| old_uid | 20087 |
|---|
| title | Moving to a World beyond p < .05 (and BF > 3): Why and how? |
|---|
| start_date | 2022/02/21 |
|---|
| schedule | 12h15 |
|---|
| online | no |
|---|
| details | Invité par Tanja Atanasova |
|---|
| summary | Numerous authors have highlighted the limitations of the Null-Hypothesis Significance Testing (NHST) approach and the (near-exclusive) reliance on p-values and significance testing. Beyond the methodological advantages and disadvantages of NHST compared to other paradigms of statistical inference, one of the main barriers to the correct use of the NHST procedure remains its complexity, often hidden by misleading intuitive interpretation. For instance, a p-value does not indicate the probability of obtaining certain data “by chance”, nor the probability of “being wrong”. In 2019, The American Statistician published a special issue named “Statistical Inference in the 21st Century: A World Beyond p < 0.05”, with the intention to provide new recommendations for users of statistics (e.g., researchers, policymakers, journalists). This issue comprises 43 original papers aiming to provide new guidelines and practical alternatives to the “mindless” use of statistics and arbitrary thresholds. In the accompanying editorial, Wasserstein, Schirm, & Lazar (2019) summarise these recommendations in the form of the ATOM guidelines: “Accept uncertainty. Be thoughtful, open, and modest.” In this talk, we will explore some consequences of these guidelines when applied to the analysis of empirical data, in the light of core concepts from the philosophy of statistics. We will illustrate how this approach nicely fits with the recently proposed “Bayesian workflow” (Gelman et al. 2020), which includes iterative model building, model checking, model understanding, and model comparison. |
|---|
| responsibles | Atanasova |
|---|
| |