Longitudinal data are important for many social science disciplines as such data facilitates the investigation of empirical research questions relating to growth, dynamics and change. Such investigations are not possible with cross-sectional data. Longitudinal data allows researchers to model the sequence of events, such as entering and exiting states, such as unemployment, poverty, marriage, divorce etc. Longitudinal data also provides greater statistical power with more than one observation per person and the ability to use fixed effect models which control for the effects of all unobserved (but stable) influences.
The purpose of the course will be providing students with a range of advanced skills for the analysis of longitudinal data. The statistical techniques taught include repeated measures, random and fixed effects models and event history analysis. There will be an emphasis of proposing plausible research questions and analysing data to investigate these questions. This course will include a mixture of presentations and practical computer based lab sessions. Students will analyse two major Australian longitudinal studies: the Longitudinal Surveys of Australian Youth (LSAY) and the Household Income and Labour Dynamics of Australian study (HILDA).
The software package used will be SAS. Other statistical packages will not be used.
Since much of the work analysing longitudinal data involves data preparation, students should be capable of understanding SAS syntax. Participants should be familiar with OLS regression and logistic regression. It would be useful for students to become familiar with both the LSAY and HILDA studies.
Course notes will be supplied
Course notes will be supplied.