Workshop in Mannheim (Germany), supported by Psycholux and Globalpark: Meta-Analysis

Psycholux is a non-commercial association, dedicated to the exchange
between science and practice in the area of marketing research
methods, consumer psychology, and advertising communication. Psycholux
initiates and facilitates this exchange by organizing workshops held
by highly respected scholars and practitioners in their respective
areas. Future topics covered include, for instance, Web-based data
collection methods, usability issues, current trends in consumer
decision making, and advanced data analysis issues, to name a few.

The first workshop supported by Psycholux and Globalpark will be held
in the near future (February 5-7, 2008) in Mannheim (Germany) on
meta-analysis, a technique to quantitatively synthesize and analyze
the outcomes of empirical studies:

http://www.psycholux.de/index.php?content=events&eventid=1000

The course will be given by Prof. Dr. Wolfgang Viechtbauer (University
of Maastricht, The Netherlands), one of the most highly distinguished
researchers in the field.

Meta-Analysis Workshop: Full event description

Researchers trying to summarize the constantly growing body of
research are increasingly using techniques summarized under the rubric
´meta-analysis´. Meta-analysis provides an entire set of statistical
methods for aggregating and comparing the results from several
thematically related studies. The focus of this course will be on
current methods and techniques of analyzing meta-analytic data.

We will start out with a short overview of the entire meta-analytic
process (consisting of five steps: problem formulation, data
collection, data evaluation, analysis and interpretation, and
presentation of results). Next, we will examine how the results from a
study can be expressed succinctly with various effect size measures.
We will then delve into fixed- and random/mixed-effects models for
combining effect size estimates and for examining whether the effect
sizes depend on one or more moderator variables. As part of this
discussion, we will also touch on the general linear mixed-effects
model, since the meta-analytic models are essentially special cases
thereof.

A major problem that may distort the results of a meta-analysis is
publication bias (the fact that the published literature may not be
representative of all the research that has been conducted on a
particular topic). Therefore, current methods for detecting and
dealing with publication bias will be discussed next. Finally, time
permitting (and depending on the interests of the participants), we
will examine methods for conducting sensitivity analyses, missing data
issues, sequential/cumulative methods in the context of meta-analysis,
meta-analytic techniques using individual subject data, methods for
dealing with dependent/correlated outcomes, and Bayesian approaches to
meta-analysis.