Statistical Inference for Proportions and Chi-Square Procedures

Estimating an Unknown Population Proportion

Sample Proportion

Estimating a population proportion involves using sample data to infer the proportion of a characteristic in the entire population.

• Sample Proportion (p̂): The ratio of the number of successes to the total sample size, calculated as $\hat{p} = \frac{x}{n}$, where x is the number of successes and n is the sample size.

• Application: Used in surveys and experiments to estimate the proportion of individuals with a certain trait.

• Example: If 45 out of 100 surveyed students prefer online classes, $\hat{p} = 0.45$.

Sampling Distribution for Sample Proportion

Properties and Normal Approximation

The sampling distribution of the sample proportion describes the distribution of $\hat{p}$ over repeated samples.

• Mean: $\mu_{\hat{p}} = p$

• Standard Deviation: $\sigma_{\hat{p}} = \sqrt{\frac{p(1-p)}{n}}$

• Normal Approximation: For large samples, $\hat{p}$ is approximately normally distributed if $np \geq 10$ and $n(1-p) \geq 10$.

• Example: If $p = 0.3$ and $n = 100$, $\sigma_{\hat{p}} = \sqrt{\frac{0.3 \times 0.7}{100}} = 0.0458$.

1-Proportion Z-Interval Procedure (Confidence Interval)

Constructing a Confidence Interval for a Population Proportion

A confidence interval estimates the range in which the true population proportion is likely to fall.

• Formula: $\hat{p} \pm z^* \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}$

• z*: Critical value from the standard normal distribution for the desired confidence level (e.g., 1.96 for 95%).

• Example: For $\hat{p} = 0.45$, $n = 100$, and 95% confidence, the interval is $0.45 \pm 1.96 \times 0.0497$.

Margin of Error

The margin of error quantifies the uncertainty in the estimate of the population proportion.

• Formula: $E = z^* \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}$

• Interpretation: The confidence interval extends $E$ units above and below $\hat{p}$.

Calculation of Margin of Error

To calculate the margin of error, substitute the sample proportion and sample size into the formula above.

• Example: For $\hat{p} = 0.45$, $n = 100$, and $z^* = 1.96$, $E = 1.96 \times \sqrt{\frac{0.45 \times 0.55}{100}} = 0.097$.

Finding the Number of Samples Needed

To achieve a desired margin of error, solve for n in the margin of error formula.

• Formula: $n = \frac{z^{*2} p^* (1-p^*)}{E^2}$, where $p^*$ is an estimated proportion (often 0.5 for maximum variability).

• Example: To achieve $E = 0.05$ with $z^* = 1.96$ and $p^* = 0.5$, $n = \frac{1.96^2 \times 0.5 \times 0.5}{0.05^2} = 384.16$ (round up to 385).

Hypothesis Tests for One Population Proportion

Testing Claims About a Population Proportion

Hypothesis testing for proportions evaluates whether the observed sample proportion differs significantly from a hypothesized value.

• Null Hypothesis ($H_0$): $p = p_0$

• Alternative Hypothesis ($H_a$): $p \neq p_0$, $p > p_0$, or $p < p_0$

• Test Statistic: $z = \frac{\hat{p} - p_0}{\sqrt{\frac{p_0(1-p_0)}{n}}}$

• P-value: Probability of observing a test statistic as extreme as, or more extreme than, the observed value under $H_0$.

• Decision: Compare p-value to significance level $\alpha$ (e.g., 0.05) to accept or reject $H_0$.

Chi-Square Procedures

Chi-Square Distribution and Degrees of Freedom

The chi-square distribution is used in tests of categorical data and is defined by its degrees of freedom.

• Degrees of Freedom (df): Number of independent values in the calculation, often $df = k - 1$ for goodness-of-fit, or $df = (r-1)(c-1)$ for contingency tables.

• Shape: Positively skewed, with mean $df$ and variance $2 \times df$.

The Chi-Square Distribution Table

The chi-square table provides critical values for various degrees of freedom and significance levels.

• Usage: Find the critical value for your test's degrees of freedom and chosen $\alpha$.

• Example: For $df = 3$ and $\alpha = 0.05$, the critical value is approximately 7.815.

Chi-Square Goodness of Fit Test

This test evaluates whether observed categorical data fit a specified distribution.

• Test Statistic: $\chi^2 = \sum \frac{(O_i - E_i)^2}{E_i}$, where $O_i$ is observed and $E_i$ is expected frequency.

• Critical Value Approach: Compare $\chi^2$ to the table value for $df$ and $\alpha$.

• P-Value Approach: Calculate the probability of observing a $\chi^2$ value as extreme as the test statistic.

• Example: Testing if dice are fair by comparing observed roll frequencies to expected frequencies.

Chi-Square Independence Test

This test determines whether two categorical variables are independent.

• Contingency Tables (Two-Way Tables): Organize data by categories for both variables.

• Expected Cell Values: $E_{ij} = \frac{(\text{Row Total}) \times (\text{Column Total})}{\text{Grand Total}}$

• Test Statistic: $\chi^2 = \sum \frac{(O_{ij} - E_{ij})^2}{E_{ij}}$

• Critical Value Approach: Compare $\chi^2$ to the critical value for $df = (r-1)(c-1)$.

• P-Value Approach: Find the probability of observing the test statistic under the null hypothesis of independence.

• Example: Testing if gender and preference for a product are independent.

Example Contingency Table

Gender	Prefer Product A	Prefer Product B	Total
Male	30	20	50
Female	25	25	50
Total	55	45	100

Additional info: The above table is a typical example for a chi-square independence test, showing observed frequencies for two categorical variables.