The Support Interval

This post summarizes Wagenmakers, E.-J., Gronau, Q. F., Dablander, F., & Etz, A. (in press). The support interval. Erkenntnis. Preprint available on PsyArXiv: https://psyarxiv.com/zwnxb/

Abstract

A frequentist confidence interval can be constructed by inverting a hypothesis test, such that the interval contains only parameter values that would not have been rejected by the test. We show how a similar definition can be employed to construct a Bayesian support interval. Consistent with Carnap’s theory of corroboration, the support interval contains only parameter values that receive at least some minimum amount of support from the data. The support interval is not subject to Lindley’s paradox and provides an evidence-based perspective on inference that differs from the belief-based perspective that forms the basis of the standard Bayesian credible interval.
(more…)

Preprint: A Cautionary Note on Estimating Effect Size

This post is a teaser for van den Bergh, D., Haaf, J. M., Ly, A., Rouder, J. N., & Wagenmakers, E.-J. (2019). A cautionary note on estimating effect size. Preprint available on PsyArXiv: https://psyarxiv.com/h6pr8/

Abstract

“An increasingly popular approach to statistical inference is to focus on the estimation of effect size while ignoring the null hypothesis that the effect is absent. We demonstrate how this common “null hypothesis neglect” may result in effect size estimates that are overly optimistic. The overestimation can be avoided by incorporating the plausibility of the null hypothesis into the estimation process through a “spike-and-slab model”.”

(more…)

Compensatory Control and Religious Beliefs: A Registered Replication Report Across Two Countries

This post is an extended synopsis of Hoogeveen, S., Wagenmakers, E.-J., Kay, A. C., & van Elk, M. (in press). Compensatory Control and Religious Beliefs: A Registered Replication Report Across Two Countries. Comprehensive Results in Social Psychology. https://doi.org/10.1080/
23743603.2019.1684821

Abstract

Compensatory Control Theory (CCT) suggests that religious belief systems provide an external source of control that can substitute a perceived lack of personal control. In a seminal paper, it was experimentally demonstrated that a threat to personal control increases endorsement of the existence of a controlling God. In the current registered report, we conducted a high-powered (N = 829) direct replication of this effect, using samples from the Netherlands and the United States (US). Our results show moderate to strong evidence for the absence of an experimental effect across both countries: belief in a controlling God did not increase after a threat compared to an affirmation of personal control. In a complimentary preregistered analysis, an inverse relation between general feelings of personal control and belief in a controlling God was found in the US, but not in the Netherlands. We discuss potential reasons for the replication failure of the experimental effect and cultural mechanisms explaining the cross-country difference in the correlational effect. Together, our findings suggest that experimental manipulations of control may be ineffective in shifting belief in God, but that individual differences in the experience of control may be related to religious beliefs in a way that is consistent with CCT.
(more…)

What Makes Science Transparent? A Consensus-Based Checklist

This post is a synopsis of Aczel et al. (2019). A consensus-based transparency checklist. Nature Human Behaviour. Open Access: https://www.nature.com/articles/s41562-019-0772-6.
The associated Shiny app is at http://www.shinyapps.org/apps/
TransparencyChecklist/.

 
How can social scientists make their work more transparent? Sixty-three editors and open science advocates reached consensus on this topic and created a checklist to help authors document various transparency-related aspects of their work.
(more…)