The Future of Online Proctoring

December 4, 2021

It’s impossible to imagine how universities could have navigated the mass shift to online learning during the COVID-19 pandemic without online proctoring services. But what’s the role of these services in a post-crisis world? What are the remote proctoring trends that are here to stay? 

You can’t discuss the future of online proctoring without also considering the limitations of proctoring software. How have and how will schools respond to privacy concerns? Or to accusations that services are plagued with false positives and biases against vulnerable students? 

The convenience and flexibility of online proctoring ensure it has a place in academia. To figure out exactly where it fits, you need to know how far it’s come already and what challenges schools still face in implementing it.

Recent Developments in Online Proctoring

Academic dishonesty remains a huge concern, and online learning programs have traditionally carried additional risks when compared to in-person testing. Remote and unmonitored students could easily look up information or pass their computer off to a friend for the duration of the test.

Over the past two years, the software for remote learning has continually grown more accurate and more sophisticated to respond to these threats. Major remote proctoring trends include initiatives that have made services more secure and more efficient. 

Across the industry, lockdown browsers have helped schools to control the test-taking environment. These browsers prevent students from accessing external resources on their computers during tests. They also keep students from copying test content for later distribution.

Online proctoring has also become more convenient. New tools and templates have enabled schools to schedule exams with ease and send important reminders and updates to students.

Improved Cheating Detection Methods

But the most noteworthy trends in online proctoring concern cheating detection. While remote viewing and exam recording have long been available, the latest software uses artificial intelligence to help prevent academic dishonesty.

AI proctoring includes the following features:

  • Facial recognition to verify identity throughout the exam
  • Voice recognition to detect sounds that indicate cheating
  • Pattern recognition to detect statistical anomalies or deviations from previous performances
  • Object recognition to identify unauthorized people or objects in the environment
  • Eye movement detection to detect indications that someone is directing their attention elsewhere

These advances make it extremely difficult for students to cheat during online assessments.

Online Proctoring Controversies

The surge in popularity of online exam proctoring led to a corresponding backlash of the exams. Critics of proctoring services have raised valid concerns about both the software and its implementation.  

While some industry leaders are taking measures to address these issues, it’s still crucial to understand the limitations of these services. Only then can you use them effectively and fairly. 

Student Privacy

Critics have targeted the amount of student data that online proctoring services collect and then store — sometimes for years as a point of concern. This can leave students vulnerable, particularly if security is substandard. 

Higher education institutions need to vet the security measures taken by the proctoring services they employ. You also need to perform your own assessments of the risks of academic dishonesty and the impacts on student privacy in various situations. Do all that you can to minimize both. 

You should outline a plan that takes these considerations into account:

  • How long footage needs to be stored
  • When certain features should be enabled or disabled
  • Whether individual assessments merit the associated costs in student privacy

Be transparent with your decisions. Make sure that both students and faculty fully understand them, and demand a similar transparency from the vendors with whom you work.

AI-Only Proctoring 

Machine learning simply cannot replace human oversight — at least not yet. For effective online proctoring, you need trained people who review AI flags and exam feeds. 

Earlier this year, ProctorU decided to cancel one of its packages, a proctoring service in which AI-generated results and video were sent to the subscribing institution without being reviewed by ProctorU staff beforehand. 

Why cancel one of its most popular options? While its low price made it attractive, the company discovered that it was often used in ways that were, at best, ineffective. At worst, the proctoring could be unfair.

ProctorU discovered that only 10% of faculty members reviewed the video of exams. Its service was more often used as a deterrent, the metaphorical equivalent of the fake cameras sometimes found in convenience stores.

Overburdened faculty members simply don’t have the time to review hours of exam footage. But this lack of review creates a system that is unfair to students. 

The sensitive-by-design software creates a huge number of false positives. This can result from ordinary behavior (a quick glance at the ceiling while deep in thought) or random events (an internet crash or other crisis).

People are better equipped to tell the difference between honest and dishonest behavior. In one instance, an online proctor was able to intervene in favor of a student whose cat jumped on her computer, submitting her exam before she was finished. 

More troublingly, AI proctoring often penalizes vulnerable students unfairly. For example, proctoring services routinely hand out higher “suspicion scores” to students of color than to white students. The software suffers from the same racial bias that plagues face recognition technology elsewhere. 

Educators should take advantage of AI tools but should always leave final decisions in human hands.

The Future of Online Proctoring in Remote Learning

Online proctoring isn’t going anywhere. In fact, many schools now integrate proctoring software into the online platforms their students already use, building them into their educational infrastructure. 

And when properly used, these services significantly cut down on instances of academic dishonesty. They can even help address student inequities by making tests easier to access and allowing for greater variation in student living conditions and schedules.

Make your school’s programs more flexible and more accessible with the best in online education technology. It’s both the smart thing to do as a business and the right thing to do as an institution. 

But take measures to ensure that you’re using these resources to their full effect and without unintended consequences. The CIP blog regularly covers the tips and trends that educators need to know. We’re here to help. 

