Department of Sociology 
SOC 525 - Longitudinal Research

Instructor: Rajulton Fernando 
Room : SSC 5405 
Tel: 661-3693 
fernando@julian.uwo.ca 

The aim of this course is to introduce you to the richness of longitudinal research design and to appropriate analytical procedures. Although we shall be focusing on a variety of longitudinal analyses in the social sciences (since this course is offered in the department of sociology), you are welcome to explore, and share with others, the variety of applications in other disciplines. Special attention will be given to methodological issues and to new software that have been made available in recent times for analyzing longitudinal data. 

You are expected to be quantitative analysts at least to the extent of a working knowledge of multiple regression techniques at 300 level, and preferably at 500 level, of statistics courses offered in the department. Many components of this course will depend on your interest in longitudinal research and enthusiasm exhibited during the course work. For each major topic, I shall assign some empirical examples to work on, either with the existing packages or with the new software that I shall be suggesting to you. Discussion on these assignments during the class will be an important training aspect of this course. 

The course material is cumulative, so it is essential that you do the reading and exercises before each class. As an advanced level course, those who are taking this course for credit will be required to produce a research paper of normal length (about 20 pages as for a journal) on any topic of interest that requires using data from any longitudinal survey of your choice. [Those who are auditing this course are also strongly encouraged to work on a similar paper.] We shall aim at publication of these papers in refereed journals as well as on the web site on longitudinal research maintained by your instructor. In addition, some of the assignments given to the class (I shall let you know which ones) will also be counted for credit. These assignments will cover 30% and the project paper 70% of the final grade. 

Readings will be suggested for every week. Please keep up to date with the readings. I shall place a copy of all the reading materials, from various books, journals and research papers in the reserved section of the Population Studies Centre (SSC 3227). [Some well-known books are listed on the last page. Please do not hurry to pick them up from the library; not all topics in these books are important, neither are they useful! Read only those I suggest in the class.]  

Tentative Topics : Some important topics in the area of longitudinal research are listed below. It is obvious that we may not be able to cover all the topics during the term. I shall choose topics depending on your interests and enthusiasm! It also depends on the depth and extent of treating various topics and on the knowledge and experience you have in dealing with them. 

1. Introduction : Research designs for longitudinal studies 

Rationale for longitudinal research 
Longitudinal data and longitudinal studies 
Longitudinal surveys in Canada 

2. Advantages, Disadvantages and Problems in dealing with longitudinal data: 

Attrition and non-response 
Conditioning effects 
Age, period and Cohort effects 
Missing data 

3. Concept of Change :

Growth, change, stability/sameness 
Individual and aggregate change 
Role of longitudinal studies in assessing change 
Statistical model for the analysis of change 
Analysis of change with binary and categorical data 
Growth curve analysis methods for longitudinal data 

4. Multilevel Models :

Introduction 
Multilevel models and longitudinal data 
Unconditional models 
Conditional models 

5. Causal Models with longitudinal data : 

2W2V models 
Random effects models 
Using multilevel data in a causal framework 

6. Event history analysis :

Introduction 
Event history analysis for discrete-time data 
Event history analysis for continuous-time data 
Studies of states and durations 
Overview of survival analysis, analysis of durations 
Analysis of Sequences 

7. Regression methods for longitudinal data: 

Introduction 
(Failures/follies of) Path Analysis and Structural Equation Models 
Latent variable modeling 

8. Problem of Unobserved Heterogeneity : 

Introduction 
Ways of handling heterogeneity 

9. Nonlinear Dynamics of Change 


Bibliography 

Crouchley, Robert. Longitudinal Data Analysis, Surrey Conferences on Sociological theory and Method. Avebury, Aldershot:1987. 

Uncles, M.D. (ed.), Longitudinal Data Analysis: Methods and Applications, London papers in Regional Science 18, Pion Ltd. London, 1988. 

Giele, J.Z. and Elder Jr. G. H. Methods of Life Course Research, Quantitative and Qualitative Approaches, Sage Pub., 1998. 

Engel, U. and J. Reinecke, Analysis of Change, Advanced Techniques in Panel Data Analysis, Walter de Gruyter, N.Y. 1996. 

Magnusson, D., L. R. Bergman, G. Rudinger, and B.Torestad. Problems and Methods in Longitudinal Research: Stability and Change. Cambridge University Press, N.Y., 1991. 
(Good for studies in psychology) 

Gottman, J.M. The Analysis of Change, Lawrence Erlbaum, U.K., 1995. 

Blossfeld, H-P and G. Rohwer, 1995. Techniques of Event History Modeling. Mahwah, NJ: Lawrence-Erlbaum. 

Blossfeld, H-P, A. Hamerle, and K.U. Mayer, 1989. Event History Analysis. Hillsdale, NJ: Lawrence Erlbaum. 

Yamaguchi, K.E., 1991 Event History Analysis. Newbury Park, CA: Sage. 

Lee, E.T. 1992. Statistical Methods for Survival Data Analysis (second ed.) NY: John Wiley. 

Mayer, K.U. and N.B. Tuma (eds.) 1990. Event History Analysis in Life Course Research. Madison: The University of Wisconsin Press. 

Trussell J., R. Hankinson, and J. Tilton (eds.), 1992. Demographic Applications of Event History Analysis. NY: Oxford University Press. 

Plewis, 1985. Analysing Change. Wiley. 

Menard S. 1991. Longitudinal Research. Sage.