ABSTRACT
In this chapter we move on to the next stage in the test design cycle. The test specifications and sample items are subjected to close scrutiny in the early stages of design. In fact, evaluation begins in the test specification design workshops; this shows that the ‘stages’ of test design, as portrayed in Figure 4.1, are much more fluid than a diagram can suggest. If an item or task survives this initial scrutiny, a small pool of items is created and taken forward to the next stage, which is called prototyping . It is at this point that the tasks are tried out with two or three small groups of learners to see how they react to them. If it appears that the responses of the learners are similar to those predicted by the test designers, more items are written to form a pool large enough to pilot. This usually involves giving a set of items to a much larger number of learners so that statistical information can be collected. If the items are embedded in a test form that is created using an assembly specification, the piloting is usually referred to as ‘field testing’. If the tasks/items and the test assembly model appear to be working well after all these checks, it is possible to move towards making the test operational. This is the point at which everyone concerned with the test development has to make a critical ‘go no-go’ decision (Fulcher and Davidson, 2007: 89). If a ‘go’ decision is made, it is then possible to set up task/item shells so that large-scale production can begin. As items are produced they are subjected to another round of review before they are placed in the growing item pool. Once in the operational item pool, they can be used to create test forms.
