Introduction to Statistical Analysis: Hypothesis Testing

This introductory module provides an overview of the course and introduces the data that will be analyzed throughout the course. Students learn about course prerequisites, how to access SAS software, and the process for setting up the necessary data files for completing practice exercises. The module includes a demonstration of exploring the Ames Housing Data set, which will be used for analysis in later modules.

This module covers fundamental statistical concepts needed for hypothesis testing. Topics include types of variables, statistical modeling approaches, distinctions between explanatory and predictive modeling, population parameters and sample statistics, normal distribution, standard error, confidence intervals, and p-values. Students learn to perform and interpret one-sample and two-sample t-tests using PROC TTEST in SAS, including testing for equal and unequal variances.

This comprehensive module teaches graphical and statistical methods for determining useful predictors in data analysis. Students learn to create and interpret box plots and scatter plots to identify potential associations, perform one-way ANOVA using PROC GLM, conduct post-hoc pairwise comparisons, calculate and interpret correlation statistics with PROC CORR, and implement simple linear regression using PROC REG. The module covers hypothesis testing for ANOVA and regression, partitioning variability, coefficient of determination, F statistics, and assumptions for these statistical methods.

This advanced module expands on previous statistical techniques, introducing two-factor analysis of variance (two-way ANOVA) and multiple linear regression. Students learn to implement and interpret two-way ANOVA with and without interactions using PROC GLM, perform post-processing analysis with PROC PLM, and fit multiple linear regression models using PROC REG. The module covers hypothesis testing for multiple regression, understanding interaction effects, adjusted R-square, and comparing multiple regression to simple linear regression models, providing skills to analyze more complex relationships in data.