Skip to content
Flag Is Not Finding

A flag is a moment for review, not a finding of misconduct

Fifty notes for institutions deciding whether to deploy remote proctoring, how to run it fairly if they do, and what to use instead where it does not fit. The same questions about purpose, proportionality and human review arise when organisations evaluate employee monitoring software, but assessment stakes make those boundaries more acute.

All notesStart here

All 50 notes

Grouped by subject. If you are deciding whether to deploy at all, start with the foundations; if you already have a programme running, start with the review step.

Foundations

The systems report deviation from an expected pattern. That is narrower than cheating, and the gap is where every dispute lives.

5 notes

Types of proctoring

Five approaches with different costs and different failure modes. The stakes should choose, not the sales process.

6 notes

What is detected

Each signal has ordinary explanations, and the weakest ones generate the most flags.

6 notes

Who gets flagged

Flag rates differ by group for reasons unrelated to conduct, and the pattern tracks disadvantage that already exists.

7 notes

The process

This is what an institution is actually buying, and it is the part most often designed last.

7 notes

Legal obligations

Systematic monitoring of people in their homes sits toward the higher end of regulatory attention in most regimes.

5 notes

Alternatives

If having the internet open does not help, there is nothing to detect. That closes the channel rather than watching it.

6 notes

In practice

Procurement, deployment and the figures that show whether the programme is functioning.

5 notes

Reference

The end state as a checklist, the limits stated plainly, and where to start reading.

3 notes

What this is

Fifty notes on remote proctoring, written for institutions deciding whether to use it and how to run it fairly if they do.

No product names, no supplier material, no detection accuracy figures.

Four things that hold

A flag is not a finding. The systems report deviation from an expected pattern, which is narrower than cheating, and the gap is filled by a human process.

Student consent is not freely given. Refusing costs the qualification, which makes proportionality the test rather than agreement.

The flag burden falls unevenly, in a pattern that tracks disadvantage and is measurable in any institution willing to look.

What is bought is the review step, the appeal and the alternative — not the detector, which is the cheap part and determines nothing.

Where to start

Deciding whether to deploy: the question to ask first, and what proctoring actually detects.

Already running a programme: the review step, and measuring whether it works.

Writing policy: anomaly is not cheating, evidence standards, and the appeal.

Facing a challenge: consent, room scans, disability and accessibility obligations.

Looking for something else: assessment design and oral examination.

What this is not

Not legal advice. Data protection, biometrics, accessibility and automated decision rules differ substantially by jurisdiction.

It does not explain how to circumvent proctoring, and it does not recommend suppliers.

Independent software comparisons

Long-form shortlists for assessment, academic integrity, learning delivery and education operations.

For broader policy and education context, see UNESCO's official site.