Hypothesis Testing for Proportions: Study Notes

Hypothesis Testing with One Sample

Hypothesis Testing for Proportions

Hypothesis testing for proportions is a statistical method used to determine whether a sample proportion significantly differs from a hypothesized population proportion. This technique is commonly applied when analyzing binomial data, such as the proportion of individuals exhibiting a certain characteristic.

• Population Proportion (p): The proportion of individuals in the population with a specific characteristic.

• Sample Proportion (\hat{p}): The proportion observed in the sample, calculated as \( \hat{p} = \frac{x}{n} \), where x is the number of successes and n is the sample size.

• z-Test for Proportion: Used when the sample size is large and the binomial distribution can be approximated by the normal distribution.

• Standardized Test Statistic (z): Measures how far the sample proportion is from the hypothesized proportion, in units of standard error.

Formula for the z-Test Statistic:

$z = \frac{\hat{p} - p}{\sqrt{\frac{p(1-p)}{n}}}$

Conditions for Using the z-Test:

• Both \( np \) and \( n(1-p) \) must be greater than 5.

• The sample must be random and independent.

Steps in Hypothesis Testing for Proportions

To conduct a hypothesis test for a population proportion, follow these steps:

1. State the Hypotheses: Formulate the null hypothesis (\( H_0 \)) and the alternative hypothesis (\( H_a \)).

2. Choose the Significance Level (\( \alpha \)): Common values are 0.01, 0.05, or 0.10.

3. Calculate the Test Statistic: Use the formula above to compute the z-value.

4. Determine the Rejection Region or P-value: Based on the test type (one-tailed or two-tailed), identify the critical value or calculate the P-value.

5. Make a Decision: Compare the test statistic or P-value to the significance level and decide whether to reject or fail to reject the null hypothesis.

Example: Hypothesis Testing Using a Rejection Region

A researcher claims that less than 39% of U.S. adults have fallen asleep during a movie in a movie theater. In a random sample of 100 U.S. adults, 35% say they have fallen asleep during a movie. At the 1% significance level, is there enough evidence to support the researcher’s claim?

• Step 1: State the hypotheses: $H_0: p = 0.39$ $H_a: p < 0.39$

• Step 2: Check conditions: \( np = 100 \times 0.39 = 39 \), \( n(1-p) = 100 \times 0.61 = 61 \) (both > 5).

• Step 3: Calculate the sample proportion: \( \hat{p} = 0.35 \).

• Step 4: Compute the z-statistic: $z = \frac{0.35 - 0.39}{\sqrt{\frac{0.39 \times 0.61}{100}}}$

• Step 5: Determine the critical value for a left-tailed test at \( \alpha = 0.01 \): \( z_{critical} = -2.33 \).

• Step 6: Compare z to the rejection region. If z is not less than -2.33, fail to reject \( H_0 \).

• Conclusion: There is not enough evidence at the 1% significance level to support the claim.

Example: Hypothesis Testing Using a P-Value

A researcher claims that 25% of U.S. adults who do not have a parent with a bachelor’s degree or beyond have completed a bachelor’s degree or beyond themselves. In a random sample of 11,400 U.S. adults, 2,900 say they have completed a bachelor’s degree. At the 10% significance level, is there enough evidence to reject the researcher’s claim?

• Step 1: State the hypotheses: $H_0: p = 0.25$ $H_a: p \neq 0.25$

• Step 2: Check conditions: \( np = 11,400 \times 0.25 = 2,850 \), \( n(1-p) = 11,400 \times 0.75 = 8,550 \) (both > 5).

• Step 3: Calculate the sample proportion: \( \hat{p} = \frac{2,900}{11,400} = 0.254 \).

• Step 4: Compute the z-statistic: $z = \frac{0.254 - 0.25}{\sqrt{\frac{0.25 \times 0.75}{11,400}}}$

• Step 5: Find the P-value using the standard normal table. For z = 1.08, area to the right is 0.1401. For a two-tailed test, P-value = 2 × 0.1401 = 0.2802.

• Step 6: Compare P-value to \( \alpha = 0.10 \). Since P-value > 0.10, fail to reject \( H_0 \).

• Conclusion: There is not enough evidence at the 10% significance level to reject the claim.

Sample Proportion Formula

The sample proportion is a fundamental statistic in hypothesis testing for proportions. It is calculated as:

$\hat{p} = \frac{x}{n}$

• x: Number of successes in the sample

• n: Sample size

Summary Table: Hypothesis Testing for Proportions

This table summarizes the key steps and formulas used in hypothesis testing for proportions.

Step	Description	Formula
State Hypotheses	Null and alternative hypotheses	$H_0: p = p_0$ $H_a: p \neq p_0$ or $p < p_0$ or $p > p_0$
Sample Proportion	Calculate from data	$\hat{p} = \frac{x}{n}$
Test Statistic	Standardized z-value	$z = \frac{\hat{p} - p_0}{\sqrt{\frac{p_0(1-p_0)}{n}}}$
Decision	Compare z or P-value to critical value or significance level	Reject or fail to reject $H_0$

Applications

• Testing claims about proportions in populations (e.g., disease prevalence, voting behavior).

• Evaluating effectiveness of interventions (e.g., proportion of patients responding to treatment).

• Quality control in manufacturing (e.g., proportion of defective items).