Assessing Theoretical Conclusions With Blinded Inference to Investigate a Potential Inference Crisis

Abstract

Scientific advances across a range of disciplines hinge on the ability to make inferences about unobservable theoretical entities on the basis of empirical data patterns. Accurate inferences rely on both discovering valid, replicable data patterns and accurately interpreting those patterns in terms of their implications for theoretical constructs. The replication crisis in science has led to widespread efforts to improve the reliability of research findings, but comparatively little attention has been devoted to the validity of inferences based on those findings. Using an example from cognitive psychology, we demonstrate a blinded-inference paradigm for assessing the quality of theoretical inferences from data. Our results reveal substantial variability in experts’ judgments on the very same data, hinting at a possible inference crisis.

Consensus-based guidance for conducting and reporting multi-analyst studies

“…They found that no two teams followed the same data preprocessing steps or analysis strategies, resulting in substantial variability in their conclusions. This and other initiatives (9)(10)(11)(12)(13)(14)(15)(16)(17)(18) demonstrate that the multi-analyst approach is able to explore how research findings vary depending on the applied analytic strategy.…”

“…Before the expert consensus process, the first three authors and the last author (henceforth: proposers) created an initial multi-analyst guidance draft after brainstorming and reviewing all the previously published multi-analyst-type projects they were aware of (9)(10)(11)(12)(13)(14)(15)(16)(17)(18). This initial document is available here: https://osf.io/kv8jt/…”

“…The proposers contacted 81 experts to join the project. The contacted experts included all the organisers of the previous known multi-analyst projects at the time (9)(10)(11)(12)(13)(14)(15)(16)(17)(18), as well as experts on research methodology. Of the 81 experts, 50 accepted the invitation and participated in the expert consensus procedure (their names are available at https://osf.io/fwqvp/).…”

“…There are alternative solutions to prevent researchers from being biased by their data and results. For example, co-analysts could be provided with blinded datasets (14,16,28), simulated datasets (24), or with a subset of the data (e.g., 11).…”

“…The analytical robustness of results and conclusions can be studied by evaluating many distinct analysis options simultaneously (e.g., vibration of effects (7) or multiverse analysis ( 8)) or by involving multiple analysts who independently analyse the same data (9)(10)(11)(12)(13)(14)(15)(16)(17)(18). Rather than exhaustively evaluating all plausible analyses, the multi-analyst method examines analytical choices that are deemed most appropriate by independent analysts.…”