Learning & Teaching Methods
One two-hour lecture per week, followed by one two-hour lab session in the same week.
Assessment
Whichever is the greater: EITHER 50 per cent Coursework Mark, 50 per cent Exam Mark OR 100 per cent Exam Mark
Coursework
One piece of directed applied coursework using Stata.
Examination
Two-hour exam during Summer Examination period
Aims of the course
This course gives students a practical grounding in the theory and methods of panel data analysis. It has the following key aims:
- To allow students to interpret and critically assess published studies using panel data
- To provide students with the skills and confidence to manipulate panel datasets on their own in the future
- To give an overview of different approaches to panel data analysis
- To develop practical skills in selecting and conducting different types of panel data analysis
- To provide an opportunity for students to compare results of analysing the same data with different panel methods
- The course reviews standard regression methods (OLS, logit and probit) and covers longitudinal data manipulation, transition matrices, continuous and discrete fixed and random effects models, and survival analysis.
Lab Sessions
Each lecture is followed by a lab-based session where students will use Stata to implement the methods covered in the lectures. Please note that this is an intensive course, and most students will need to spend one or two hours in the lab each week, in addition to these scheduled sessions, in order to cover the work.
The data used will be a subset from the British Household Panel Survey (BHPS), and the exercises will involve the sort of analysis that professional social scientists might need to undertake.
Most sessions will build on the work of a previous session; it is therefore important that students keep copies of all their do-files and outputs.
Students should already be familiar with the fundamentals of Stata, including:
- basic data management techniques
- working in interactive and batch mode
- basic analytical techniques such as OLS, logit and probit
If you are unsure about your competence in Stata, please talk to the course tutors well before the start of the course.
Lecture Outline
- Week 16
A review of concepts for regression modelling, or what you should know already - Week 17
Understanding and using panel data; transition matrices - Week 18
Fixed effects and random effects models - Week 19
More on fixed and random effects models, and an introduction to survival/event history analysis - Week 20
More on survival/ event history analysis
Detailed course outline
Detailed course outline (.pdf)