Taking into account sources of between-study heterogeneity is most simply done by modeling the effects of study membership for each participant directly in the model. As described earlier, fixed-effects IDA treats the study membership of participants as a fixed and known characteristic of each individual observation nested within a given study. Analytic techniques associated with this approach are straightforward; we incorporate one of several available coding schemes (e.g., dummy codes, effect codes, weighted effect codes) to denote study membership as a fixed characteristic of each individual observation (as we would gender or ethnicity) and enter these dummy- or effect-coded variables as predictors in our fitted models in a way consistent with Fisher's model-based inferential approach described earlier. A key advantage of this strategy is that we can also estimate multiplicative interactions between individual characteristics (e.g., gender, ethnicity) and study membership. This in turn allows us to test the differential impact of individual characteristics on outcomes across the set of studies.
