Co-authored by Dr. Melissa Swisher, Lecturer, Purdue University
Khan Academy is a hugely popular nonprofit educational organization with many courses [www.khanacademy.org] and YouTube videos on subjects such as computing, science, economics, and mathematics. Recently, Sal Khan of Khan Academy adopted a mastery learning criterion [https://www.cbsnews.com/news/sal-khan-academy-mastery-learning/]: Users won’t move onto the next content area until they master with high accuracy the current content area. Khan Academy’s previous model was the traditional fixed pace of instruction and test with noncontingent content advancement. While incorporating mastery learning makes sense, it provides some practical difficulties. How could education be tailored to individual students with mastery learning?
Bloom (1984) indicated that learners who take practice tests and receive corrective feedback (see also Slowiak & Lakowske, 2017) prior to unit exams outperform learners who don’t have these practice tests prior to unit exams. This test-feedback-test approach is the one that Khan uses with Khan Academy’s mastery learning through mastery challenges [https://www.khanacademy.org/badges/challenge-accepted]. In mastery challenges, learners earn badges—gamification applied to education (see Subhash & Cudney, 2018 for a review)—when they understand the material, which no doubt leads to teachers’ anecdotes (https://youtu.be/45jzZCMdXaE) about increased student motivation and self-esteem. Learners can continue to take tests and restudy until they have mastered a unit; some learners may need to complete several tests before becoming proficient, but other students may only need to take one test.
Learners study the material and retest until they pass the unit exam.
Parallel to the mastery learning research in education, behavior analysis offers personalized systems of instruction and programmed instruction. Fred Keller developed the personalized system of instruction (see Eyre, 2007 for a review) so learners could progress at their own pace, meet learning objectives prior to moving to the next unit, and attend lectures for motivation over student-selected topics. In Keller’s system, learners write short answers on unit tests which are graded and corrected by (several) proctors—typically learners who have previously taken and passed the course. Learners continue to take exams until they reach 80-90% accuracy and gain access to lectures only once they’ve passed prerequisite material.
Online classes and learning management systems (e.g., Blackboard, Canvas, Brightspace, Google classroom) make it easier for instructors to incorporate personalized system of instruction elements (e.g., corrective feedback, administering multiple versions of exams, and self-pacing; see also Martin, Pear, & Martin, 2002). Programmed instruction is also easily adaptable to present on learning management systems. Learners can meet their stated behavioral objectives for mastery, receive high rates of reinforcement, and benefit from successive approximations to reduce errors both at home and within the classroom (e.g., Cummings & Saunders, 2018; Escobar & Lattal, 2001, Fienup, Hamelin, Reyes-Giordano, & Falcomata, 2011).
Incorporating mastery criterion should prevent learners from feeling bored or unfairly challenged in class. They’ll be able to learn the material at their own pace—with more practice when they need it and faster advancement when they don’t.