Informative hypothesis testing in the EffectLiteR framework: A tutorial

In this paper, we illustrate how the typical workflow in analyzing psychological data, including analysis of variance and null hypothesis significance testing, may fail to bridge the gap between research questions and statistical procedures. It fails, because it does not provide us with the quantities of interest, which are often average and conditional effects, and it is insufficient, because it does not take the expectations of the researcher about these quantities into account. Using a running example, we demonstrate that the EffectLiteR framework as well as informative hypothesis testing are more suitable to narrow the gap between research questions and statistical procedures. Furthermore, we provide two empirical data examples, one in the context of linear regression and one in the context of the generalized linear model, to further illustrate the use of informative hypothesis testing in the EffectLiteR framework. peerReviewed publishedVersion