26 - 27 March 2018
This two-day course is designed to introduce students to the statistical techniques used in meta-analysis, the synthesis of data from different studies. It will enable participants to utilize these techniques in their own meta-analyses and facilitate the interpretation and critique of existing meta-analyses. Practical examples will be drawn from the mental health field. Upon completion of this course it is intended that students will be able to:
- Calculate the statistical tests to run a meta-analyses under different settings and assumptions
- Know how to run a variety of common meta-analyses; for single-group means, two-sample mean differences (independent and dependent), two-group binary outcomes and correlation coefficients
- Understand and be able to explain the difference between fixed-effects and random-effects meta-analyses and choose which is appropriate in relevant context
- Assess heterogeneity across studies
- Understand and calculate the following statistics data: Cohen’s d, Hedges g, risk ratio, odds ratio, risk difference, Fisher’s z, T-squared, Q, I-squared
- Use a funnel plot to assess publication bias
- Perform a meta-analysis when outcome metrics differ across studies
Students should have a basic understanding of applied statistics (e.g. confidence intervals, t-tests), experimental design and the statistical software STATA. Some understanding of random effects would be useful (such as the longitudinal and clustered data course run in previous weeks) but will be revised within the course.
Academic Lead: John Hodsoll, BRC Biostatistician
Cost: £100 (KHP discounts available)
Click here to book.