If cheating is optimisation, then assessment must not be pure

Abstract

(This is a paper published in Learning Letters in 2024, on modelling evidence of authenticity across a curriculum.) One way of viewing cheating is from the perspective of optimisation: producing a result without performing the work. This creates an interesting parallel with computer programming. Programmers want their compilers to optimise their code, invisibly deviating from the instructions to get the same answer with less effort. In assessment, we want to know students did the work. So, we can borrow some techniques from computer programming (“effect tracking”) to model what evidence of learning we expect across a curriculum. i.e. we want to see “Course and Effect”.