to reject" instead of "accept" to avoid implying proof of the null hypothesis.

Condition	Conclusion
Original claim does not include equality	There is sufficient evidence to support the claim
Original claim includes equality	There is not sufficient evidence to support the claim
Fail to reject H0	There is not sufficient evidence to support the claim
Reject H0	There is sufficient evidence to support the claim

Confidence Intervals for Hypothesis Testing

Relation to Hypothesis Testing

Confidence intervals can be used to test hypotheses. If a confidence interval for a population parameter does not contain the value in a claim, we should reject that claim.

Parameter	Claim	Confidence Interval	Conclusion
Proportion	p = 0.5	0.55 to 0.63	Reject claim
Mean	μ = 100	98 to 102	Fail to reject claim

Type I and Type II Errors

Definitions and Implications

Errors can occur in hypothesis testing:

• Type I Error (α): Mistakenly rejecting the null hypothesis when it is true. $\alpha = P(\text{Type I error})$

• Type II Error (β): Failing to reject the null hypothesis when it is false. $\beta = P(\text{Type II error})$

Example: Claim: p > 0.87. H0: p = 0.87, Ha: p > 0.87

• Type I error: Rejecting H0 when p = 0.87 is actually true.

• Type II error: Failing to reject H0 when p > 0.87 is actually true.

α and β are related; choosing one affects the other. Typically, α is set at a level deemed acceptable, and β is determined by sample size and other factors.

Summary Table: Hypothesis Testing Procedures

Step	Description
1. State Hypotheses	Formulate H0 and Ha in symbolic form
2. Select Significance Level	Choose α (e.g., 0.05)
3. Identify Test Statistic	Calculate z, t, etc.
4. Determine Critical Value or P-value	Find cutoff or probability
5. Make Decision	Reject or fail to reject H0
6. State Conclusion	Express result in non-technical terms

Additional info: These notes provide a comprehensive overview of hypothesis testing, including definitions, procedures, examples, and error types, suitable for college-level statistics students.