Identity Verification in Online Exams: Methods and Questions

Identity verification is the first and most important process in an online exam to check whether the registered candidate is taking the exam and not another person, to maintain exam integrity. It typically involves comparing the registered candidate's information, such as an image and other details, with the candidate present before or during the assessment.

Identity verification is different from login authentication:

• Authentication: Based on the system configuration, the candidate may receive an OTP to log in to their account.

• Identity Verification: Webcam, AI, or a human can check whether the person using the account is linked to the claimed real-world identity.

If the identity verification fails or is interrupted, the examination provider may take necessary precautions and alternative options to make the process smooth.

The ID verification methods include several steps based on the online proctoring software and the conducting body's requirements.

A review of online assessment systems published in Smart Learning Environments reported that, among the 22 systems examined, 19 (86.4%) used government-issued ID checks and 16 (72.7%) used face recognition before the exam.

Here are some of the most common methods used during and before  online exam monitoring to ensure fairness and security.

1. Government-Issued ID Verification

The candidate must provide a government-issued ID card, such as a Voter ID, passport, or driving license, to compare it with the registration details. Based on the process, the document would be captured and scanned by an automated proctoring system or a human proctor.

2. Facial Recognition

Facial recognition that includes eyes and other features to ensure the identity of the candidate. This method can reduce reliance on manual visual comparison, but a match should be interpreted according to the system's validated performance and the exam's procedures.

3. One-Time-Password (OTP)

To ensure login details and candidate authentication, the system may send an OTP through the registered email ID and contact number. However, an OTP verifies control of a channel or account; it does not independently prove the user's real-world identity. A person may share account access or receive a code on behalf of someone else.

4. Live Human Identity Verification

As per the exam requirement and system configuration, a human proctor or authorized reviewer checks the candidate’s ID and compares it with the person appearing on the camera.

5. Liveness and Anti-Spoofing Checks

Liveness checks attempt to determine whether the biometric input comes from a live person rather than a presentation attack, such as a photograph or replayed video. Implementations vary and may use prompts, motion cues, or automated analysis.

6. Continuous Identity Checks During the Exam

Not only at the beginning, some of the platforms continuously monitor candidate identity using AI or an automated system. This can help identify a possible change in the person taking the assessment.

Before selecting online exam software, universities, certification bodies, and recruitment teams should ask vendors the following questions related to identity verification.

Top 8 Questions to Be Asked:

• How is identity verification performance measured and tested?
• What happens when the system cannot match a face or read an ID?
• Does the provider test performance across different lighting conditions, devices, and candidate groups?
• Does the system check identity only at login, or can it perform additional checks during the exam?
• How are identity records and biometric information protected from unauthorized access?
• Are facial images stored, and if so, for how long?
• Who can access identity records, and when are they deleted?
• Can candidates receive help when verification fails?

A failed ID check does not mean that the candidate is attempting impersonation and prevents them from taking the exam. Sometimes system limitations, poor camera angle, network issues, and dim lighting may generate false alerts that require immediate assistance.

Well, in that case, check why the system failed and what the reason is. Allow and assist the candidate to solve the issues, fulfill the requirements, and try again. Use other verification methods or escalate issues to the right person.

Final Thoughts

Identity verification is an important part of online exam security, but no single method is suitable for every assessment. ID checks, facial comparison, OTPs, liveness detection, and human verification each serve different purposes and have different limitations. For institutions choosing online exam software, the priority should be to select a proportionate combination of methods, test its reliability, protect candidates' personal information, and provide a fair way to resolve verification failures.