Empirical research inevitably includes constructing a data set by processing raw data into a form ready for statistical analysis. Data processing often involves choices among several reasonable ...options for excluding, transforming, and coding data. We suggest that instead of performing only one analysis, researchers could perform a multiverse analysis, which involves performing all analyses across the whole set of alternatively processed data sets corresponding to a large set of reasonable scenarios. Using an example focusing on the effect of fertility on religiosity and political attitudes, we show that analyzing a single data set can be misleading and propose a multiverse analysis as an alternative practice. A multiverse analysis offers an idea of how much the conclusions change because of arbitrary choices in data construction and gives pointers as to which choices are most consequential in the fragility of the result.
A commonly voiced concern with the Bayes factor is that, unlike many other Bayesian and non-Bayesian quantitative measures of model evaluation, it is highly sensitive to the parameter prior. This ...paper argues that, when dealing with psychological models that are quantitatively instantiated theories, being sensitive to the prior is an attractive feature of a model evaluation measure. This assertion follows from the observation that in psychological models parameters are not completely unknown, but correspond to psychological variables about which theory often exists. This theory can be formally captured in the prior range and prior distribution of the parameters, indicating which parameter values are allowed, likely, unlikely and forbidden. Because the prior is a vehicle for expressing psychological theory, it should, like the model equation, be considered as an integral part of the model. It is argued that the combined practice of building models using informative priors, and evaluating models using prior sensitive measures advances knowledge.
Subjective well-being changes over time. While the causes of these changes have been investigated extensively, few attempts have been made to capture these changes through computational modelling. ...One notable exception is the study by Rutledge et al. Rutledge, R. B., Skandali, N., Dayan, P., & Dolan, R. J. (2014). A computational and neural model of momentary subjective well-being. Proceedings of the National Academy of Sciences, 111(33), 12252-12257. https://doi.org/10.1073/pnas.1407535111, in which a model that captures momentary changes in subjective well-being was proposed. The model incorporates how an individual processes rewards and punishments in a decision context. Using this model, the authors were able to successfully explain fluctuations in subjective well-being observed in a gambling paradigm. Although Rutledge et al. reported an in-paper replication, a successful independent replication would further increase the credibility of their results. In this paper, we report a preregistered close replication of the behavioural experiment and analyses by Rutledge et al. The results of Rutledge et al. were mostly confirmed, providing further evidence for the role of rewards and punishments in subjective well-being fluctuations. Additionally, the association between personality traits and the way people process rewards and punishments was examined. No evidence for such associations was found, leaving this an open question for future research.
Most empirical papers in psychology involve statistical analyses performed on a new or existing dataset. Sometimes the robustness of a finding is demonstrated via data-analytical triangulation (e.g., ...obtaining comparable outcomes across different operationalizations of the dependent variable), but systematically considering the plethora of alternative analysis pathways is rather uncommon. However, researchers increasingly recognize the importance of establishing the robustness of a finding. The latter can be accomplished through a so-called multiverse analysis, which involves methodically examining the arbitrary choices pertaining to data processing and/or model building. In the present paper, we describe how the multiverse approach can be implemented in student research projects within psychology programs, drawing on our personal experience as instructors. Embedding a multiverse project in students’ curricula addresses an important scientific need, as studies examining the robustness or fragility of phenomena are largely lacking in psychology. Additionally, it offers students an ideal opportunity to put various statistical methods into practice, thereby also raising awareness about the abundance and consequences of arbitrary decisions in data-analytic processing. An attractive practical feature is that one can reuse existing datasets, which proves especially useful when resources are limited, or when circumstances such as the COVID-19 lockdown measures restrict data collection possibilities.
