META-REP 2024: Workshops
Please sign in for your preferred workshop.
Note: Workshop capacities are limited. We will contact you after the conference registration deadline to confirm (or disconfirm) your participation in the workshop. Workshop attendance requires conference registration.
| Workshop 1 | Monday, October 28, 9:00 - 16:00, Konferenzraum | |
| Abel Brodeur (University of Ottawa) | ||
| Replication Games Participants will be matched with other researchers working in the same field. Each team will work on reproducing a recently published study in a leading social science journal (e.g., Psychological Science and Nature Human Behaviour). All participants will be given co-authorship to a meta paper aggregating the reproduction/replications results. The event will be from 9am to 4pm. To participate, please register. This event is part of games organized all over the world by the Institute for Replication (https://i4replication.org/). All researchers (e.g., graduate students, post-doc, faculty) are welcome. |
||
| Workshop 2 | Monday, October 28, 9:00 - 16:00, Viereckhof EG | |
| Cassie Ann Short & Daniel Kristanto (Carl von Ossietzky Universität Oldenburg) | ||
| Multiverse Analysis: From Theory to Practical Implementation in R Participants: Scientists from any discipline who apply statistical models in the framework of regression analysis, without prior experience with multiverse analysis implementations. Prerequisites: Basic statistics (including GLM) and statistical programming skills in R are required. Participants are required to bring their laptops with pre-installed R packages, as specified in the pre-workshop materials. Abstract: Throughout the study design, data pre-processing and data analysis workflow, multiple defensible alternative options are available for selection, described by Gelman and Loken (2013) as the 'garden of forking paths'. In a traditional analysis approach, researchers select one workflow of defensible options, thereby constructing one dataset, and often do not explicitly disclose the decision-making process. However, this creates a multiple comparison problem, as a large variety of workflows are theoretically possible. Multiverse analysis, as proposed by Steegen et al. (2016), addresses this issue by performing repeated hypothesis testing across a multitude of datasets derived from different defensible decisions - such as variable selection, categorization, and outlier management. After combining all identified defensible choices and eliminating contradictory combinations, datasets are generated and submitted to statistical analyses. Results provide as many outcomes of statistical tests as datasets created in the multiverse, which can be visualized to understand how statistical conclusions depend on methodological decisions, or they can be statistically integrated. This workshop will guide attendees through both the theory and application of multiverse analysis using R Software for Statistical Computing. Participants will engage in practical exercises to apply these concepts, learning to analyze, integrate, and visualize outcomes from multiple analytical paths. |
||
| Workshop 3 | Monday, October 28, 9:00 - 16:00, Studienbibliothek | |
| Jan Kustermann & Ruben Brück (ZPID Trier) | ||
| FAIR Data Management Research and science greatly benefit from access to high-quality, curated, open research data. However, scientific publications have to fulfil various criteria so that they can be labelled as such. This requires comprehensive research data management. Therefore, research data should be made findable, accessible, interoperable, and reusable. But what does this ‘FAIRfication’ mean for everyday research, and what measures result from it? In this workshop, we will take a closer look at the topics of FAIR data documentation, data management plans (DMP), and meta-analysis tools. We will provide brief insights into the content and combine this with the presentation and application of helpful tools to support participants in their day-to-day research. |
||