Online Short Course Longitudinal Data Analysis

February 23-24 and February 27-28, 2012

The Canadian Society for Epidemiology and Biostatistics is proud to offer an interactive online short course on longitudinal data analysis. This series of webinar workshops is designed for those who wish to enhance their knowledge and skills in advanced quantitative statistical methods used for longitudinal data analysis.

COURSE OBJECTIVES:

By the end of the course, participants should be able to:

1.	Understand and apply the following statistical analysis techniques used for longitudinal studies:
	•	Statistical models based on the generalized estimating equations approach
	•	Statistical models based on Maximum Likelihood Theory (Random Effects Models)
2.	Use statistical software (SAS or SPSS) for data analysis; and
3.	Interpret the results from statistical output.

TEACHING METHODS & MATERIALS:

Online webinar software will be used to enable students to participate interactively from their home or remote location in a lecture-style course given by a highly trained and experienced instructor. The interactive software will allow students from a distance to view the instructor's screen, listen to the lecture in real time, and participate and ask questions either orally or in written form (chat enabled). The instructor will provide lecture slides (PowerPoint) as well as programming code for use with the statistical packages: SAS and SPSS.

The webinar series is divided into two parts and each part is spread out over two consecutive days. A 3-hour lecture with short break will be given each day for a total of 12 hours of lecture time.

There is no required text but recommended texts are:

(i)	Diggle PJ, Liang KY, Zeger SL. Analysis of Longitudinal Data. Oxford Science Publication, 1995.
(ii)	Twisk JWR. Applied Longitudinal Data Analysis for Epidemiology – A Practical Guide. Cambridge University Press: Cambridge, UK, 2003.

All relevant notes will be provided by the instructor.

PRIOR REQUIRED KNOWLEDGE:

Understanding of multivariable linear and logistic regression techniques and familiarity with concepts such as confounding, interaction, and multi-collinearity.

Texts recommended for these topics:

• Applied Regression Analysis and Other Multivariable Methods:  Third Edition.  Kleinbaum, D. G., Kupper, L. L., and Muller, K. E., and Nizam, A., Duxbury Press, 1998.
• Logistic Regression: a Self-Learning Text, Kleinbaum, D. G., Springer-Verlag, 1994.

COURSE CONTENT:

PART I: Longitudinal Data Analysis – Models based on Quasi-Likelihood Theory

Dates: February 23 and February 24, 2012
Duration: Total 6 hours - 3 hours per day
Times: 10:00 - 13:00 EST

1.	Introduction
	•	Cross-sectional vs. Longitudinal Studies
	•	Advantages of Longitudinal Studies
	•	Characteristics of Longitudinal Studies
	•	Univariate Quasi-Likelihood
	•	Multivariate Quasi-Likelihood

2.	Longitudinal Data Analysis Using Generalized Estimating Equations Approach
	•	Marginal Models for Continuous and Binary Outcomes
	•	Transitional Models for Continuous and Binary Outcomes

PART II: Longitudinal Data Analysis – Models Based on Maximum Likelihood Theory

Dates: February 27 and February 28, 2012
Duration: Total 6 hours - 3 hours per day
Times: 10:00 - 13:00 EST

1.	Mixed Effects Models for Continuous Response
2.	Mixed Effects Models for Discrete Response
3.	Approaches for Handling Missing Data

REGISTRATION COSTS:

Registrant type	Full course – Parts I and II	Half course – Part I or II
CSEB member	$500 + HST	$350 + HST
Non-member	$800 + HST	$500 + HST
Student (member or non-member)	$350 + HST	$275 + HST

METHOD OF ASSESSMENT:

There are no formal assessment criteria for this course. However, each of the sessions includes practice examples which can be used to assess your knowledge of each of the topics.

CERTIFICATE OF COMPLETION:

Participants will be issued an electronic certificate of completion issued by the Canadian Society for Epidemiology and Biostatistics upon completion of the course.

REQUIRED MATERIALS:

Participants must have the following in order to participate in the webinar course:

•	A computer/laptop
•	Speakers/headphones/headset

OPTIONAL MATERIALS:

•	SAS software/SPSS software
•	A headset or microphone is required should the participant wish to ask questions or participate orally to the session

INSTRUCTOR BIOGRAPHY:

The workshop will be given by Dr. Punam Pahwa, MSc, PhD.

Dr. Pahwa is a Professor of Biostatistics in the Department of Community Health and Epidemiology, College of Medicine, University of Saskatchewan. Her research focus is to analyze complex health data sets by integrating recently developed statistical techniques in order to develop and improve population health research. There are two facets of her research: methodology and application. The methodological aspect of her research consists of integrating advanced statistical theories to analyze longitudinal complex health data. The applied aspect of her research consists in evaluating health determinants that are associated with chronic conditions, with a special focus on rural occupational and environmental exposures. This methodological aspect is not only an indispensable component of population health research, but it also helps many health researchers who face challenges in understanding large complex data sets.

INSTRUCTOR EXPERIENCE:

Pahwa P. Lecture on “Longitudinal Data Analysis” at Kumamoto University, Japan in 2010.

Pahwa P., and Senthilselvan A. Two-day workshop on “Longitudinal Data Analysis” for Health Canada in 2008.

Pahwa P., and Senthilselvan A. Pre-conference workshop on “Longitudinal Data Analysis” at the Sixth International Symposium, PHARE, Saskatoon, SK, October 18 - October 22, 2008.

Pahwa P. Sponsored by the Section on Statistics in Epidemiology, American Statistical Association to lead a roundtable (fee event) discussion on “Analysis of Longitudinal Complex Survey Datasets” at the Joint Statistical Meeting in August 2008, Denver, U.S.A. The focus of the roundtable session was to discuss recent developments and issues related to statistical theories to analyze longitudinal complex survey data.

---