
Interpreting learning progress using assessment scores: what is there to gain?
Author(s) -
Nathan Zoanetti
Publication year - 2021
Language(s) - English
Resource type - Conference proceedings
DOI - 10.37517/978-1-74286-638-3_17
Subject(s) - information gain , trajectory , computer science , term (time) , test (biology) , artificial intelligence , machine learning , psychology , cognitive psychology , paleontology , physics , quantum mechanics , astronomy , biology
Using assessment scores to quantify gains and growth trajectories for individuals and groups can provide a valuable lens on learning progress for all students. This paper summarises some commonly observed patterns of progress and illustrates these using data from ACER’s Progressive Achievement Test (PAT) assessments. While growth trajectory measurement requires scores for the same individuals over at least three but preferably more occasions, scores from only two occasions are naturally more readily available. The difference between two successive scores is usually referred to as gain. Some common approaches and pitfalls when interpreting individual student gain data are illustrated. It is concluded that pairs of consecutive scores are best considered as part of a longer-term trajectory of learning progress, and that caveated gain information might at best play a peripheral role until additional scores are available for individuals. This review is part of a larger program of research to inform future reporting developments at ACER.