ABSTRACT
It is important that all tests be fair and be seen to be fair in their use. However, fairness only makes sense when viewed within a wider social and psychological perspective. Taken from an individual perspective, unfairness occurs when a wrong decision is made about an individual on the basis of a test result. Yet wrong decisions are made all the time, particularly when an individual’s score is near the cut-off. If, for example, a student needs a B grade to gain entrance to a degree course, and the examiners have decreed that a B is equivalent to 80% on the test, and the student has obtained 79%, then there must be a statistical chance of at least one in three that a mistake has been made. Now from the point of view of the examiners, who will be looking at the overall perspective, the best that can be hoped for is that mistakes will be kept to a minimum. Given the limited number of places available, and the lack of definitive knowledge of what makes for success in a course, it would be unrealistic to hope for any high degree of accuracy near the cut-off point. Yet it is unlikely that the unsuccessful student will find this argument acceptable. A consideration of this issue rapidly shows us that there exists a series of social conventions within society concerning what is considered as fair and unfair in these circumstances, and these conventions are informative as they often do not agree with conceptions of bias as viewed from the point of view of the statistician.
