Search

Researchers have become increasingly aware that data-analysis decisions affect results. Here, we examine this issue systematically for multinomial processing tree (MPT) models, a popular class of cognitive models for categorical data. Specifically, we examine the robustness of MPT model parameter estimates that arise from two important decisions: the level of data aggregation (complete-pooling, no-pooling, or partial-pooling) and the statistical framework (frequentist or Bayesian). These decisions span a multiverse of estimation methods. We synthesized the data from 13,956 participants (164 published data sets) with a meta-analytic strategy and analyzed the magnitude of divergence between estimation methods for the parameters of nine popular MPT models in psychology (e.g., process-dissociation, source monitoring). We further examined moderators as potential sources of divergence. We found that the absolute divergence between estimation methods was small on average (

People judge repeated statements as more true than new ones. This illusory truth effect is a robust phenomenon when statements are ambiguous and plausible. However, previous studies provided conflicting evidence on whether repetition also affects truth judgments for highly implausible statements. Given the lack of a formal theory explaining the interaction between repetition and plausibility on the illusory truth effect, it is important to develop a formal model to explicitly represent the assumptions regarding this phenomenon. In this study, we develop a Bayesian cognitive model that builds on the simulation-based model by Fazio, Rand, and Pennycook (2019). Thereby, we formalize how repetition and plausibility jointly influence the illusory truth effect in light of nonlinear transformations of binary truth judgments. We test our model using experimental data from two previous studies by computing Bayes factors for four competing model variants. Our findings vary across studies but indicate that the observed interaction of repetition and plausibility may be explained by a constant, additive effect of repetition at a latent probit scale.

Objective: Among basic personality traits, Honesty‐Humility yields the most consistent, negative link with dishonest behavior. The theoretical conceptualization of Honesty‐Humility, however, suggests a potential boundary condition of this relation, namely, when lying is prosocial. We therefore tested the hypothesis that the association between Honesty‐Humility and dishonesty weakens once lying benefits someone else, particularly so if this other is needy. Methods: In two online studies (Study 1: N = 775 in Germany; Study 2: N = 737 in the UK, preregistered), we measured self‐reported Honesty‐Humility and dishonest behavior in incentivized cheating paradigms in which the beneficiary of participants' dishonesty was either the participants themselves, a "non‐needy" other (e.g., another participant), or a "needy" other (e.g., a charity). Results: We found support for the robustness of the negative association between Honesty‐Humility and dishonesty, even if lying was prosocial. Conclusion: Individuals high in Honesty‐Humility largely prioritize honesty, even if there is a strong moral imperative to lie; those low in Honesty‐Humility, by contrast, tend to lie habitually and thus even if they themselves do not directly profit monetarily. This suggests that (un)truthfulness may be an absolute rather than a relative aspect of Honesty‐Humility, although further systematic tests of this proposition are needed. [ABSTRACT FROM AUTHOR]

Bayesian t tests have become increasingly popular alternatives to null-hypothesis significance testing (NHST) in psychological research. In contrast to NHST, they allow for the quantification of evidence in favor of the null hypothesis and for optional stopping. A major drawback of Bayesian t tests, however, is that error probabilities of statistical decisions remain uncontrolled. Previous approaches in the literature to remedy this problem require time-consuming simulations to calibrate decision thresholds. In this article, we propose a sequential probability ratio test that combines Bayesian t tests with simple decision criteria developed by AbrahamWald in 1947. We discuss this sequential procedure, which we call Waldian t test, in the context of three recently proposed specifications of Bayesian t tests. Waldian t tests preserve the key idea of Bayesian t tests by assuming a distribution for the effect size under the alternative hypothesis. At the same time, they control expected frequentist error probabilities, with the nominal Type I and Type II error probabilities serving as upper bounds to the actual expected error rates under the specified statistical models. Thus, Waldian t tests are fully justified from both a Bayesian and a frequentist point of view. We highlight the relationship between Bayesian and frequentist error probabilities and critically discuss the implications of conventional stopping criteria for sequential Bayesian t tests. Finally, we provide a user-friendly web application that implements the proposed procedure for interested researchers. Bayesian t tests have become increasingly popular in psychological research. In contrast to classical test procedures, Bayesian tests can measure statistical evidence in favor of the null hypothesis and allow for optional stopping. Yet, probabilities of statistical decision errors (i.e., falsely rejecting a hypothesis when it is true) are not explicitly controlled. In this article, we propose a sequential test procedure where Bayesian t tests are calculated repeatedly after each additional observation. The sample size is increased until the test exceeds a predefined threshold. We call the proposed procedure Waldian t test because it is a straightforward combination of Bayesian t tests with Abraham Wald's sequential probability ratio test. We illustrate the procedure in the context of three different types of default and informed Bayesian t tests, and show how it satisfies both frequentist (i.e., controlling error probabilities) and Bayesian (i.e., measuring statistical evidence) desiderata. We also highlight the relationship between frequentist and Bayesian error probabilities and critically discuss the implications of conventional stopping criteria for sequential Bayesian t tests. Finally, we provide a userfriendly web application that implements Waldian t tests for interested researchers. [ABSTRACT FROM AUTHOR]