A new version of dexter has just been submitted to CRAN. It was something of an urgent affair because of unexpected changes to one of the packages on which dexter depends. However, there are numerous smaller fixes and improvements, and some new functions, e.g. probability_to_pass. A new vignette explains how to equate a pass-fail score with data missing by design and without relying explicitly on unobserved variables. We suspect that the ability to compute sensitivity, specificity, and a ROC curve under such conditions may be of interest beyond the field of psychometrics.