Multilevel Multivariate Meta-Analysis With Application to Choice Overload

Contemporary psychological research can be dizzying in its complexity, and this complexity results in patterns of variation and covariation among the observations from a set of studies that requires careful treatment in meta-analysis. For example, individual studies in a given domain can vary considerably in terms of their dependent measures and moderators; examine multiple conditions that result from the experimental manipulation of those moderators and give rise to multiple dependent effects of interest (e.g., simple effects and interaction effects); employ a mix of study designs (e.g., unmoderated versus moderated, between-subjects versus within-subjects, univariate versus multivariate); and feature different contexts, treatment manipulations, and measurement scales. Further, individual papers feature multiple studies that, while different, are quite similar particularly in comparison to studies featured in other papers.

Citation:

Blakeley B. McShane and Ulf Bockenholt (2017) ,"Multilevel Multivariate Meta-Analysis With Application to Choice Overload", in LA - Latin American Advances in Consumer Research Volume 4, eds. Enrique P. Becerra, Ravindra Chitturi, and Maria Cecilia Henriquez Daza and Juan Carlos Londoño Roldan, Duluth, MN : Association for Consumer Research, Pages: 1-4.