Atomized feedback, an approach to automatic formative assessment

Abstract

In this article, we describe an approach to give automatic formative evaluation by atomizing the feedback. The mechanics in this approach is a large number of multiple-choice tests. The multiplechoice tests are used as voluntary assignments and gives controlled progression by releasing the tests in an order. Instruction is in the spirit of flipped classroom with short video lectures accompanied by one to several above-mentioned assignments. The preliminary results show reduced dropout, higher student satisfaction, and less fail on final exam. The background for this approach submerged from a multi-campus/online course with 300+ students. The aim was to create the same environment for learning for all categories of students, on campus, remote campus and online students. Atomized formative evaluation is achieved by giving feedback down to each individual multiple-choice answer. The multiple-choice questions are in the nature of practical tasks where students are expected to use external tools, calculations and simulation to find answers. The multiple-choice test collects the students’ findings. The number of correct answer gives a summative evaluation. After turning in the test, answers are available together with feedback for each answer. Correct answers explain why this is correct and a wrong answer explains why this is wrong in addition to how and why the student should go by finding the correct answer. The questions are designed with no “dummy” answer, so when choosing a wrong answer, it is a plausible answer, and an explanation for this is given as feedback. The goal is to encourage the student to try again both to improve the summative score and to fill the gap in knowledge. When forming the feedback, one must consider both the type of question and the choices made available. Using atomized feedback improves the formative evaluation greatly by being instant, continuous, to the point and informative. This is evaluation for learning that both guides and encourage the student. In addition, since instructors/TA do not use time to evaluate results on assignments, they can guide / tutor individual students who need follow up. Student performance is evaluated by achieved grade, student satisfaction, progression, and the number of assignments compared to grade. The preliminary findings compared with earlier student performance in the same course but without atomized formative feedback show a reduced number of dropouts and reduced number of students failing the final exam.