Recently, I eavesdropped as my sister, a lecturer in the paramedical field, had Zoom viva voce exams (oral exams) with a succession of students. A panel of teachers admitted students one by one into a video call and asked them questions. During the exam, it seemed that some students were possibly reading answers from a website or textbook on their computer screen. Three students in, it was clear that students were sharing the examination questions with their classmates on WhatsApp, as later students seemed (somewhat) more prepared for the questions. The teachers then began changing the questions and topics for each student.
For a teacher, this illustrates the biggest concern with online assessment: ensuring academic integrity while assessing learners fairly.
For teachers, planning and grading exams and assignments is usually an exhausting process. Add the online element and the process is likely to be even more time-consuming. What can teachers do to make this process better? Your options will depend on the capabilities of the platform you use, the size of your class, the subject you teach, the requirements of your school or college, and technology and network limitations (for you and your learners).
Online Assessment Options
Among the challenges of conducting school completely online, the most complex aspect that instructors have to deal with is assessing student performance. Instructors have used different methods: timed multiple-choice exams, exams in which learners have to write the answers by hand and scan or send a picture to instructors by a certain time, video viva voce or oral exams, assessing discussion board participation, online presentations, and authentic assessments or projects.
Every method has its advantages and drawbacks. Some require more initial effort from instructors, some are more objective than others, and some are easier to reuse. The table below summarizes various options. (And the section below the table takes a deeper dive into several of these options.)
Assessment type | Grading | Instructor effort in creation | Instructor effort in grading | Objectivity | Advantages | Drawbacks | Tips |
---|---|---|---|---|---|---|---|
Multiple-choice/objective exams | Usually autograded, if the learning platform allows it | High initial effort in creating the questions and answer keys | Low or no effort (if autograded) | High | Infinitely reusable; you can reuse these questions year after year; the best option for large class sizes/MOOCs | Doesn’t test writing skills or the ability to formulate arguments or explain a point | Create questions across difficulty levels covering all the topics. |
Open-ended/essay-type questions | Manually graded | Moderate initial effort | High | Moderate | These can be typed or handwritten according to technology facilities;they allow evaluation of thought processes and writing skills | Evaluation can be time-consuming, especially if learners submit scans of handwritten essays | Avoid questions based on mere memorization and ask for learners’ perspectives or solutions. |
Video viva voce/oral exams | Manually graded | Low | High | Moderate | Helpful in gauging learner’s understanding of concepts and ability to explain them | Difficult to manage with large numbers of learners | Use rubrics for grading. Use variants of questions from the same type of content or difficulty level for different learners. |
Authentic assignments | Manually graded | High | Moderate | Moderate | Helpful in gauging learner’s application of concepts in a real setting | Evaluation can be time-consuming for large groups | Limit these to one or two a semester. |
Presentations | Manually graded | Low | Moderate | Moderate | Allow assessment of learner understanding of a topic | Time-consuming and best for small class sizes | Try to have learners teach their peers certain topics. |
Discussion board participation | Manually graded | Low | High | Moderate | Allows learners to formulate responses at their own pace | Prompts must be carefully designed to be effective;evaluation can be time-consuming | Use a rubric and very specific prompts. |
Online proctored exams (Software/proctor) | – | – | – | – | Can curb cheating to some extent | Proctor software can be cheated, while a person proctoring requires instructor time, vigilance, and effort | Given the resources involved, these are best for high-stakes exams. |
Additional Considerations
Here are some more options to consider when assessing learners online.
Assign authentic assessments: Authentic assessments require that learners perform tasks as they would in the real world. They are a direct assessment of a learner’s skills at performing the tasks, in contrast to multiple-choice or written exams that test the learner’s recall and understanding of the tasks. Using authentic assessments in subjects that allow it can be helpful to instructors. For instance, learners in a marketing course can be required to develop a marketing plan for a company, or learners in an engineering course can be required to create a design or working model. Authentic assessments somewhat reduce the risk of plagiarism and help you gauge how well learners can apply concepts.
Evaluate using rubrics: One challenge with evaluating direct or open-ended assessments is staying objective and evaluating different assignments using the same criteria. For instance, you want to be sure that the excellent design or presentation of one assignment does not overshadow the excellent content of another assignment. A useful tool here is to create a set of rubrics or parameters to be used in evaluation. A rubric states the parameters that the instructor is grading learners on (for example, design, concept, writing, originality, presentation, grammar) and what the learner needs to do to achieve each parameter. There are different styles of rubrics (some better suited to evaluation and some better suited to giving learners feedback), and you can pick the style that suits your needs and those of your students.
Use a combination of assessment types: Instead of basing the entire grade on a single assessment, consider using a number of smaller assessments of different types, for example, a multiple-choice exam and a presentation. This can give learners with different strengths a chance to ace the assessment type they are best at.
Have many shorter assessments: If possible, use shorter assessments spread throughout the semester instead of one long examination at the end. These might be easier to grade and will spread the burden of manual grading over time. In addition, performance on earlier assessments can help learners understand the areas they need to improve before the next assessment.
Use discussion board responses: If the online platform you use has a discussion board feature, you can evaluate learners’ responses to a prompt or discussions with each other. You can create a rubric to evaluate responses fairly. The prompts and the methods for evaluating discussions must be well-crafted to be most effective. Click here to read more about how to use discussion boards effectively.
Have proctored exams or use a proctoring software: You can monitor students during an exam by having all students connect to a video conference and keep their cameras on as they write. Though this method is not fool proof, it does help to curb cheating. Many companies also offer different types of proctoring software with a range of features, from blocking access to other websites while students take the exam to using artificial intelligence to verify the person writing the exam and flag suspicious behavior.
While finding the right balance for you and your learners might take a while, you should feel free to experiment with different methods and change your approach until you find what works best for you. Remember that wholly online assessment is new for a lot of teachers and that everyone is learning as they go.