Most tests are, of course, done with books closed and notes put away, but here is an argument for why giving open-book tests makes sense during the pandemic — and even after it has ended.
This was written by Annie Murphy Paul, a writer who focuses on learning and cognition. She is a former senior editor at Psychology Today magazine and was awarded the Rosalynn Carter Fellowship for Mental Health Journalism.
Paul is the author of several books, including “The Cult of Personality Testing,” a cultural history and scientific critique of personality tests, and of “Origins,” about the science of prenatal influences, and her newest work, available in June 2021, titled “The Extended Mind: The Power of Thinking Outside the Brain.”
By Annie Murphy Paul
How can we test what students have learned — in a way that’s accurate and fair — when they’re taking those tests at home?
Last spring, when the arrival of the coronavirus pandemic led most schooling to suddenly shift online, assessment was forced to take place remotely, too. Teachers and administrators were dismayed to see evidence that some students seized the opportunity to cheat, apparently looking up answers online or in textbooks.
This fall, as end-of-semester exams approach, many students are still learning remotely. College students who did attend classes on campus were sent home for Thanksgiving break, not to return until January, February or even later. Middle schools and high schools that have not already moved to a distance-learning model may soon do so, as cases of the coronavirus rise across the country.
This time, with more time to plan, many instructors have instituted measures intended to prevent cheating. These may involve training a camera on students as they complete their tests, or disabling students’ Web browsers so that they can’t access the Internet during test administration.
But such measures come with their own problems. Some students have objected to being surveilled by webcam observers and eye-tracking algorithms; others have pointed out that anti-cheating efforts may inadvertently advantage affluent students. Programs that limit access to the Web on students’ computers, for example, won’t deter cheating by those who are rich enough to afford a second device.
There does exist an equitable, non-intrusive, low-tech method of preventing cheating, however: making exams open-book for everyone.
Educational psychologists have been researching the effects of open-book vs. closed-book tests for years. Perhaps surprisingly, many of these studies find that test format produces little difference in learning or assessment outcomes. In addition, psychologists report, open-book exams come with some distinct advantages. They generate less anxiety in students, for example — an important factor to consider at a time when the pandemic is already imperiling young people’s mental health.
Properly constructed, open-book tests can promote the development of higher-order thinking and problem-solving skills over the rote memorization of factual information. And, of course, they render moot the temptation to cheat, reducing the incidence of unethical behavior.
Most important, open-book (and open-Web) exams more closely replicate the kinds of tasks students will need to carry out after they graduate. In our lives outside of school, it’s rare that we’re asked to answer questions or solve problems using only the knowledge in our heads.
Indeed, successful performance usually involves marshaling information and insights gleaned from a number of sources — print publications, the Internet, other people — and combining these with our own mental store of facts. Real life, you could say, is an open-book test of our ability to efficiently employ a wide range of resources, both internal and external.
Some philosophers and cognitive scientists have even argued that the mind itself isn’t contained within the head, but rather extends outward into our surroundings. Resources and tools on which we depend become, effectively, parts of our thought processes — a notion called “the extended mind.”
David Chalmers, a philosophy professor at New York University, is one of the original formulators of the theory of the extended mind. “If you take the extended mind thesis seriously, then you ought to be testing the whole extended self,” Chalmers has said. “If a calculator or a computer is going to be with you, coupled with you, reliably available in the future, then it is part of your extended self, and you ought to be testing the whole extended self.”
Such philosophical justifications aside, it is true that closed-book exams appear to be superior to their open-book counterparts in two ways.
The first is that students tend to study harder for closed-book tests; they assume that open-book tests will be easier, and so they prepare for them less intensively.
The second is that the act of taking the test itself can improve long-term retention of the material that is tested, via a psychological mechanism known as “retrieval practice” (or, aptly, “the testing effect”). Each time we recall an item of information from memory, we strengthen and streamline the neural connections that support remembering that item again in the future.
These effects, while worth taking into consideration, don’t necessarily mean that closed-book tests are preferable; instead, they suggest several steps instructors can take to make open-book testing even more beneficial.
The first of these: Teachers can caution students that open-book exams are not as easy as they may assume, and allow them to experience this reality for themselves by offering practice open-book tests during class time. Managing multiple flows of information while working under time pressure to solve a challenging problem, it turns out, isn’t easy at all.
Doing all of that demands skills that most students are in need of developing, leading to a second step: Instructors can intentionally cultivate the capacities required by open-book testing. Not incidentally, these are skills that students will lean on throughout their lives — such as the ability to search for and identify relevant data, to evaluate the credibility of sources, and to assemble disparate pieces of information into a coherent argument.
Finally, teachers can help students garner the memory benefits of the testing effect even without a (closed-book) test.
Retrieval practice can be made into a regular part of online instruction — for example, by asking students to write down their answers to a question during class, and then prompting them to raise their hands when they hear their own response among a series of potential responses spoken aloud by the teacher. Those who answered the question correctly will have fixed that information more firmly in memory; those who got the question wrong will learn why they were mistaken.
The challenges of educating students during a pandemic, including the need to prevent cheating, have presented us with an opportunity to explore the advantages of open-book testing.
But even after the coronavirus recedes, we may want to keep the practice in place — perhaps as part of a hybrid model, incorporating some open-book and some closed-book assessment. Research in cognitive science has demonstrated the importance of knowing some material by heart: a set of fundamental facts about a subject, organized into a framework that supports understanding and elaboration.
Closed-book exams could be used to test whether students have internalized this basic knowledge. Open-book exams could be reserved for the assessment of the abilities associated with the extended mind: identifying, organizing and integrating information from a variety of external sources.
The coronavirus has tested students, teachers, and schools in myriad ways. How gratifying it would be to use the tough choices the virus has forced upon us to better prepare students for a future beyond the pandemic.