Are you an educator looking for turn-key curriculum solutions for your health science course?
Get a free demo

RELATED STORIES

The Future of Online Proctoring

January 19, 2021

It’s impossible to imagine how universities could have navigated the mass shift to online learning during the COVID-19 pandemic without online proctoring services. But what’s the role of these services in a post-crisis world? What are the remote proctoring trends that are here to stay? 

You can’t discuss the future of online proctoring without also considering the limitations of proctoring software. How have and how will schools respond to privacy concerns? Or to accusations that services are plagued with false positives and biases against vulnerable students? 

The convenience and flexibility of online proctoring ensure it has a place in academia. To figure out exactly where it fits, you need to know how far it’s come already and what challenges schools still face in implementing it.

Recent Developments in Online Proctoring

Academic dishonesty remains a huge concern, and online learning programs have traditionally carried additional risks when compared to in-person testing. Remote and unmonitored students could easily look up information or pass their computer off to a friend for the duration of the test.

Over the past two years, the software for remote learning has continually grown more accurate and more sophisticated to respond to these threats. Major remote proctoring trends include initiatives that have made services more secure and more efficient. 

Across the industry, lockdown browsers have helped schools to control the test-taking environment. These browsers prevent students from accessing external resources on their computers during tests. They also keep students from copying test content for later distribution.

Online proctoring has also become more convenient. New tools and templates have enabled schools to schedule exams with ease and send important reminders and updates to students.

Improved Cheating Detection Methods

But the most noteworthy trends in online proctoring concern cheating detection. While remote viewing and exam recording have long been available, the latest software uses artificial intelligence to help prevent academic dishonesty.

AI proctoring includes the following features:

  • Facial recognition to verify identity throughout the exam
  • Voice recognition to detect sounds that indicate cheating
  • Pattern recognition to detect statistical anomalies or deviations from previous performances
  • Object recognition to identify unauthorized people or objects in the environment
  • Eye movement detection to detect indications that someone is directing their attention elsewhere

These advances make it extremely difficult for students to cheat during online assessments.

Online Proctoring Controversies

The surge in popularity of online exam proctoring led to a corresponding backlash of the exams. Critics of proctoring services have raised valid concerns about both the software and its implementation.  

While some industry leaders are taking measures to address these issues, it’s still crucial to understand the limitations of these services. Only then can you use them effectively and fairly. 

Student Privacy

Critics have targeted the amount of student data that online proctoring services collect and then store — sometimes for years as a point of concern. This can leave students vulnerable, particularly if security is substandard. 

Higher education institutions need to vet the security measures taken by the proctoring services they employ. You also need to perform your own assessments of the risks of academic dishonesty and the impacts on student privacy in various situations. Do all that you can to minimize both. 

You should outline a plan that takes these considerations into account:

  • How long footage needs to be stored
  • When certain features should be enabled or disabled
  • Whether individual assessments merit the associated costs in student privacy

Be transparent with your decisions. Make sure that both students and faculty fully understand them, and demand a similar transparency from the vendors with whom you work.

AI-Only Proctoring 

Machine learning simply cannot replace human oversight — at least not yet. For effective online proctoring, you need trained people who review AI flags and exam feeds. 

Earlier this year, ProctorU decided to cancel one of its packages, a proctoring service in which AI-generated results and video were sent to the subscribing institution without being reviewed by ProctorU staff beforehand. 

Why cancel one of its most popular options? While its low price made it attractive, the company discovered that it was often used in ways that were, at best, ineffective. At worst, the proctoring could be unfair.

ProctorU discovered that only 10% of faculty members reviewed the video of exams. Its service was more often used as a deterrent, the metaphorical equivalent of the fake cameras sometimes found in convenience stores.

Overburdened faculty members simply don’t have the time to review hours of exam footage. But this lack of review creates a system that is unfair to students. 

The sensitive-by-design software creates a huge number of false positives. This can result from ordinary behavior (a quick glance at the ceiling while deep in thought) or random events (an internet crash or other crisis).

People are better equipped to tell the difference between honest and dishonest behavior. In one instance, an online proctor was able to intervene in favor of a student whose cat jumped on her computer, submitting her exam before she was finished. 

More troublingly, AI proctoring often penalizes vulnerable students unfairly. For example, proctoring services routinely hand out higher “suspicion scores” to students of color than to white students. The software suffers from the same racial bias that plagues face recognition technology elsewhere. 

Educators should take advantage of AI tools but should always leave final decisions in human hands.

The Future of Online Proctoring in Remote Learning

Online proctoring isn’t going anywhere. In fact, many schools now integrate proctoring software into the online platforms their students already use, building them into their educational infrastructure. 

And when properly used, these services significantly cut down on instances of academic dishonesty. They can even help address student inequities by making tests easier to access and allowing for greater variation in student living conditions and schedules.

Make your school’s programs more flexible and more accessible with the best in online education technology. It’s both the smart thing to do as a business and the right thing to do as an institution. 

But take measures to ensure that you’re using these resources to their full effect and without unintended consequences. The CIP blog regularly covers the tips and trends that educators need to know. We’re here to help. 

Are you an educator looking for turn-key curriculum solutions for your health science course?
Get a free demo

RELATED STORIES