Preregistration is a method to increase research transparency by documenting research decisions on a public, third-party repository prior to any influence by data. It is becoming increasingly popular ...in all subfields of psychology and beyond. Adherence to the preregistration plan may not always be feasible and even is not necessarily desirable, but without disclosure of deviations, readers who do not carefully consult the preregistration plan might get the incorrect impression that the study was exactly conducted and reported as planned. In this paper, we have investigated adherence and disclosure of deviations for all articles published with the Preregistered badge in
Psychological Science
between February 2015 and November 2017 and shared our findings with the corresponding authors for feedback. Two out of 27 preregistered studies contained no deviations from the preregistration plan. In one study, all deviations were disclosed. Nine studies disclosed none of the deviations. We mainly observed (un)disclosed deviations from the plan regarding the reported sample size, exclusion criteria and statistical analysis. This closer look at preregistrations of the first generation reveals possible hurdles for reporting preregistered studies and provides input for future reporting guidelines. We discuss the results and possible explanations, and provide recommendations for preregistered research.
How feelings change over time is a central topic in emotion research. To study these affective fluctuations, researchers often ask participants to repeatedly indicate how they feel on a self-report ...rating scale. Despite widespread recognition that this kind of data is subject to measurement error, the extent of this error remains an open question. Complementing many daily-life studies, this study aimed to investigate this question in an experimental setting. In such a setting, multiple trials follow each other at a fast pace, forcing experimenters to use a limited number of questions to measure affect during each trial. A total of 1398 participants completed a probabilistic reward task in which they were unknowingly presented with the same string of outcomes multiple times throughout the study. This allowed us to assess the test–retest consistency of their affective responses to the rating scales under investigation. We then compared these consistencies across different types of rating scales in hopes of finding out whether a given type of scale led to a greater consistency of affective measurements. Overall, we found moderate to good consistency of the affective measurements. Surprisingly, however, we found no differences in consistency across rating scales, which suggests that the specific rating scale that is used does not influence the measurement consistency.
The credibility of scientific claims depends upon the transparency of the research products upon which they are based (e.g., study protocols, data, materials, and analysis scripts). As psychology ...navigates a period of unprecedented introspection, user-friendly tools and services that support open science have flourished. However, the plethora of decisions and choices involved can be bewildering. Here we provide a practical guide to help researchers navigate the process of preparing and sharing the products of their research (e.g., choosing a repository, preparing their research products for sharing, structuring folders, etc.). Being an open scientist means adopting a few straightforward research management practices, which lead to less error prone, reproducible research workflows. Further, this adoption can be piecemeal – each incremental step towards complete transparency adds positive value. Transparent research practices not only improve the efficiency of individual researchers, they enhance the credibility of the knowledge generated by the scientific community.
Societies invest in scientific studies to better understand the world and attempt to harness such improved understanding to address pressing societal problems. Published research, however, can be ...useful for theory or application only if it is credible. In science, a credible finding is one that has repeatedly survived risky falsification attempts. However, state-of-the-art meta-analytic approaches cannot determine the credibility of an effect because they do not account for the extent to which each included study has survived such attempted falsification. To overcome this problem, we outline a unified framework for estimating the credibility of published research by examining four fundamental falsifiability-related dimensions: (a) transparency of the methods and data, (b) reproducibility of the results when the same data-processing and analytic decisions are reapplied, (c) robustness of the results to different data-processing and analytic decisions, and (d) replicability of the effect. This framework includes a standardized workflow in which the degree to which a finding has survived scrutiny is quantified along these four facets of credibility. The framework is demonstrated by applying it to published replications in the psychology literature. Finally, we outline a Web implementation of the framework and conclude by encouraging the community of researchers to contribute to the development and crowdsourcing of this platform.
Openness is one of the central values of science. Open scientific practices such as sharing data, materials and analysis scripts alongside published articles have many benefits, including easier ...replication and extension studies, increased availability of data for theory-building and meta-analysis, and increased possibility of review and collaboration even after a paper has been published. Although modern information technology makes sharing easier than ever before, uptake of open practices had been slow. We suggest this might be in part due to a social dilemma arising from misaligned incentives and propose a specific, concrete mechanism—reviewers withholding comprehensive review—to achieve the goal of creating the expectation of open practices as a matter of scientific principle.