Assessing mathematical communications about projects and investigations

Abstract

The provision of valid evidence of learning achievement and the need for accountability are two crucial factors in assessment. Projects and investigations, which provide evidence of communication in mathematics are discussed as a means to the direct assessment of the true objectives of the curriculum meeting the criteria of Eisner (1993). Using the methodology of Griffin and Forwood (1991), developed for adult literacy and numeracy scales, research in UK universities addresses the assessment of such tasks in mathematical modelling, pure mathematics and statistics. Analysis from this research, using item response modelling and the techniques of Linacre (1990) and Adams and Khoo (1992), describes student, judge and descriptor behaviour on rating scales which attempt to assess student achievement comprehensively. This paper focuses on the example of oral communication skills as an illustration. Implications of the research are discussed and questions are raised which require further investigation.