Meta-analysis, interaction strength and effect size: Application of biological models to the synthesis of experimental data

Meta-analysis holds considerable promise as a rigorous, quantitative approach to the synthesis of experimental results; however, its future application and success will depend on our ability to extract suitable metrics of effect size from a broad pool of experimental studies. An appropriate metric should be interpretable in the context of the ecological theory that motivated the studies and be statistically tractable (although the latter is of secondary importance). Accomplishing these goals is a difficult task and, at a minimum, requires clarification of the meaning of "effect size" or "interaction strength", and elucidation of the links among ecological theory, mathematical models, experimental design, and the extraction of appropriate metrics from experimental data. Here we propose a series of workshops aimed at defining the future of meta-analysis in ecology, which rests on our ability to define quantitative and conceptually relevant definitions of "effect size" and apply these metrics to experimental studies conducted by different investigators, in different times, and in different places.