Introduction to Meta-Analysis using Stata

Tuesday, 28 May 2019, 10am to 4pm, Burwood Corporate Centre

This full day workshop aims to introduce participants to Meta-Analysis using Stata. In this hands-on workshop, using health related datasets, participants will be introduced to Stata’s metan features and learn the basics of conducting a meta-analysis. This course will introduce the theory and concepts behind a meta-analysis, while allowing for practical time during the day to apply these concepts to datasets using Stata.

The concepts we will introduce are:

  • What are the differences between fixed-effect and random-effects analyses?
  • What are the types of heterogeneity that we need to be aware of?
  • What is publication bias and how can we check?
  • How to interpret the output from a meta-analysis?
  • What are Forest Plots and Funnel plots how are they used?
  • What are possible criticisms of a meta-analysis?

In the practical sessions, using health related datasets, participants will be introduced to Stata’s user written programs: metan 

There will be a strong emphasis on study designs and interpretation of Meta-Analysis results from:

  • Fixed-effect and random-effects models
  • Forest Plots
  • Tests of statistical heterogeneity involving Q and checking I2
  • Funnel plots – publication bias
  • Trim and fill method

The main objectives of this workshop are to increase the likelihood of upskilling those who are attending so that they are able to:

  • decide if a meta-analysis is suitable and what approach to use after a systematic review of the literature,
  • understand what aspects of the approach they need to seek further assistance/knowledge on, and
  • interpret a published meta-analysis with all of the above concepts in mind.

Prerequisites

No prior knowledge of conducting a Meta-Analysis is expected. However, prior knowledge and understanding of systematic reviews and the following statistical concepts is essential:  types of study designs, power and sample size, effect sizes, standard errors and confidence intervals.

Participants are encouraged to bring their own laptop with Stata installed. Those students using Stata on Demand are also encouraged to attend. Guidance on how to install the Stata packages for metan will be provided in the week prior to the workshop.

Who should attend from the Faculty of Health?

PhD students, PhD supervisors and early career researchers with an interest or possible need to conduct a Meta-Analysis and are not sure where to start.

Presented by

Dr Steve Bowe, PhD, Senior Researcher - Biostatistics

Dr Patrick Owen, PhD, School of Exercise and Nutrition Sciences

Register Here

Please note: courses may be subject to cancellation if insufficient number of participants have registered.

Last updated:
Page custodian: Faculty of Health