What This Document Is
This document is a class participation activity for Biostatistics (HCD 300) at Arizona State University, focused on applying two-sample t-tests for independent samples. It presents a series of problems and solutions related to hypothesis testing and statistical analysis using real-world health data.
Why This Document Matters
This activity is designed for students enrolled in the course to practice and demonstrate their understanding of t-tests. It’s used to reinforce concepts covered in lectures and associated PowerPoint presentations, specifically regarding comparing means between two independent groups. Successful completion indicates a grasp of statistical inference and its application to public health datasets.
Common Limitations or Challenges
This activity provides *completed* examples. It does not offer a comprehensive tutorial on the underlying theory of t-tests, nor does it cover all possible scenarios or variations of the test. Users will still need to understand the broader statistical concepts and be able to apply these tests to new datasets independently. It assumes prior knowledge of statistical concepts.
What This Document Provides
The full document includes: formulated null and alternative hypotheses for comparing physical health data between Boston and Seattle residents, explanations of the uses and assumptions of two-sample t-tests, a justification for choosing a t-test over a z-test, calculations of variance, degrees of freedom, and p-values using both Excel functions and the Data Analysis Toolpak, interpretation of Cohen’s d for effect size, and a second analysis comparing mental health days between males and females, including hypothesis formulation, variance checks, p-value determination, and effect size interpretation. It provides completed calculations and conclusions for each step. This preview does *not* include the actual Excel calculations or the full output from the Data Analysis Toolpak.