Categorical Data Analysis
This course focuses on categorical data analysis techniques common in clinical and health outcomes research. Students will learn how to apply and interpret bivariate analysis, measures of agreement, logistic regression, proportional odds models, and models for matched/correlated data.
Knowledge commensurate with undergraduate intermediate algebra (equivalent to Loyola's MATH 100 course), undergraduate introductory statistics (equivalent to Loyola's STAT 103 course), GNUR 546 (Introduction to Linear Models), and good working knowledge of a statistical package.
Outcomes: Upon successful completion of the course, the student will be able to: 1) Apply appropriate categorical analysis techniques based on research questions about differences in proportions or trends; 2) Evaluate diagnostic tests through sensitivity, specificity, and positive and negative predictive value calculations; 3) Explain sample size or statistical power for studies with binary outcomes; 4) Interpret goodness of fit and calibration of the model to categorical outcome data; 5) Interpret findings of statistical categorical analyses from software output for select tests; 6) Critique categorical data designs, including randomization and power analysis.
Knowledge commensurate with undergraduate intermediate algebra (equivalent to Loyola's MATH 100 course), undergraduate introductory statistics (equivalent to Loyola's STAT 103 course), GNUR 546 (Introduction to Linear Models), and good working knowledge of a statistical package.
Outcomes: Upon successful completion of the course, the student will be able to: 1) Apply appropriate categorical analysis techniques based on research questions about differences in proportions or trends; 2) Evaluate diagnostic tests through sensitivity, specificity, and positive and negative predictive value calculations; 3) Explain sample size or statistical power for studies with binary outcomes; 4) Interpret goodness of fit and calibration of the model to categorical outcome data; 5) Interpret findings of statistical categorical analyses from software output for select tests; 6) Critique categorical data designs, including randomization and power analysis.