Usage and applications of meta-analysis techniques
June 24 @ 11:00 am - 5:00 pm$250
A 6-hour workshop taught by Mariola Moeyaert, Ph.D.Register
Meta-analysis is used to: combine evidence across different research studies; integrate multiple studies into a single statistical framework; yield more precise estimates of effect sizes; allow for unique treatment comparisons; explain differences arising from conflicting study results; and identify areas for future research. Meta-analytic methods are employed in a diverse array of disciplines across the social, behavioral, health, and agricultural sciences.
This workshop provides hands-on exposure to the process involved in conducting a meta-analysis from the planning stage, through the selection of appropriate statistical techniques, through the issues involved in analyzing data, to the interpretation of results. Examples and case studies from the social sciences will be integrated into the discussions and lectures.
Note: Lecture examples and demonstrations will rely on free software: Excel, the R Statistical Computing Environment, and specific software for meta-analysis (RevMan and CMA). RevMan is completely free and CMA has a free trial version. Meta-analysis techniques are also available and easily accessible in other major software packages (e.g., Stata, SAS, and SPSS). The instructor can provide the students with Stata, SAS, and SPSS syntax upon request.
During the workshop, participants will learn how to calculate various kinds of effect sizes and to use them to conduct and make appropriate inferences from meta-analyses. Participants will have the opportunity to work on a mini meta-analysis project, using data provided by the instructor. Using the data set “Mobile digital technology”, they can practice the steps involved in conducting a meta-analysis. The participants will end with synthesizing studies’ estimates of one of the standardized mean differences (Hedges’ g) using a fixed effect and random effects model.
Pricing and schedule
Time: Wednesday, June 24, 11AM to 5PM (EST)
We offer $25 graduate student and multiple workshop discounts. Find out about our discounts here.
The main workshop material is scheduled for six hours. Time permitting, Dr. Moeyaert will answer questions about participants’ specific research projects. Participants can ask questions via chat, microphone, or telephone. In order to allow sufficient time for questions, the number of workshop participants is limited to 30.
Who should attend?
The course assumes a basic knowledge of statistics and linear regression analysis. Researchers interested in conducting a meta-analysis.
- Introduction to meta-analysis
– Examples of meta-analyses, highlighting the importance of the technique
– Overview meta-analytic datasets
– Statistical Software
- Part 1 of the meta-analysis process – Study retrieval
Step 1. Formulating meta-analytic research questions
Step 2. Planning of the meta-analysis (protocol)
Step 3. Systematic search for studies
Step 4. Selection of relevant studies using
Step 5. Data extraction
- Part 2 of the meta-analysis process – Data analysis
Step 6. Effect size calculation
Step 7. Fixed effect versus random effect meta-analysis
Step 8. Moderator analysis (subgroup analysis and meta-regression)
- Part 3 of the meta-analysis process – Discussing and reporting the results
Step 9. Interpretation of the results
Step 10. Reporting the results
All the 10 steps will be illustrated using free meta-analysis software packages in RStudio – RevMan – CMA (10 days free trials) and Excel.
About the instructor
Dr. Mariola Moeyaert is an Associate Professor Statistics within the division of Educational Psychology and Methodology at UAlbany. She received her PhD in Educational Statistics from the Katholieke Universiteit Leuven (KU Leuven) in Belgium and worked as a Postdoctoral fellow at the Center of Advanced Study in Education at the City University of New York. Her major research interests and publications are in the field of multilevel analysis, meta-analysis and interrupted time series analysis. She is also interested in Bayesian statistics, bootstrapping, structural equation modeling, longitudinal data analysis, and international comparative research. She (co)authored about 47 international publications, in such journals as Psychological Methods, Multivariate Behavior Research, Journal of School Psychology, Behavior Research Methods, School Psychology Quarterly, etc., reporting about developments in research methodology (including several computing intensive Monte Carlo simulation studies) and about applications of statistical models on educational/medical data. Currently, she is involved as PI and Co-PI in research projects funded by the Institute of Education Sciences about methodological advancements in the field of multilevel meta-analysis. You can find more information about Dr. Mariola Moeyaert by following these two links (1) https://www.