[Formal requirements for assessments-what is important?]

Assessments in undergraduate and graduate medical education are crucial for evaluating the competencies of learners. Assessments have both a control function (summative) and a feedback function (formative). This distinction is important, as the two aspects cannot easily be combined, and, therefore, it must be considered which function should be emphasized in each case.

This article examines the formal requirements and quality criteria that assessments must meet, particularly with regard to objectivity, reliability, and validity.

Referring to classical test theory, the three main criteria that are decisive for the quality of examinations are discussed. In addition, potential judgment biases and how to avoid them are outlined, as is how to improve assessment items by means of typical item characteristics.

Assessments should be independent of the examiners. Written exams, especially multiple-choice tests, offer greater objectivity than oral exams. Objective structured clinical examinations (OSCEs) increase the objectivity of practical exams. A reliable exam reflects the actual performance of the examinees and is as free as possible from random effects. Verifying the validity of an exam is an ongoing process that involves many variables. Simple forms of validity are easier to determine but are less meaningful.

The quality of exams depends heavily on whether the abovementioned quality criteria are met. Systematic analysis of exam questions can be used to improve them in a targeted manner.