Validating national curriculum indicators
The eight stages are likened to eight links of a chain, with weakness of any one link weakening the chain as a whole.
Further guidance is offered through the identification of several possible validity threats associated with each link. A model of educational assessment for use in the validation and planning of assessments (Crooks et al.1996 p268) The importance of the eight links in the model can be illustrated by mentioning for each link just one example of the threats to validity associated with that link.
It complements the BERA paper by Mick Quinlan and Alex Scharaschkin on National Curriculum Testing: Problems and Practicalities.Validation will only flourish if approaches are developed which help to organise our thinking about important validation questions and to identify issues which need particularly close scrutiny (Shepard, 1993) In practical terms the unitary approach makes it difficult to know where to start reviewing the validity of something as complex as national curriculum assessment.This difficulty has been addressed by Crooks, Kane & Cohen (1996).For the purpose of this paper I will focus on the high profile subjects and key stages: English, mathematics and science at key stages 2 and 3.
This focus parallels the accompanying review by Mick Quinlan and Alex Scharaschkin of the reliability of the national tests in these subjects at these key stages. There was a time when validity is the extent to which a test measures what it purports to measure would pass as a definition.
The form of reporting at other key stages is not prescribed.