Research Thesis Topic
Statistical Meta-Analysis with applications in health and social sciences
This is a statistical method to combine data from several independent studies conducted using randomised control trails or for making inferences on effect measures or outcome variables. Analyses are done for relative risks and odd ratios for binary data, and weighted mean difference, using precision as weight, for continuous variables. Both classical and Bayesian approaches can be used. Forest plots and funnel plots are used to study the outcome variables. Issues such as study bias and heterogeneity of outcome measures are required to be handled properly. Although initially used in clinical studies involving randomised control trials, the methods are now being used in many areas of education, criminology, psychology, pharmacy, and business. The combination of data from independent studies is likely to provide better quality of inference due to increased sample size. The results of studies involving binary outcomes are generalised for ordinal categorical data. Methods for ordinal categorical data are also investigated. Analyses of data under various popular statistical models such as the fixed effect, random effects, inverse variance heterogeneity and quality effect models are explored. This project area provides students with a good opportunity to mix theoretical developments with real life applications using various medical and epidemiological data. Issues such as heterogeneity, publication bias, network meta-analysis, multivariate meta-analysis may also be investigated.
- Centre for Health Sciences Research
- School of Agricultural and Environmental Sciences
- Public Health and Health Services
- Statistics
- Doctor of Philosophy (DPHD)
- Master of Research (MRES)
Please review the admission requirements for the academic program associated with this Thesis Topic