Causality and Meta-Analysis

Our group has recently started a collaboration with Judea Pearl’s group on utilizing ideas from causal inference to develop methods for meta-analysis. This website will provide information on the results we attain in this project including the paper, software and curricular materials we develop.

This project is funded by the National Science Foundation. The information on our NSF grant is below.

NSF Proposal IIS – 1302448 III: Medium: Meta-analysis reinterpreted using causal graphs

Statistical conclusions from research studies may often be misleading due to a variety of reasons including small sample sizes for the studies or confounding factors which are unknown to the investigators of the study.  One way to reduce the possibility of misleading conclusions is to combine the results of multiple research studies using a technique referred to as “meta-analysis.”  Meta-analysis is one of the most widely used techniques to infer knowledge from data in science.  The idea behind meta-analysis studies is that the combined statistical conclusions from multiple research studies reflect the information in all of the studies and are more likely to be accurate.  The conclusions from meta-analyses are considered “better” or “more likely to generalize” than conclusions from single studies.   However, this notion is not well formalized and formalizing this question is a goal of this project.   In addition, meta-analysis methods do not take into account any knowledge of the similarities and differences between the studies.  Taking advantage of these similarities and differences can improve the effectiveness of meta-analysis.

This project takes advantage of recent developments in the area of “causal inference” which is the study inferring cause and effect relationships from data.  These types of inferences utilizes a type of graph called a causal graph which graphically represents cause and effect relationships.  This project develops an alternate framework for meta-analysis based on a novel type of causal graph, a selection graph.  A selection graph formally represents the similarities and differences between the studies.   This project provides a unifying framework for meta-analysis and leads to more powerful methodology.  The methods developed in this project are applied to genetic studies where meta-analyses have discovered thousands of variants involved in common human disease in the past few years.

Causal graphs have had a major impact on the way causality is taught and understood in cognitive science, statistics, and the health and social sciences. The proposed research promises to have similar impacts by transforming the approach to meta-analysis, one of the work horses of statistical inference in the physical, life and social sciences.  The development of the techniques presented in this proposal will be used to perform meta-analyses of genetic studies which can lead to the discovery of variation involved in disease.  The project website, http://zarlab.cs.ucla.edu/causal-meta-analysis/, contains the results of this project including information on courses which are taught by the investigators covering topics relevant to the project.