whatever the level of the standards employed by a given superior there is a possibility that all subordinates will be rated within a narrow range. For one reason or another , the difference between the best and the worst employee on a particular scale often turns out to be minimal , even though , much larger differences do in fact exist. In small units it is not unheard of to find the whole group clustered around one point on the scale. Errors of this kind tend in large part to obviate the value of the evaluations. This clustering is commonly toward the middle of a scale , and it is this particular tendency that is implied by the error of central tendency. The rater does not place anyone very far from the midpoint, in either a positive or a negative direction