Log in

Queensland Branch Meeting: Statistical Methods in Meta-Analysis with Applications in Health Sciences

  • 24 Nov 2021
  • 3:00 PM - 4:30 PM
  • Zoom

Registration is closed
Please join us online for the November Queensland Branch Meeting, on the 24th of November. The seminar will start at 15:00 (AEST), with a branch meeting starting at 16:00 (AEST). Please find details for the seminar below.


3:00 PM - 4:00 PM (AEST) Seminar

4:00 PM - 4:30 PM (AEST) Branch Meeting

Wednesday 24th November 2021

Location: Online - Zoom

Please note that the seminar will be recorded and might be put on YouTube or similar platform.

Statistical Methods in Meta-Analysis with Applications in Health Sciences

Abstract: Although originated in education (c.f. Glass, 1976), and widely used in health sciences, meta-analysis is now being used to estimate the common effect size in almost all disciplines. At the age of evidence-based decision-making meta-analysis has been increasingly used to synthesize the findings of multiple independent studies on the same topic often as part of systematic reviews and with conflicting outcomes. Meta-analysis refers to the statistical analysis of summary results from independent primary studies focused on the same intervention or treatment effect to decide on policies, or clinical practices, or programs. The main objective of meta-analysis is to find an estimate of the common effect size using data from all relevant independent studies and present the results in a forest plot representing confidence intervals along with some key summary statistics. Like many other statistical methods, heterogeneity is an important issue in meta-analysis, and hence measuring heterogeneity is an essential part of it. Different statistical models use different weighting system to estimate the common effect size and find the standard error of the estimator in the way of calculating confidence intervals. As covered in Khan (2020), this talk presents meta-analysis as a key component of systematic review, highlights selected effect size measures, discusses different statistical models, and illustrates meta-analyses using commonly used statistical models for several data sets from health sciences.

Glass G. V (1976). Primary, secondary, and meta-analysis of research. Educational Researcher. 5 (10): 3–8. doi:10.3102/0013189X005010003. S2CID 3185455.

Khan, S (2020). Meta-Analysis: Methods for Health and Experimental Studies. Springer Nature

Presenter: Prof. Shahjahan Khan, School of Sciences & Centre of Health Research, University of Southern Queensland, Toowoomba, Australia

Professor Shahjahan Khan obtained his PhD and MSc degrees in Mathematical Statistics from the University of Western Ontario (UWO), Canada. He is currently the founding Professor of Statistics at the University of Southern Queensland (USQ) which he joined in 1993. His research interests include systematic review and meta-analysis, inference with non-sample prior information, predictive inference, multivariate analysis, and linear models. He has published over 250 research papers, and presented 18 workshops, 28 keynote addresses, and over 50 invited talks in scholarly gatherings. He has supervised 16 PhD and 3 MPhil students. Professor Khan is the Founding Chief Editor of Journal of Applied Probability and Statistics (JAPS) since 2005. He recently authored a book on “Meta-Analysis for Health and Experimental Studies” published by Springer Nature, 2020. His presentation is based on some of the work from this book.

Powered by Wild Apricot Membership Software