What Proctoring Cannot Fix

Proctoring observes conduct during an exam. Most integrity problems are not located there. The limits are easier to state beside the workflow reference, because no operational platform can repair weak assessment design, unclear policy or absent human judgement.

Contract cheating

Work produced by somebody else and submitted as the student's own.

The dominant integrity problem for coursework in most institutions, and proctoring does not touch it.

What works: staged assessment, personalised inputs, oral defence on a sample.

Paper security

A leaked question bank is undetectable in session monitoring, because nothing anomalous happens.

The students who have seen the paper behave entirely normally.

What works: item rotation, staged release, monitoring for anomalous cohort performance rather than anomalous individual behaviour.

An assessment that rewards recall

If the exam tests what can be looked up, the incentive to look it up is structural.

Surveillance manages the symptom at considerable cost.

What works: changing what is assessed, which the alternatives section covers and which is also better education.

A cohort that does not trust the institution

Proctoring imposed without explanation, on students who believe it is inaccurate and unfair, produces resistance and complaints independent of any technical question.

No configuration fixes a trust problem.

What works: explanation, published figures, a real appeal, and a genuine alternative.

Determined, organised cheating

Substitutes with genuine documents, prepared rooms, material outside the frame.

The measure is best against the careless.

What works: identity verification proportionate to the risk, and assessment design that makes the effort not worth it.

Institutional anxiety

Frequently the real driver: a fear of a story about cheating.

A proctoring contract answers that anxiety without changing much, which is why it is bought.

What works: deciding what the institution actually stands behind, and being able to describe its integrity approach as a whole rather than pointing at a product.

Using this before procurement

For each thing you hope proctoring will address, ask what would change if it worked perfectly.

Where the answer requires redesigning the assessment, securing the papers, or changing the relationship with students, the software is not the constraint.

Removing those leaves what it actually delivers.

What it does deliver

Deterrence, announced.

Triage, reducing what a human must watch to a reviewable set.

A record that supports a fair process where one is properly designed.

And, in genuinely high-stakes remote assessment, the ability to run it at all.

Four things. None of them is academic integrity, and all of them are real.

The point

For each thing you hope proctoring will address, ask what would change if it worked perfectly.

Where the answer requires redesigning the assessment or securing the papers, the software was never the constraint.

Worth stating

No configuration fixes a trust problem.

Proctoring imposed without explanation, on students who believe it inaccurate and unfair, produces resistance independent of any technical question.

Also worth knowing

What it does deliver is deterrence when announced, triage that makes review possible, a record supporting a fair process, and the ability to run high-stakes remote assessment at all.

Four things, none of them academic integrity.

And finally

Determined organised cheating — substitutes with genuine documents, prepared rooms, material outside the frame — is what the measure is worst against.

Identity verification proportionate to the risk and assessment design do more.

Summary

Ask what would change if it worked perfectly. Where the answer requires redesigning the assessment or securing the papers, the software was never the constraint.

In summary

Contract cheating, paper security, recall-based assessment, institutional anxiety and a cohort that does not trust you.

None of these is fixed by any configuration, and each has a different remedy. For wider institutional context, consult the European Commission.