This image shows an examination question from the GNVQ (intermediate level) 2D Visual Language paper, September 1997 (Edexcel 1997: 7).

The question asks, 'What BEST describes the marks made with this pencil?’ The instructions given to students on the front of the exam paper state that, '[E]ach question has FOUR possible answers; a, b, c and d. Only ONE is correct' (Edexcel 1997: 7). The one correct answer is somewhere among the choices given; even black shading, dark-textured shadows, lines of variable thickness, or, very fine lines. But surely the marks that the (linear representation of a drawing) implement might make are dependent on a range of factors. For example; the use to which the instrument is being applied, the grade of lead that might be in the 'pencil', the pressure being applied by the person holding it and their desire and concept?

There cannot be only one correct answer to this question or to the others like it. They present the struggle that examiners have in trying to objectively measure learning in art education. As a result the questions asked in the exam paper need to follow a more ambiguous format relying upon semi-auxiliary verbs for support, for example, ‘What BEST describes...?’ or, ‘Which medium is MOST LIKELY to...?’ (Edexcel 1997: 7). The question I present offers an example of the tension-ridden conflicts that lie between art education and the audit culture. The contradictions existed before the paper was compiled and the people who devised and approved the question simply materialised the problems by attempting to measure a subject area in which wrong could work out to be right and visa versa. The presentation of the question minimises risks because the tick box format does not permit critical responses from students and questions always relate to auditable norms as being correct.