SAS Essential Training: 1 Descriptive Analysis for Healthcare Research

Introduction

• SAS essential training: Descriptive analysis
• What you should know

1. Preparing for Analysis

• BRFSS dataset and documentation
• Organizing your work
• Hypotheses for the demonstration
• Explanation of SAS XPT
• Reading in SAS XPT file
• KEEP and DROP commands
• Reviewing confounders

2. Applying Exclusions

• Alternatives to labels and formats
• Using PROC FREQ to prepare
• Removing rows by categorical criteria
• Removing rows by continuous criteria
• Removing rows where values are missing
• Observations from the log file
• Updating data reduction documentation

3. Developing the Analytic Dataset

• Recoding a grouped variable
• Recoding the smoking variable
• Creating flags: Indicator variables
• Grouping a continuous variable
• Cleaning a continuous variable
• Completing the analytic dataset

4. Helpful Plots

• Bar chart
• Pie chart
• Histogram
• Box and whisker plot
• Scatter plot

5. Categorical Descriptive Analysis

• Choosing an analytic approach
• Creating a categorical table shell
• PROC FREQ with options
• Putting results in a table
• Chi-square and Fisher’s exact tests

6. Continuous Descriptive Analysis

• Normal plot
• Creating a continuous table shell
• PROC SORT and PROC UNIVARIATE
• Transferring continuous results to a table
• Conducting t-tests
• Conducting one-way ANOVAs

7. Alternative Approaches to Descriptive Analysis

• Tricks in Excel
• PROC TABULATE for categorical analysis
• PROC TABULATE for continuous analysis
• Explanation of the ODS
• Example of ODS and PROC FREQ
• Example of ODS with PROC UNIVARIATE

8. Tips and Tricks

• Using weights with PROC FREQ
• Recoding into quartiles
• Creating an index variable
• Using the Import wizard
• Sorting and merging
• Introduction to PROC SQL

Conclusion

• Review of the process
• Next steps