In our modern digital era, the online examination has undoubtedly become an integral component of exams. The student assessment is submitted remotely without any physical interaction. This method is not just more affordable, but also less time-consuming. However, student authentication and compliance have their own set of challenges. Ensuring online exam integrity has become a necessity.
One of the safest methods to ensure that the student is not cheating during an exam is typing biometrics or biometric authentication. Generally, a biometric system is designed to solve a matching problem through the live measurements of human body features which are more appropriate for online learner verification.
There are seven levels of security when it comes to proctoring.
While the first two levels are way too insecure, and the last two levels are not used anymore due to the COVID-19 lockdown, we are stuck at levels three through five. ObservED is an affordable tool that offers examiners the security benefits of Level 3 and Level 4 proctoring.
How does Typing Biometrics Operate?
Using Typing Biometrics, or Keystrokes Dynamics, a typing pattern profile of the user is created, which is later used to authenticate that user, by comparing with great accuracy, of over 90%, the two typing patterns.
In compiling the typing pattern, all sorts of telemetry and information are gathered, from the time-space between two keypresses to the timing of each individual word the user has typed. If the similarity is high, the verification is successful, otherwise, it will be considered unsuccessful.
Why Typing Biometrics?
A growing number of academies and learning institutions across the world are beginning to leverage the power of typing biometrics. For instance, in the US, the Department of Motor Vehicles (DMV) was one of the first official institutions to allow typing biometrics as a mandatory technique for digital identity validation.
They explicitly state that their validation methods come together with a proctored exam. Typing biometrics comes as a necessary component to prevent cheating and is a major pillar in providing the means for a cost-efficient remote proctoring solution.
Now, with the COVID-19 restrictions, the need for students to take classes online, as well as for proctoring tools is greater than ever. Especially with over 70% of students admitting to cheating during exams at one point in their lives.
Typing biometrics comes as a necessary component of a remote proctoring solution in order to automate the examination process and overcome all the shortcomings of physical proctoring, which include human error, lack of scalability, inefficiency, and high wait time, and bandwidth.
Typing biometrics and automated proctoring come with the following benefits:
- A scalable and robust technique that allows universities and learning centers to ensure the fairness of exams to a large number of students, without having to hire an additional workforce
- It’s accessible to all students
- It’s cost-effective, as it doesn’t involve any costs associated with physical proctoring.
A frictionless authentication by the way you type – Meet TypingDNA
One of the promoters of typing biometrics in virtual proctoring is TypingDNA. This biometric authentication tool released by TypingDNA is a highly-accurate, complex product that minimizes the risk of online fraud and promotes fairness across students. Being powered by AI-driven behavior, this technology can be applied in numerous areas, from insurance and financial services to online learning.
TypingDNA is a primary component of the ecosystem that makes ObservED a complete proctoring tool. The other two major components of the ObservED platform are advanced analytics powered by machine learning and a face recognition system.
How does TypingDNA Works?
Basically, this AI-driven software can recognize people based on how they type. During the course, a constant pop-up shows up to determine if the student has not switched places with someone else. Each iteration is recorded, coming with its own matching score based on the way the user types.
When the matching score released is higher than the risk threshold, the system instant reports any anomalies. This system integrates with multiple LMSs and works flawlessly inside Tremend’s ObservED solution.
In the next article of this series, we’ll explore the other two components of ObservED, namely the advanced analytics powered by machine learning and the face recognition system.
Until the next time, you can watch a recording of a highly successful eProctoring webinar conducted by Tremend in collaboration with TypingDNA. Get free access to it here.