Introduction

What is the solution for secure online exams? The current technological solutions are to lock down the browser or to use advanced surveillance software to observe students while they write. The locked browser has little value when people have multiple devices, and the proctoring surveillance solutions require cameras and bandwidth that many African students cannot afford. The solution may work, but it is prohibitively expensive at present. Is there a better way? I have always loved the story about how the American space programme spent millions of dollars trying to invent a pen that could write in a zero-gravity environment; the Russians just used a pencil. The Russian pencil for online exams is to make them open-book tests. Why waste vast amounts of time and money trying to proctor secure closed-book exams when open-book exams are as effective without the security challenges?

While most online institutions are looking for ways to ensure that their students cannot cheat in closed-book online exams, I argue that the best solution for higher education is to replace closed-book papers with open-book testing because it has been shown to be as effective educationally while significantly reducing the cost and risk of cheating. Turning from closed- to open-book testing spares online institutions the headache of proctoring online exams, while posing no risk to the educational quality of their courses. I will show this to be the case by a brief survey of a small sample of the recent literature on the advantages and disadvantages of open-book testing.

Advantages of open-book testing

Although there are a few exceptions, most studies of the relative educational merits of open-book versus closed-book testing have found similar results (Rummer, Schweppe, and Schwede 2019). The two testing methods have been found to be equally effective, with closed-book exams more suitable for testing knowledge and recall, while open-book exams are better for testing higher-order thinking. In the literature surveyed, I found nine advantages of open-book tests and exams.

  1. There is less cheating. Since students are allowed to consult study materials and resources, there is no need to ensure that they are not consulting resources during the test (Senkova et al. 2018). They can still cheat by collaborating with other students or consulting experts, so this does not solve all forms of dishonesty. However, there are two practical ways to reduce cheating further. Firstly, draw random questions from question pools to make it more difficult for students to collaborate. Secondly, to discourage students from consulting experts or having third parties take their tests for them, use multiple assessments, and do not weight the final assessment too heavily. It is more difficult to get someone else to take many small assessments for you than to have someone help with one large, final exam. Switching to open-book testing lies at the heart of the Russian-pencil solution for writing in space—it is not worth trying to prevent copying on closed exams when we can test application on open ones.
  2. They are less stressful. Students find open-book testing much less stressful than closed-book exams. Because of the stress they feel, many students do not do themselves justice at traditional examinations. Dicarlo (2009) argues that they are a major contributor to student burnout. Therefore, reducing the number of closed-book exams will improve students’ resilience, resulting in better completion rates (Teodorczuk, Frazer, and Rogers 2018).
  3. They are more authentic. In real-life practice, engineers don’t need to remember their formulas verbatim and doctors don’t need to memorise every diagnostic and treatment. Teodorczuk, Frazer, and Rogers (2018) argue that it is unprofessional to rely on fallible human memory to recall facts in high-stakes real-life situations. Open-book testing allows students to consult their sources for facts, testing their ability to apply them to solve realistic problems. This kind of testing presents a much closer approximation of the real-world use of knowledge.
  4. They focus on higher-level outcomes. Traditional examinations tend to focus on the lower levels of Bloom’s cognitive taxonomy—remembering facts and concepts. Since the testing values recall, students study accordingly. I have yet to meet a tertiary educator who aspires to teach primarily for recall. We all long to prioritise the higher levels, such as applying, analysing, evaluating, and creating. Open-book testing is much better suited to this aspiration than closed-book methods. When students have their study materials open, there is no point asking them to list, name, or state things. The questions have to be framed to call for higher-order thinking (Deneen 2020; Dicarlo 2009; Teodorczuk, Frazer, and Rogers 2018).
  5. They require better designed questions. This can be a positive or a negative. Closed-book testing lends itself to simple, factual questions. These kinds of questions are quick and easy to write, but they are pointless in open-book contexts. Formulating challenging, thought-provoking questions for open-book exams takes more time and skill. The downside is that the examiner may not do a good job of it, resulting in an unsuitable test. However, if the examiner does a respectable job, the questions will be better designed (Hughes 2021).
  6. They produce equivalent long-term recall. This was a surprise finding in the study by Senkova et al. (2018). Their study measured the impact of open- and closed-book testing on students’ long-term retention of knowledge outcomes, expecting to find that closed-book methods would fair best here because students prioritise memorising facts when studying for closed-book tests. Why would they not do significantly better on later tests of the facts? There are two reasons. Firstly, if students study for the closed-book test using massed repetition instead of spaced repetition, the forgetting curve predicts that they will remember little after a few weeks. Secondly, desirable difficulty predicts that students will retain more of the lower-level information if forced to engage in higher-order thinking tasks. Thus, a welcome spin-off of open exams with more complex questions is better long-term retention of the foundational knowledge.
  7. The testing effect works equally well. What is “the testing effect”? It is the well-established principle that students’ understanding and retention of their work improves more by regular testing than by re-studying it. The associated pedagogy is to make regular use of low-stakes quizzes and tests. The question is whether those low-stakes, formative quizzes need to be closed-book tests? The answer is a resounding, “No!” If the goal is to help students retain knowledge by using multiple low-stakes quizzes, the testing effect works as well when they are open-book quizzes as when they are closed (Senkova et al. 2018).
  8. They reduce commission errors. Commission errors occur when a student is asked a question and gives an incorrect answer. Research shows that if they do not receive prompt feedback to correct the error, they are likely to encode the wrong response and reproduce the error on later tests. Open-book testing reduces the incidence of commission errors because students can search for the right answers (Senkova et al. 2018).
  9. The free space in crowded curricula. This is the argument that Teodorczuk, Frazer, and Rogers (2018) Lamenting the fact that the medical school curriculum is overloaded with information that students need to commit to memory, they argue that switching to open-book exams would free time and space for concentrating more on higher-order outcomes. This problem is not limited to medical schools. I cannot think of a curriculum that is not overloaded with content, even my son’s grade 6 curriculum has far too much information crammed into it.

There is no good reason to be concerned about the educational legitimacy of open-book testing. If they are well developed, they provide a rigorous means of testing. For online programmes, they avert many of the security concerns that plague closed-book exams.

 

Short Bio: Kevin, who is the Principal of SATS, obtained his first doctorate from Stellenbosch University and his Ph.D. from SATS. He has a deeply insightful approach to theology and has already made a significant contribution in his relatively short career.