Categorical Data Analysis (PUBH5212)

UNIT OF STUDY

In this unit the biostatistical concepts covered in earlier units are extended to cover analysis of epidemiological studies where the outcome variable is categorical. Topics of study include: testing for trend in a 2 x r contingency table; the Mantel-Haenszel test for the combination of several 2 x 2 tables, with estimation of the combined odds ratio and confidence limits; multiple logistic regression; Poisson regression; modelling strategy. The assignments will involve practical analysis and interpretation of categorical data. Data analyses will be conducted using statistical software (SAS).

Further unit of study information

Classes

1 x 2hr lecture, 5 x 1hr lectures, and 5 x 1hr tutorials over 6 weeks. Also available online - such students must have access to a computer running a version of Microsoft Windows compatible with the latest version of SAS.

Assessment

1x 3 page report (30%) and 1x 8 page report (70%)

Textbooks

Course notes are provided.

Faculty/department permission required?

No

Unit of study rules

Prerequisites and assumed knowledge

PUBH5018

Corequisites

PUBH5211

Study this unit outside a degree

Non-award/non-degree study

If you wish to undertake one or more units of study (subjects) for your own interest but not towards a degree, you may enrol in single units as a non-award student.

Cross-institutional study

If you are from another Australian tertiary institution you may be permitted to underake cross-institutional study in one or more units of study at the University of Sydney.