What This Document Is
This document presents detailed output from a statistical analysis performed using SAS software, specifically within the context of Biostat 411: Analysis of Correlated Data at UCLA. It focuses on the application of mixed-effects models to hierarchical data, stemming from the Television, School, and Family Smoking Prevention and Cessation (TVSFP) project. The output showcases a practical example of how these models are implemented and interpreted. It’s a record of the analytical process, detailing the steps taken to prepare and analyze the data.
Why This Document Matters
This resource is invaluable for students enrolled in advanced biostatistics courses dealing with correlated data structures. It’s particularly helpful for those learning to apply mixed-effects modeling techniques to real-world research scenarios. Individuals preparing to conduct similar analyses, or needing to interpret the output of such analyses, will find this a useful reference. It’s best utilized *after* gaining a foundational understanding of mixed-effects models and SAS programming, as it dives directly into the application of these concepts.
Topics Covered
* Hierarchical data structures
* Mixed-effects modeling
* Longitudinal data analysis
* SAS programming for statistical analysis (PROC MIXED)
* Estimation and hypothesis testing in correlated data
* Data transformation (wide to long format)
* Interpretation of statistical output
What This Document Provides
* SAS code used for data manipulation and analysis
* Output from PROC CONTENTS, detailing data structure
* Output from PROC PRINT, showing a sample of the dataset
* Results from PROC MIXED, including model information, convergence diagnostics, and parameter estimates
* Details of contrast statements used for specific comparisons
* A listing of variables and their attributes within the dataset
* Sample data observations to illustrate the data structure.
