Source-linked AI summary
Conceptualizing and Evaluating Replication Across Domains of Behavioral Research
Jennifer L. Tackett, Blakeley B. McShane
TL;DR
The paper challenges ZELD’s direct-versus-conceptual replication dichotomy and dichotomous evaluation methods. It proposes broader replication concepts for psychological domains and hierarchical/meta-analytic models that incorporate heterogeneity and related evidence, accepting uncertainty to support greater learning.
Problem
ZELD’s replication definitions leave direct replications’ critical elements unspecified, make nearly all behavioral replications conceptual, and exclude hybrid designs and domains where large-scale prospective studies are impractical.
Method
The paper proposes broader replication conceptualization, retrospective use of shared archival data, hierarchical/meta-analytic models, and holistic evaluation incorporating contextual variability, methods, prior evidence, plausibility, design, and data quality.
Results
The proposed approach generally avoids labeling replications successes or failures, instead accepting uncertainty and variation while enabling more learning about the world.
Takeaways & Limitations
Replication evaluation should generalize beyond fast, inexpensive domains and replace binary judgments with analyses that account for heterogeneity and broader evidence.
Abstract
from arXiv · showhide
We discuss the authors' conceptualization of replication, in particular the false dichotomy of direct versus conceptual replication intrinsic to it, and suggest a broader one that better generalizes to other domains of psychological research. We also discuss their approach to the evaluation of replication results and suggest moving beyond their dichotomous statistical paradigms and employing hierarchical / meta-analytic statistical models.