Hypothesis Testing, Confidence Intervals, and Inference in Statistics: Study Guide

Hypothesis Testing

Types of Hypotheses

Hypothesis testing is a fundamental method in statistics for making inferences about population parameters based on sample data. The two main types of hypotheses are:

• Null Hypothesis (H0): States that there is no effect or no difference. It is the hypothesis that the researcher tries to disprove.

• Alternative Hypothesis (Ha or H1): States that there is an effect or a difference. It is what the researcher wants to prove.

For example, to test if the mean price of a single-family home has changed, the hypotheses might be:

• $H_0: \mu = 243,714$

• $H_a: \mu \neq 243,714$

Types of Errors

• Type I Error: Rejecting the null hypothesis when it is actually true.

• Type II Error: Failing to reject the null hypothesis when the alternative hypothesis is true.

Example: If a test incorrectly concludes that the mean price has changed when it has not, a Type I error has occurred.

Steps in Hypothesis Testing

1. State the null and alternative hypotheses.

2. Choose a significance level (commonly $\alpha = 0.05$).

3. Collect sample data and calculate the test statistic.

4. Find the P-value or compare the test statistic to a critical value.

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

P-Values and Significance

Definition of P-Value

The P-value is the probability of obtaining a sample statistic as extreme as the one observed, assuming the null hypothesis is true.

• If the P-value is less than or equal to the significance level $\alpha$, reject the null hypothesis.

• If the P-value is greater than $\alpha$, fail to reject the null hypothesis.

Example: If $P = 0.03$ and $\alpha = 0.05$, we reject $H_0$.

Confidence Intervals

Constructing Confidence Intervals

A confidence interval estimates a population parameter (such as mean or proportion) with a specified level of confidence (e.g., 95%).

• For the mean: $\bar{x} \pm z^* \frac{\sigma}{\sqrt{n}}$ (if population standard deviation $\sigma$ is known)

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

Example: A 95% confidence interval for the mean age of victims is between 41.2 and 45.8 years.

Inference on Two Population Parameters

Comparing Two Means or Proportions

When comparing two populations, we often test hypotheses about the difference between their means or proportions.

• For means: $H_0: \mu_1 = \mu_2$ vs. $H_a: \mu_1 \neq \mu_2$

• For proportions: $H_0: p_1 = p_2$ vs. $H_a: p_1 \neq p_2$

The test statistic for the difference in means (assuming equal variances) is:

• $t = \frac{(\bar{x}_1 - \bar{x}_2) - (\mu_1 - \mu_2)}{\sqrt{s_p^2 \left(\frac{1}{n_1} + \frac{1}{n_2}\right)}}$

• Where $s_p^2$ is the pooled sample variance.

Example: Testing if the mean waiting time at a restaurant has decreased after a new system is implemented.

Inference on Categorical Data

Proportions and Survey Data

Categorical data analysis often involves estimating proportions and testing hypotheses about them.

• For a single proportion: $H_0: p = p_0$

• For two proportions: $H_0: p_1 = p_2$

Example: Testing if the proportion of adults who consider themselves to be liberal is higher than a polling organization's reported value.

Sampling Methods and Data Types

Types of Sampling

• Independent Sampling: Samples are selected independently from each population.

• Dependent Sampling: Samples are paired or matched in some way.

Types of Data

• Qualitative (Categorical) Data: Data that can be categorized based on traits or characteristics.

• Quantitative Data: Data that can be measured numerically.

Example: Survey responses about political philosophy (Conservative, Liberal, Moderate) are qualitative data.

Normal Probability Distribution and Tables

Z-Table and Critical Values

The Z-table provides the area under the standard normal curve to the left of a given z-score. It is used to find probabilities and critical values for hypothesis tests and confidence intervals.

• Critical value $z^*$ for a 95% confidence interval is approximately 1.96.

Example: To find the probability that a value is less than $z = 1.25$, look up 1.25 in the Z-table.

Correlation and Regression

Correlation Coefficient

The correlation coefficient ($r$) measures the strength and direction of a linear relationship between two variables.

• $r$ ranges from -1 to 1.

• Critical values for $r$ depend on sample size and significance level.

Example: If $r$ is greater than the critical value for the sample size, the relationship is statistically significant.

Tables

Z-Table (Excerpt)

Z	Area to Left
0.00	0.5000
1.00	0.8413
1.96	0.9750
2.00	0.9772

Correlation Coefficient Critical Values (Excerpt)

Sample Size	Critical Value
5	0.878
10	0.632
20	0.444

Example Data Table: Golf Ball Diameters

Golf Ball Diameter (inches)
1.682
1.684
1.687
1.681
1.685

Example Data Table: Political Philosophy

Political Philosophy
Conservative
Liberal
Moderate
Liberal
Conservative

Example Data Table: Employment and Response

Employment	Response
Full-Time	Yes
Part-Time	No
Full-Time	No
Part-Time	Yes

Additional info:

• Some questions involve using technology to compute P-values and confidence intervals, which is common in modern statistics courses.

• Critical values and normal probability plots are used to assess normality and the strength of relationships between variables.

• Many questions require interpretation of survey and experimental data, reflecting real-world applications of statistical inference.