The selection of a test group is based on the examiner’s judgement, rather than a formal statistical method.