Redesigning Assessment in a Large Business Statistics Course
A commerce college teacher handles Business Statistics for 120 first-year students in two sections. The current 100-mark scheme is one three-hour external examination with 100 multiple-choice questions; internal assessment is 20 marks based only on attendance. Last year the mean score was 58, but the same students scored poorly in a later research-methods course where they had to choose a statistical test and interpret output. The teacher proposes a take-home assignment (20 marks), a group project (20 marks) and a reduced end-semester examination (60 marks). The principal is concerned about plagiarism in take-home work and free-riding in groups; students also complain that attendance marks are arbitrary. The academic calendar allows only two weeks for feedback before the project deadline. The teacher must present a defensible assessment redesign aligned with the intended learning outcomes: selecting tests, computing, and interpreting results.
1. Using constructive alignment, evaluate the current scheme and propose a revised 100-mark plan with justifiable weights.
Read the worked answer
The current scheme is misaligned. A 100-item MCQ examination mainly samples recall and formula recognition, while the stated outcomes require test selection, computation and interpretation. Attendance is an input measure, not evidence of learning, and it gives no reliable information about competence. A defensible redesign would use the 60-mark end-semester examination for breadth and security, but include short numerical problems and one data-interpretation item rather than only MCQs. The 20-mark individual take-home should give each student a unique dataset and require a brief justification of the chosen test, the computation and a written interpretation. The 20-mark group project should require a shared dataset, a group report and an individual oral defence or contribution note. These weights preserve summative rigour while creating direct evidence of the higher-order outcomes. The two-week feedback window means the take-home should be staged: a proposal in week one, feedback, then the final submission.
2. How should the teacher address plagiarism and free-riding without abandoning the assignments?
Read the worked answer
The teacher should design out the opportunity rather than rely only on policing. For the take-home, unique datasets, personal interpretation questions and a short in-class oral defence reduce copying and make ghost-writing less useful. Students should submit drafts or checkpoints so the teacher can see the work develop. For the group project, a group contract should specify roles, meeting logs and deliverable ownership; peer evaluation should adjust individual marks within a limited range, and each member should answer questions on the whole project. The teacher should state the plagiarism policy, use a similarity check as one signal rather than as proof, and require proper citation of software output and sources. These measures protect fairness while still assessing collaboration and communication, which are legitimate learning outcomes.
3. What evidence should the teacher collect in the first implementation to decide whether the redesign worked?
Read the worked answer
The teacher should not judge success only by a higher mean score, because easier assignments or marking leniency could inflate it. Useful evidence includes rubric scores by outcome, item analysis of the end-semester paper, the correlation between assignment performance and the later research-methods course, and the proportion of students who can justify a test choice orally. A short pre/post diagnostic on test selection and interpretation would show change in the targeted skills. Student feedback should ask whether the assignments were clear and whether group work was fairly assessed. The teacher should also record plagiarism cases, free-riding complaints and marking time. Comparing the first cohort with the previous cohort is possible, but differences in students and content must be acknowledged; the strongest evidence is whether students independently demonstrate the intended outcomes.