Each Reader Decides If a Replication Counts; Reply to Schwarz and Clore (2016)
Year: 2016
Type: preprint
Abstract: I agree with Schwarz & Clore on the importance of considering differences between original and replication studies when interpreting replication failures. I disagree on the proposition that without manipulation checks replications cannot be statistically analyzed as such, and disagree on their approach to considering hypotheses for why a replication failed. They stop after generating hypotheses, I show the benefits of also testing them. I propose a heuristic for deciding which design differences replicators should highlight in their articles. A unifying theme is that replicators are communicating with all readers, not just the original authors; the goal is informing the former rather than persuading the latter.
Source: SSRN Electronic Journal
Author: Uri Simonsohn
Institution: University of Pennsylvania
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W3123687901