2 Validity

This section of the paper is premised on the definition of validity as being the warrants for the inferences that users will make in the light of the information that the process produces. That is, that the assessment tool does not merely provide valid evidence for that which the achievement standard describes, but for that which the end user (whether the student, an employer, an educational institution or society at large) believes it to describe. Professor Black emphasises that traditional academic and vocational assessments are profoundly unsatisfactory in this area. He postulates that this is because skill learning is context specific, while assessment works with evidence from only a limited number of contexts.

As the paper notes, both those designing the achievement standards and those devising the assessment tasks will need to keep the complexity of the possible inferences in mind. The key issue here is not to promise more than the qualification is able to deliver. As with any assessment, the demonstration of learning occurs at a specific time (or during a specific period), and under specific circumstances.

Having determined the area of each achievement standard, panels have then considered the better site of assessment - external or internal. Professor Black suggests that it may in some cases be necessary to consider a third option - that of including both for the same achievement standard. While this may be desirable in some circumstances, further consideration of workload and cost factors have lead to this option not being taken.