Statistics Exam 2 Study Guide: Regression, Sampling, Probability, and Confidence Intervals

Linear Regression

Regression Equations and Interpretation

Linear regression is a statistical method used to model the relationship between two quantitative variables. The regression equation allows us to predict the value of one variable based on the value of another.

• Regression Equation: The general form is $y = \beta_0 + \beta_1 x$, where $\beta_0$ is the intercept and $\beta_1$ is the slope.

• Interpretation of Slope: The slope $\beta_1$ represents the expected change in $y$ for a one-unit increase in $x$.

• Prediction: Use the regression equation to predict $y$ values for given $x$ values.

Assessing the Fit of Regression

Evaluating how well the regression model fits the data is crucial for understanding the strength and reliability of the relationship.

• Correlation Coefficient ($r$): Measures the strength and direction of the linear relationship between variables. Values range from -1 to 1.

• Coefficient of Determination ($r^2$): Represents the proportion of variability in $y$ explained by $x$.

• Residuals: The differences between observed and predicted values. Analyzing residuals helps assess model fit and identify outliers.

• Interpretation of $r^2$: Higher $r^2$ values indicate a better fit; for example, $r^2 = 0.8$ means 80% of the variability in $y$ is explained by $x$.

Example

• Given data on students' study hours ($x$) and exam scores ($y$), a regression equation can predict exam scores based on study hours.

Sampling

Populations vs. Samples

Sampling is the process of selecting a subset of individuals from a population to estimate characteristics of the whole population.

• Population: The entire group of interest.

• Sample: A subset of the population used to make inferences.

• Sampling Methods: Simple random sampling, stratified sampling, cluster sampling, etc.

• Sampling Error: The difference between sample statistics and population parameters due to random variation.

Sampling Bias and Representativeness

• Bias: Systematic error that leads to inaccurate estimates.

• Representative Sample: A sample that accurately reflects the population.

• Example: Surveying only morning students may not represent all students.

Observational Studies vs. Experiments

Key Differences

Understanding the distinction between observational studies and experiments is essential for interpreting research findings.

• Observational Study: Researchers observe subjects without intervention.

• Experiment: Researchers manipulate variables to observe effects.

• Confounding Variables: Factors that may affect the outcome and are not controlled for.

• Randomization: Assigning subjects to groups randomly to reduce bias.

Example

• Observational: Studying the link between smoking and lung cancer by observing smokers and non-smokers.

• Experimental: Testing a new drug by randomly assigning patients to treatment and control groups.

Probability

Basic Probability Concepts

Probability quantifies the likelihood of events occurring and is foundational for statistical inference.

• Law of Large Numbers: As the number of trials increases, the sample mean approaches the population mean.

• Probability Rules: Addition Rule (for mutually exclusive events), Multiplication Rule (for independent events).

• Mutually Exclusive Events: Events that cannot occur together.

• Independent Events: The occurrence of one event does not affect the probability of another.

Example

• Flipping a coin: Probability of heads is $0.5$.

• Rolling a die: Probability of rolling a 4 is $\frac{1}{6}$.

Sampling Distributions and Confidence Intervals for Proportions

Sampling Distribution of a Proportion

The sampling distribution describes the distribution of sample proportions over repeated samples from the same population.

• Mean of Sampling Distribution: $p$ (population proportion).

• Standard Error: $\sqrt{\frac{p(1-p)}{n}}$ where $n$ is the sample size.

• Normal Approximation: For large $n$, the sampling distribution of the sample proportion is approximately normal.

Calculating Probabilities Using Sampling Distributions

• Use the normal model to estimate probabilities for sample proportions.

• Example: Probability that more than 55 heads occur in 75 coin flips.

Confidence Intervals for Proportions

Confidence intervals provide a range of plausible values for a population proportion based on sample data.

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

• Margin of Error: The term $z^* \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}$ quantifies uncertainty.

• Effect of Confidence Level and Sample Size: Higher confidence level or smaller sample size increases the width of the interval.

Example

• Estimating the proportion of voters who approve of a new law based on a sample.