Remote proctoring consists of monitoring the test sessions of students to deter them from gaining an unfair advantage on the test by several common cheating strategies. Common strategies are:
A test may be proctored remotely by a person, an automated system including artificial intelligence (AI), or a combination of the two.
Remote proctoring goes hand in hand with traditional assessments like multiple-choice questions (MCQ) given online. Some argue that conventional assessments aren’t needed at all, hence no need for remote proctoring. There may be valid arguments for moving toward authentic assessment, which tests for the skills as they are applied in the real world. More relevant in the humanities and social sciences, these arguments break down for the coursework in STEM fields.
Calculus and Statistics don’t lend themselves to free-form assessments like a final paper or project. There’s a vast range of concepts and techniques at every level that a student needs to master. It’s unlikely that any project or paper is going to cover all of them.
The argument that MCQs are “inauthentic” doesn’t hold because the kind of work problems commonly used in these kinds of assessments mirror closely the application of these methods in the real world. In the STEM field of computer programming, for example, “authentic assessments” consist of actual programming tasks coded and executed in a development environment. These assessments also need to be proctored to ensure that students rely on their own skill and experience and not gain an unfair advantage from either pre-knowledge of the test content or outside help during the test session.
Suppose we dismiss for the moment questions about the necessity of traditional assessments. In that case, we are left with the question of how we best administer these assessments in the context of distance learning. During the pandemic, every educational institution has had to wrestle with this question. And they will continue to do so. After the pandemic has eased and students have returned to classrooms, the momentum of distance learning will continue, and more and more higher-value degree courses and credentials will be delivered and assessed remotely. What follows is a list of the benefits of remote proctoring of assessments for Universities and their students.
How much cheating is there in academia? A lot, according to the expert, Donald McCabe, of Rutgers Business School. Whether it is 50% of students or 10%, any significant amount of cheating is corrosive to the learning environment. Students of high ability, who have little inclination or motivation to cheat, feel ripped off when students of lesser ability or work ethic cheat and get the same score on a test. Endemic cheating puts the lie to the meritocratic values extolled by universities. Invigilation of online exams has been proven to reduce cheating and make assessments more accurate.
Aggressive invigilation of tests gives students confidence that they are going to get a fair shake. Institutions can take steps to ensure academic integrity that don’t involve remote proctoring. Test design that randomizes test forms and question order and paraphrases ubiquitous test bank questions, so they are not quickly looked up online is one technique. Including an “Honor Code” of the academic policy in the syllabus for the course is another. These techniques deter a lot of potential academic dishonesty. Simple remote proctoring techniques like locking down the browser may be a marginal improvement. Students recognize, however, that there are still plenty of opportunities for mischief. A comprehensive academic integrity regime that includes active remote monitoring by AI-assisted human proctors addresses this last window and gives students confidence that the school cares about academic integrity and is making a considerable effort to uphold it.
The main benefit of remote proctoring is the same as the main benefit of remote learning in general: Convenience. Test takers don’t need to travel, and they can take tests anywhere in the world. What is mere convenience here in the developed world is access to educational opportunities for students living in areas more remote to urban centers of learning. Remote proctoring allows distance learning programs to rise in the level of rigor to those of in-person colleges.
A single parent can wait until the kids are asleep. Many students are juggling multiple responsibilities, and having to stick to a precise schedule for an exam seating at a test center may not be the most convenient. Also, as the experience with MOOCs has shown, students go at the pace that works best for them. They may do a semester’s coursework in a few weeks and be ready to test out halfway through the semester. On-demand testing with remote proctoring makes all this possible.
The difference between a certification and a certificate is an assessment. Ironically, students collaborating on exams may improve their learning outcomes. If the curriculum is designed for group work and the professor allows it, students are free to collaborate. For most exams, collaboration is not by pedagogical design and is a violation of the test rules.
For collaboration to lead to better learning outcomes, students need to interact and debate face-to-face. In the context of distance learning, collaboration looks less like a Socratic colloquium than hiring online tutors to give you answers on the sly. Test scores on poorly proctored exams may better measure students’ financial resources and ruthlessness than their genuine ability. Moreover, there is less opportunity for close collaboration in distance learning situations, so being lax about deterring these kinds of test violations puts remote and less social students at a distinct disadvantage when taking exams.
Standardized testing has been the great equalizer. College placement exams like the ACT and SAT have allowed traditionally disadvantaged students to earn a place at top-ranked schools despite lacking legacies or connections to gain admission. The tests were designed for that very purpose. (That’s why UC eliminating the SAT for admission decisions may have unintended consequences.)
Placement testing only works if everyone is sure that the system can’t be gamed. That is why potential students and their parents felt betrayed when the Varsity Blues testing scandal hit the news. It is a case where students who had every advantage, from money to social connections, violated the integrity of the standardized tests to get unearned access. Remote proctoring is necessary to make sure that those tests are fair.
Using traditional assessments with remote proctoring helps to remove human bias from the equation. It is possible to control for human bias in high-stakes testing. Still, it can be expensive to have multiple redundant readers of final papers, and even then, the precision of the rubric or formulas graders use may be suspect. Standardized testing with remote proctoring, on the other hand, lends itself to precise psychometrics that allows the creation of multiple forms and items with the same approximate level of difficulty. With automatic scoring on multiple-choice questions and remote proctoring to ensure that no one is gaining an unfair advantage, it is easier to measure the validity of test instruments.
When students and faculty know that it is very rare to get away with academic dishonesty, people’s scores on assessments have much more value. Employers, customers, and the whole of our modern economy rely on the prestige and reputation of educational institutions to guarantee the quality and character of graduates.
Technology has made it possible to educate and train millions of people all over the world. MOOCs and online programs of major universities offer an opportunity to people in the farthest-flung corners of the world to reach their potential through education and training. When it comes to an assessment of that training, cost should not be a barrier to climbing that final rung on the ladder. Manual test processes and physical test centers can add expenses. Staffing up graduate student instructors to read lengthy papers and projects and manage uniform grading also raises costs. The best remote proctoring with artificial intelligence costs a fraction of live proctoring and offers much more measurable consistency than the same kind of assessment with reader grades.
When you add it all up, the benefits of remote proctoring are overwhelming. There is an epidemic of cheating that, if left unchecked, will continuously erode confidence in our educational institutions. That does not diminish the valid critiques of the implementation of remote proctoring in practice. These concerns, however, are more questions of how the technology is used rather than the technology itself. Questions of student rights and privacy can be addressed by appropriate policies and corporate governance of the proctoring technology companies. Questions of algorithmic bias in the systems themselves can be addressed by technical solutions that ensure strict human supervision of automated systems.
Paying close attention to the student experience in taking exams points to simple, pragmatic solutions for minimizing needless intrusion by remote proctoring systems. Engaging with students also offers the opportunity to uncover the motivations, triggers, and behavioral economics that encourage students to fall short of academic integrity. We can then design solutions, technological and not, that help them stay on the straight and narrow path of academic excellence.
We are excited to announce some major updates to Rosalyn that will change the way universities conduct online exams. We have identified a need for more control over settings for any given exam based on customer feedback, and we are happy to announce that our new exam rules, web resources and calculator features have satisfied that need.
Learn how four students rate their online exam experience using the most popular proctoring platforms.
As educators and certifying organizations increase their reliance on remote testing, students’ voicing of concerns about privacy and the intrusiveness of the technology is reaching a crescendo. Ultimately, the issue is about much more than protecting the privacy of test-takers’ confidential information.