Why can’t we be more idiographic in our research? It is the individual organism that is the principle unit of analysis in the science of psychology (Barlow & Nock, 2009).

Why can’t we be more idiographic in our research? It is the individual organism that is the principle unit of analysis in the science of psychology (Barlow & Nock, 2009).

As put by Gelman et al. (2013, page 148): ‘because a probability model can fail to reflect the process that generated the data in any number of ways, posterior predictive *p*-values can be computed for a variety of test quantities in order to evaluate more than one possible model failure’.

What is the difference between the errors and the residuals ? What does it mean for a model to *predict* something ? What is a link function ? In the current post, we use four R functions (viz., the `predict`

, `fitted`

, `residuals`

and `simulate`

functions) to illustrate the mechanisms and assumptions of the generalized linear model.

In the current post, we present and compare three methods of obtaning an estimation of the ICC in multilevel logistic regression models.

I have strong doubts that someone reading a blog called *barely significant* in this troubled period might have escaped the saga of the summer. However, as this story relates closely to the name of this blog and to the motivations behind its creation, I could not help myself to write a summary of this saga (or of its beginning, at least), to inform any unfortunate people that might have missed it.

This post continues our exploration of the logistic regression model by extending it to a multilevel logistic regression model, using the brms package.

This post aims to assess the average probability of participant presence in psychological experiments and, in the meantime, to introduce Bayesian logistic regression using R and the rethinking package.

In this post I will argue that the “match point situation” (i.e., the situation in which your entire life is dictated by chance only), is an everyday’ situation for the researcher who uses *Null Hypothesis Significance Testing* (NHST) without controlling power.

According to Rubin (1984), a Bayesianly justifiable analysis is one that “treats known values as observed values of random variables, treats unknown values as unobserved random variables, and calculates the conditional distribution of unknowns given knowns and model specifications using Bayes’ theorem”