Predictive Validity of the "Ready of Not" System for the Assessment of Students Needing Remediation in Music Theory

Abstract

A perennial problem in the teaching of music theory is how to identify students who are insufficiently prepared for the harmony/ear-training sequence. If it is a logistical problem - how to convene, grade, and advise such a throng before registration, it is also a problem of assessment - what skills and knowledge should one test, how exactly should a student proficient with the major key signatures (but not minor) be advised, and how poorly must one perform to be steered onto the remedial route? To solve the first problem, some instructors have devised computerized versions of paper-and-pencil tests. While such assessments have the advantage of walk-in administration, online grading and advising, they do not ordinarily employ the more powerful computations of so-called "expert" systems: non-linear and branching pathways, algorithmic computation of probabilities, content shaped by in-progress assessment, random generation of problems, databases, empirically informed cutoffs or norm-referenced advisement.