Norm and Criterion Referencing

Norm Referencing

The main purpose behind the use of norm-referenced assessment is to rank the performance of students in a particular group in order to generate a final grade. It involves fitting of the marks of students to a normal distribution (or 'bell curve') using a qualitative or statistical technique, allowing the assessor to re-interpret students' marks after the assessment has taken place (James, 2002). The resulting normal distribution is then used to allocate grades to students based on a pre-determined proportion of grades. This means that the quantity of each grade awarded to students can be standardised from year to year.

The method of distributed grades in this fashion is based on the assumption that the performance of student groups is always roughly the same and can therefore be validly placed in such a normal distribution. This is an assumption that has been refuted by many authors (Bond, 1996; James, 2002, Shacklock & Morgan, 2000), yet despite this, norm referencing is still a very popular method of determining student grades in both secondary and higher education.

Another criticism of the norm referencing method is that as the quantity of each grade that may be awarded is pre-set which does not adequately cater for student groups where there is either a very high or very low level of performance. A student that is awarded the highest grade in a course is assumed to have performed to a high standard, however, if the class as a whole had a poor level of performance, a person with the highest grade may not have performed as well as students awarded that same grade in previous years. Conversely, if a group of students has a very high level of performance, the norm referencing method means that only the same number of students as in previous years receive the top grade even though more students may have 'deserved' this grade on the basis of their outstanding knowledge and skills.

The criterion referencing method involves...