Coding, Outputs, & Analysis
In this sample project I analyzed health data including the relationship between density and weight, from over 250 male participants. Sourced from BYU's Human Performance Research Center, the data was examined using SAS programming to identify trends. The project involved performing descriptive statistics on both the explanatory and response variables, as well as conducting Case QQ, Case CC, and Case QC analyses to assess data distribution and relationships. The following are snippets of my code as well as examples of my various outputs which were then summarized in a report.
Initial Coding and Prepping the Data
This section allows for the management and organization of the data from general explanatory and response variables to then be transformed into multilevel variables and binary variables for purposes of analysis.
General Analysis Code
This section of code is written allowing for initial analysis to occur such as the five number summary and various data visualizations which gives a basic level of insight into data trends.
The Regression Procedure
This output provides insight into the data including how well the model fits through R-Square and Adjusted R-Square. Additionally, the regression procedure tells us how important the predictors are , such as the p-value and t-test, which operates to explain whether or not the variables have a meaningful relationship and if that relationship has a predictor that really matters (is significantly different from zero).
The Frequency Procedure
Used for categorical data, this output allows for the count of how often categorical variables arise in the data set, provides contingency tables, and allows for the Chi-Squared test to be run on the data set. This test gives a lot of insight not only on the distribution of variables but also on the relationship between them.
