Who Decides and On What

A flag becomes an outcome by passing through hands. Which hands, in what order, with what authority, is the design question institutions skip. The responsibility model can be compared with the project example, where access to records and authority to interpret them should remain separate and explicit.

The chain

The system flags. It concludes nothing.

A reviewer assesses the recording and may close, mark inconclusive, or escalate.

An academic or integrity officer decides whether there is a case to answer and puts it to the student.

A panel or officer reaches a finding, on the whole evidence including the student's account.

And an appeal body reviews the process and the conclusion.

Five stages. Most institutions have three and improvise the rest.

What each may conclude

The reviewer may not find misconduct. They may say what the recording shows.

The case officer may not find misconduct either. They decide whether there is enough to ask.

Only the panel finds, and only with the student's explanation in front of it.

Writing this down prevents the commonest failure, which is a reviewer's note reading as a finding by the time it reaches the student.

Separation

The person who flags, the person who reviews, the person who brings the case and the person who decides should not all be the same.

In small institutions complete separation is not possible; separating the reviewer from the decider is the minimum and is achievable everywhere.

Document which separations you have and which you cannot manage, because an appeal will ask.

What the standard of proof is

State it. Most academic processes use balance of probabilities; some use a higher bar for serious findings.

And state what proctoring evidence can contribute to it: a flag is not evidence, a recording is, and an unexplained recording of a second person supplying answers is strong.

An institution that has not written this down will apply it inconsistently, which is visible across cases and is the basis of successful appeals.

The student's account

Must be taken before any finding, not after.

In writing or in person, with the recording available to them.

And genuinely weighed: a process where the account is collected and the outcome was already drafted is a process in name.

Reviewers and panels should be able to point to cases where the account changed the outcome. If there are none, that is the finding.

Records of the decision

Who decided, on what evidence, with what reasoning, when.

Kept for the appeal period and long enough for any external review.

And written in terms that could be shown to the student, because frequently they will be.

The small-institution problem

One person doing everything is common in small providers and is the highest-risk arrangement.

Mitigations: external review of any upheld case, written criteria applied strictly, and a second opinion from another institution where reciprocal arrangements exist.

Not ideal, and better than a single unchecked judgement.

What to check

Can you name the person at each of the five stages?

Is the reviewer separate from the decider?

Is your standard of proof written down?

And can anybody point to a case where the student's account changed the outcome?

The point

The reviewer may not find misconduct; they may say what the recording shows.

Only the panel finds, and only with the student's account in front of it.

Worth stating

In small institutions complete separation of roles is not possible; separating the reviewer from the decider is the minimum and is achievable everywhere.

Document which separations you have and which you cannot manage, because an appeal will ask.

Also worth knowing

State the standard of proof and what proctoring evidence can contribute to it.

An institution that has not written this down will apply it inconsistently across cases, which is visible and is the basis of successful appeals.

And finally

Records of the decision — who decided, on what evidence, with what reasoning, when — should be written in terms that could be shown to the student, because frequently they will be..

Summary

Separate the reviewer from the decider at minimum, document which separations you cannot manage, and be ready to explain both.

In summary

Five stages, five different authorities, and only the last may make a finding.

Most institutions operate three and improvise the rest, which is visible in how a flag is worded by the time it reaches a student. For wider institutional context, consult the Association for Computing Machinery.