Inter-rater reliability is always less than ideal because each rater has a different set of exemplars of the variables being assessed. Exemplar categorization is the process of comparing an instance of something with an easily identified example in memory.