Recent developments in setting and using objective tests in mathematics using QM Perception
2006-05-26T09:40:38Z (GMT) by
This short paper builds upon work described at the last CAA Conference, Greenhow & Gill (2004), in setting objective tests in various areas of mathematics using Question Mark Perception. Current activities continue to exploit the QML language and template files, coupled with MathML mathematics mark-up and the Scalable Vector Graphics (SVG) syntax for producing diagrams. There are many advantages to using such mark-up languages, primarily the use of random parameters at runtime that thereby produce dynamic equations, distracters, feedback and diagrams. An unlooked for, but welcome, advantage, is that one can also resize and recolour these elements by reading the preferences that have been set up in a user-defined cookie. This means that “reasonable provision” for disabled students as required by the SENDA legislation, is built-in. The MathML and SVG technology can be exported to any web-based system, or indeed ordinary web pages that can provide an inexhaustible set of realisations at the click of the reload button. Being central to the display of mathematics on the web, MathML’s WebEQ applet has recently been considerably extended to include graphing of MathML expressions, naturalistic input of equations with syntax checking and math-action tags. These math-action tags can be used to define a specific part of an equation, and mouse actions can then be acted upon, for example to provide a commentary on that part of the equation, toggle to another equation (perhaps a derivation of the tagged term or similar) or, possibly, to set a variable that can be used for marking (as in a hot spot question). The first part of this paper will show how these new facilities can be input into new question types for effective questions and feedback design. It is clear that much useful technology already exists, but setting effective questions that benefit students’ learning requires equal attention to their content and pedagogy. The second part of this paper looks at a possible methodology for setting much more advanced questions than hitherto, looking closely at an example from the ordinary differential equations section of Mathletics. The third part of this paper looks at a series of experiments with a first year mechanics group at Brunel University, as part of the Formative Assessment and Feedback (FAST) project. Students’ reactions were studied, especially the effect of the feedback on their subsequent behaviour when faced with similar/dissimilar questions after a variable time delay. Students spent a lot of time and energy considering the feedback provided, sometimes copying it down or printing it out. Somewhat surprisingly, it seems that a “learning resource” has actually been written, whose formative nature is of equal or more importance than the assessment function originally intended. It can be concluded that plentiful formative feedback is of great importance in the students’ ability to learn mathematics from the tests, rather than simply get their grades or marks in an efficient manner.