Top 5 Myths Of Online, Remote Test Proctoring
Chinnapong/Shutterstock.com

Online Test Proctoring May Not Do What You Think

Test proctoring, observing someone taking a test, has been a hot topic since thousands of colleges had to move millions of students online to grapple with the COVID-19 pandemic. Most of those students were taking classes and tests online for the first time and under unusual and extraordinary circumstances, leading to questions and concerns about online testing and the proctoring process. Unfortunately, myths and assumptions came along with those questions. And, as is frequently the case, those assumptions and myths have been often repeated but seldom corrected.

As someone who’s written extensively about online learning, testing, and academic integrity, here are some of the bigger myths about online proctoring and what’s actually going on instead.

Myth 1: Online Proctoring Is An Invasion Of Privacy  

The truth is that if you’re taking a test in college, someone is probably going to be watching. Usually, that’s a professor or graduate student teaching assistant. Sometimes, proctoring can be done in a test center either on campus or somewhere locally. But someone is always watching to ensure the person taking the test is actually the student and that the test rules are followed.

Naturally, when a student takes a test online, their professor can’t be there. And they’re not in a test center. When a test is online, the only alternative is remote proctoring, using technology such as webcams to watch and record a test session to be sure no one is disadvantaged.

Contrary to popular myth and considering the alternatives, online proctoring is actually less of a privacy invasion than taking the same test in person. Online, students can pick where they test—in their bedrooms, at a kitchen table, in front of a blank wall, just about wherever they want. At a test center, students have no similar agency. And that does not even count the other students or strangers who may be watching too. All things considered, taking an online test with a remote proctor is far more private than most any other option.

Myth 2: It’s Surveillance, Snooping, Invasive Technology

It’s not possible to review or verify every testing system used by every proctoring provider. Still, the systems used by three or four largest, most well-known proctoring companies are similar enough to dispel the rumor that what they do is invasive or surveillance.

Test proctoring software requires the permission of the test-taker at every step of installation and use. Proctors are trained to explain what they are doing and get permission from students to, for example, disable their web browser during a test or lock “cut and paste” functions. Everything a test proctor does before, during, and after a test is visible to the student and recorded for review. When a test session is done, the software is easily uninstalled or deleted just as can be done with any other software. Most proctoring companies recommend doing exactly that.

In all cases, observation and recording of test sessions only happen during the test session. There are no known cases of proctors even attempting to access student records or property outside a testing window, which seems technologically improbable anyway. In other words, the software used by proctoring providers works the same way as the software used by a help desk or IT department or when employees work remotely—it offers temporary, limited, and highly visible remote access during a specific, verifiable session.

Myth 3: AI And “False Flags” Accusing Honest Students Of Cheating

Most online proctoring and security platforms use some form of review system beyond simply watching a test session. They may include sophisticated systems that can monitor behaviors of the test-taker as well as what’s happening on the student’s computer—recording keystrokes or analyzing background noises, as examples.

But however they are configured, those tools do not accuse students of cheating, as many people believe. Instead, when these systems detect something unusual, the best proctoring systems then require humans to review the video and the event—and usually two or more humans—before a student is accused of misconduct. Further, one of the people who reviews a test incident should be the student’s professor or other leader at their school, presumably people who know the student, the test, and the material best and know the difference between normal testing behavior and cheating.

The technologies that proctoring platforms use do not ever determine who is or who is not cheating, people do, and usually professors. If a student is “flagged” by a technology, professors and trained proctors intercede to separate routine conduct like a sneeze from an attempt to cheat. If a school is using non-human technology by itself to levy cheating accusations or impose penalties, they are using it wrong.

 Myth 4: Cheating Is Rare, Proctors Treat Students Like Suspected Criminals 

It may not actually be a myth that cheating is rare since many people, and especially students, do understand how common cheating is. Recent academic studies have used words such as “widespread” and “commonplace.” One paper found, “Students in online courses have the highest tendency to cheat, with more than 70% admitting to cheating.” In August, I wrote in the Washington Post about the dramatic spike in cheating brought on by the necessity of remote learning during COVID-19.

The myth here is that, even though cheating is common, remote test proctors want to catch students cheating, suspect all students, or are seeking “gotcha” moments. The truth is that test proctors are far more motivated to deter cheating than to detect it. The reason is simple. A cheating incident makes more work—long, complicated, expensive work. Moreover, cheating is embarrassing to everyone involved.

Proctoring companies work for schools and schools absolutely do not want to catch cheaters. Catching and punishing cheaters involves paperwork, meetings, reviewing evidence, appeals, and potential legal challenges. It’s a pain. What schools want is to dissuade and avoid cheating—stopping cheating before it starts. Accordingly, proctors want that too. They’d much rather remind a student to put away a cell phone or study notes than “flag” them for cheating.

Myth 5: Data Security And Vast Commercial Troves Of Testing Data

While it’s true that everything that happens online creates data and that no online system can ever be completely bulletproof, it’s a myth that student testing data is shared, for sale, or that it even exists to be sold or shared.

The key to untangling this myth is understanding that test data, student data from testing, belongs to the school giving the test, not the proctoring company. That means that the commercial enterprises that run the tests or provide the testing platform can’t share it or sell it. It does not belong to them. Instead, schools set the policies about how and for how long any test data is saved. Some schools delete it within days. Others keep it longer in case it’s needed for review within a program or in the case of a challenge or appeal of some kind.

Moreover, selling or sharing student data is nearly universally prohibited. In the United States and Europe, federal or international law prohibit it outright. There is not a single case of a company or school selling or sharing student test information for commercial purposes. With all that, or perhaps in spite of all that, the myths of online proctoring are powerful and have triggered students to petition their schools to end online proctoring or allow students to “opt-out” of it. And unfortunately, the idea that students can “opt-out” is just as much mythology as AI determining cheating.

The thought that some students would take a supervised test with enforced rules while others are allowed not to is academically untenable, to say nothing of deeply unfair. In reality, the only alternative to proctoring an exam is not proctoring an exam. And according to recent research, non-proctored testing not only increases the rate of cheating, it actually invites cheating. One peer-reviewed paper found, “that when a test is not proctored, students perceive cheating as more acceptable and are more likely to cheat or commit test fraud.” Further, it found that “any inaction on the part of the faculty to provide a secure exam administration was seen [by students] as an indication that the faculty did not care about test security or cheating.”

Everyone should be able to agree that encouraging cheating is a bad idea and that proctoring, at a minimum, deters it. That ought to be good enough.

Educators can, of course, give exams in person or send students to supervised test centers or change the way they structure their assessments so cheating is less linked to success. But those options are limited or impractical in many online learning environments. So long as there are online tests designed to measure acquired skills or knowledge in a direct, question-and-solution way, online proctoring is essential. And because it is, we should set aside some of the misguided mythology that's sprung up around it.

Close