The aims of this paper are:
 

1. to introduce the student of education to the Rasch model of test calibration;
2. to discuss ways in which the model  could be put to use in the Caribbean context.

 
The treatment is non-mathematical, so that the underlying concepts may be clearly grasped by students who may not be equipped to deal with a more rigorous exposition of its statements. A fuller treatment will be found in some of the works cited in the bibliography.
 
Introduction
 
Objectivity of measurement in the behavioural sciences does not have the same connotation as in physical science. In the latter, it is assumed that the measuring device is neutral. We may have the choice of what units to express the measured quantity in; but, having decided to use a balance graduated in grammes, we do not expect our measurements of mass to depend upon which particular balance we use. We cannot make the same assumption with regard to test scores in education. A raw score is meaningless without a considerable amount of knowledge of the particular test that was used to derive it. To assign a meaning to an individual’s score we need to know the distribution of scores from other testees who have taken the test in the past. We must consider also what selection of items has gone into making up the test battery, since a different selection of items might well result in a different measurement.