Applied Longitudinal Data Analysis

This course provides an overview of Longitudinal Data Analysis. As well as the statistical theory, an overview of the many applications and capabilities of LDA is given.

This course is designed as an introductory course for applied researchers and as such, is suitable for participants who want to develop a fundamental knowledge of LDA techniques.

 
Level 3 - runs over 5 days
Instructor: 

Dr Mark Griffin is the Director of Insight Research Services Associated (www.insightrsa.com), and holds Adjunct appointments within the School of Public Health, University of Queensland and the Sydney Medical School, University of Sydney. Mark serves on the Executive Committee for the Statistical Society of Australia, and is Chair of their Section for Business Analytics. Mark also serves as the Asia-Pacific Regional Director for the International Institute of Business Analysis, is Chair of their Business Analytics Special Interest Group, and is an IIBA Endorsed Education Provider. He is currently doing research with the Queensland Ambulance Service analyzing their incident reports, where the QAS visits approximately 700,000 incidents per year. To date he has presented over 80 two-day and 10 five-day workshops in statistics around Australia.

About this course: 

Longitudinal Data Analysis is a very popular statistical method in a range of fields including medicine, natural resource management, business and economics.

 

As well as allowing a researcher to elicit the changes in a subject (person, business, etc) over time, longitudinal data is also exceedingly powerful as it allows the within-subject and between-subject variance to be estimated independently (generally allowing the parameters in the statistical model to be estimated with a much tighter accuracy than traditional models).

 

This course provides an overview of Longitudinal Data Analysis. As well as the statistical theory, an overview of the many applications and capabilities of LDA is given. The course is not particularly mathematical, but instead places emphasis on the fundamental concepts of LDA and how it is used by applied researchers.

 

The workshop will focus on linear mixed effects models, however we will also go through generalized linear mixed effects models and any new material a participant will need to understand in order to successfully use generalized LME’s. There will be exercises using both types of models.

 

The computer exercises for this workshop will be conducted in Stata. Stata has been chosen as it provides both basic and advanced functionality for conducting longitudinal data analyses. A full set of exercises and solutions in Stata will be provided in this workshop, where they include the Stata syntax required to perform each individual exercise. In addition the workshop presenter will demonstrate how to use Stata for the first exercise in each session prior to workshop participants individually working though the exercises for that session.

 

General aims of the course are for students to develop a readiness for using LDA software and to develop the requisite knowledge for applying LDA methods and models in an intelligent way. Note that participants may be invited to briefly present their own research on the last day of class. This exercise, along with the formal lecture material, might help participants to chart a direction forward in their study and application of LDA.

 

This course has also been developed in consultation with staff from the National Centre for Longitudinal Data, Dept of Social Services. As such students will have access to a number of datasets through this workshop including one or more of:

  • The Housing, Income and Labour Dynamics in Australia (HILDA) survey
  • The Longitudinal Study of Australian Children (LSAC)
  • The Longitudinal Study of Indigenous Children (LSIC)
  • Building a New Life in Australia (BNLA) – a longitudinal study of humanitarian migrants

These studies have followed around 10,000 participants for approximately 10 years (further details, including precise study characteristics, can be found at https://www.dss.gov.au/about-the-department/national-centre-for-longitud...). It is also expected that this workshop will include a guest presentation from a DSS representative.

Course syllabus: 

Day 1

  • Study design and reporting (including Setting goals and objectives, Inclusion and exclusion criteria for participant selection, Data Management and data linkage, Reporting styles including the CONSORT statement, Ethics including privacy and confidentiality)    
  • Revision of linear, logistic and Poisson regression

 

Day 2 and Day 3

  • Mixed effects models
  • Fixed and random effects
  • Random intercept and random slope
  • Goodness of fit measures (including Likelihood and AIC)
  • Model choice
  • Having more than one clustering level

 

Day 4

  • Survival or time-to-event analysis
  • Kaplan-Meier curves
  • Survival and hazard functions

 

Day 5

  • Missing data
  • Multiple imputation
  • Heckmann Selection models

 

Course format: 

This course may run in a computer lab, or you may be advised to bring your own laptop with specified software.

We will let you know in advance.

 

Approximately half of the time during this course will be spent in PowerPoint presentations, and half of the time in computer demonstrations and self-paced computer exercises (conducted in Stata).

 

Recommended Background: 

Participants must have completed the course Fundamentals of Multiple Regression or an equivalent course at university level and/or have equivalent experience.

Familiarity with analysis of variance, factor analysis or regression is desirable, but not strictly necessary. It is assumed that participants have little or no familiarity of structural equations with latent variables.

Recommended Texts: 

The instructor's bound, book length course notes will serve as the course texts.

 

Other references include:

  • Singer J.D., Willett J.B. (2003). Applied Longitudinal Data Analysis: Modeling Change and Event Occurrence
  • Hedeker D., Gibbons R.D. (2006). Longitudinal Data Analysis
  • Diggle P., Heagerty P. (2013). Analysis of Longitudinal Data
Program where course next likely to be offered: 
Summer Program 2019
Notes: 

The instructor's bound, book length course notes will serve as the course texts.

Supported by: 

Stata is distributed in Australia and New Zealand by Survey Design and Analysis Services.