Statistical Inference, Hypothesis Testing, and Confidence Intervals: Study Guide

Foundations of Statistical Inference

Introduction to Statistical Inference

Statistical inference is the process of using data collected from a sample to make conclusions or predictions about a larger population. This process is fundamental to statistics and underpins hypothesis testing and confidence interval estimation.

• Sample: A subset of the population from which data is actually collected.

• Population: The entire group about which information is desired.

• Parameter: A numerical summary of a population (e.g., mean μ, proportion p).

• Statistic: A numerical summary of a sample (e.g., sample mean \( \bar{x} \), sample proportion \( \hat{p} \)).

Hypothesis Testing

Key Components of a Hypothesis Test

Hypothesis testing is a formal procedure for comparing observed data with a claim (hypothesis) about a population parameter.

• Null Hypothesis (\( H_0 \)): The statement being tested, typically representing no effect or the status quo. It always contains an equality sign (\( = \)).

• Alternative Hypothesis (\( H_a \)): The statement you are seeking evidence for, using \( > \), \( < \), or \( \neq \).

• p-value: The probability, assuming \( H_0 \) is true, of obtaining a result as extreme or more extreme than the observed sample result.

• Significance Level (\( \alpha \)): The threshold for determining whether a p-value is small enough to reject \( H_0 \). Common values are 0.05 or 0.01.

Decision Matrix for Hypothesis Testing

Decisions in hypothesis testing are based on comparing the p-value to the significance level \( \alpha \):

Condition	Action	Conclusion
p-value \( \leq \alpha \)	Reject \( H_0 \)	There is enough evidence to support \( H_a \).
p-value \( > \alpha \)	Fail to Reject \( H_0 \)	There is not enough evidence to support \( H_a \).

Proportions vs. Means

Testing Proportions

Used when the variable of interest is categorical (e.g., Yes/No, Success/Failure).

• Parameter: Population proportion (\( p \)).

• Test Statistic: z-test statistic is used for large samples.

• Example: Testing if more than 50% of residents support a traffic idea.

Formula for the z-test statistic for a proportion:

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

• \( \hat{p} \): Sample proportion

• \( p_0 \): Hypothesized population proportion

• \( n \): Sample size

Testing Means

Used when the variable of interest is quantitative (e.g., weights, test scores).

• Parameter: Population mean (\( \mu \)).

• Test Statistic: t-test statistic is used when the population standard deviation is unknown and the sample size is small.

• Example: Testing if the average weight of candy bars is 5 ounces.

Formula for the t-test statistic for a mean:

$t = \frac{\bar{x} - \mu_0}{s / \sqrt{n}}$

• \( \bar{x} \): Sample mean

• \( \mu_0 \): Hypothesized population mean

• \( s \): Sample standard deviation

• \( n \): Sample size

Confidence Intervals

Understanding Confidence Intervals

A confidence interval (CI) provides a range of plausible values for a population parameter, calculated by adding and subtracting a margin of error from a sample statistic.

• Interpretation: "We are 95% confident that the true population parameter lies between [lower bound] and [upper bound]."

• Evaluating Claims: If a claimed value is not within the CI, there is evidence against the claim.

• Comparing Two Means: If the CI for the difference between two means contains zero, the population means may be the same.

General formula for a confidence interval:

$\text{Statistic} \pm \text{(Critical Value)} \times \text{Standard Error}$

• For means (when \( \sigma \) unknown): $\bar{x} \pm t^* \frac{s}{\sqrt{n}}$

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

Problem-Solving Checklist

1. Identify the variable: Is it a proportion (%) or a mean (average)?

2. Set up hypotheses: \( H_0 \) always uses \( = \); \( H_a \) uses \( > \), \( < \), or \( \neq \).

3. Run the test: Use technology or formulas to find the test statistic (z or t) and the p-value.

4. Compare: Is p-value \( \leq \alpha \)?

5. Conclude: State if there is enough evidence to support the original claim.

Calculator/Technology Practice

• 1-PropZTest: For single proportion claims.

• 2-PropZTest: For comparing two proportions.

• T-Test: For single mean claims.

• 2-SampleTTest: For comparing two means.

• T-Interval: For finding mean confidence intervals.

Example Applications

• Proportion Example: Testing if more than 50% of residents support a new policy using a 1-PropZTest.

• Mean Example: Testing if the average test score is different from 75 using a T-Test.

• Confidence Interval Example: Calculating a 95% CI for the average weight of a product.

Additional info: The notes synthesize core concepts from chapters on hypothesis testing for proportions and means, confidence intervals, and the use of statistical technology, corresponding to Ch. 8 and Ch. 9 of a typical statistics curriculum.