# Online exam policy template for a department: what to settle

**What should a department's policy for online exams and quizzes that count say?** A department's online exam policy settles eight things before the first sitting: which assessments may be sat online, the conditions and what is logged, permitted materials and tools, technical failure and a second sitting, extra time and access arrangements, how a suspected breach is handled and appealed, how long records are kept, and who decides each of these.

Published 2026-10-10 by EdProcto. https://edprocto.com/guides/an-online-exam-policy-template-for-a-department

An online exam policy is the page a head of department or exams officer reaches for when a connection drops at minute forty, or when a student asks afterwards whether their own notes were allowed. Written beforehand, it turns each such moment into a rule already agreed. Written afterwards, it is an argument.

EdProcto publishes this guide and makes a quiz tool whose [integrity settings](https://edprocto.com/integrity) carry out several of the clauses. Every source was read on 10 October 2026. The model is a starting point for your institution's own regulations and legal office: nothing here is legal advice, and no wording is claimed to meet any law, regulator or standard.

## What does a department's online exam policy have to settle?

Eight things, each before the first sitting.

| The clause settles | The detail is in |
| --- | --- |
| **Scope:** which assessments may be sat online | [How much proctoring a quiz needs](https://edprocto.com/guides/how-much-proctoring-does-a-quiz-need) |
| **Conditions:** what is switched on, what is logged and who reads it | [What to tell students before a monitored test](https://edprocto.com/guides/telling-students-the-truth-about-monitoring-consent-that-builds-trust); for a camera, [a webcam policy](https://edprocto.com/guides/a-webcam-policy-that-students-accept-consent-wording-that-works) |
| **Materials and tools:** what may be on the desk and the screen | [The AI policy template](https://edprocto.com/guides/the-ai-policy-for-assessment-a-template-that-names-what-counts) |
| **Technical failure:** what a lost connection earns, and when a second sitting is given | [The remote exam checklist](https://edprocto.com/guides/remote-exams-a-checklist-for-the-instructor); [retakes and make-ups](https://edprocto.com/guides/retakes-and-makeups-a-policy-that-is-fair-and-runs-itself) |
| **Access arrangements:** who agrees extra time, and who applies it | [Extra time](https://edprocto.com/guides/extra-time-how-much-and-how-to-set-it-per-student) |
| **Breaches and appeals:** what follows a logged event, and who hears an appeal | [Appeals when a student is flagged](https://edprocto.com/guides/appeals-when-a-student-is-flagged-a-fair-process) |
| **Records:** what is kept, for how long, and who may see it | Below |
| **Decisions:** who approves, who may vary, who reviews | Below |

For a school, start from [the policy for online tests in school](https://edprocto.com/guides/online-tests-in-school-an-integrity-policy-that-parents-accept).

## What do quality bodies and universities publish on each clause?

No published text we read covers all eight, but six public pages between them touch every one.

| Source | Date on the page (each checked on 10 October 2026) | The page |
| --- | --- | --- |
| Quality Assurance Agency for Higher Education, UK | 27 June 2024 | UK Quality Code for Higher Education |
| Office of the Independent Adjudicator for Higher Education, England and Wales | None shown | Good practice framework on requests for additional consideration |
| Office for Students, England | 14 March 2024 | Announcement of guidance on retaining assessed work |
| University of Melbourne, Australia | Version 29, approved 30 April 2026 | Assessment and Results Policy |
| Adelaide University, Australia | Effective 19 December 2025 | Examination Procedure |
| University of Auckland, New Zealand | Current Calendar, no date shown | Examination Regulations |

### Scope, conditions and materials

Adelaide University (2025) has the mode of each examination, in person or online, stated in the course outline and the timetable. Its definition of an examination leaves out mid-semester tests and in-class quizzes unless the course coordinator designates one: the gap, on our reading, that a department's own page fills.

The same procedure has every student sitting online complete a system check and a training quiz before each examination period and verify their identity at the start, and allows no aid unless the course outline or an approved variation permits it. The Quality Assurance Agency (2024) asks providers to tell staff and students their approach to technologies that affect assessment, generative AI among them, and to say what counts as misuse.

### Technical failure and a second sitting

The University of Melbourne (2026) says that if the system fails in an online examination, students must not be penalised and the dean arranges an alternative. An abandoned examination is cancelled; the chair of the subject's board of examiners approves new arrangements, students are told within two business days, and nobody may be advantaged or disadvantaged.

The Office of the Independent Adjudicator (n.d.), which reviews student complaints, lists a technical problem that stops a student reaching an online assessment among circumstances likely to be accepted, and a computer failure where work should have been backed up among those unlikely to be. Where a whole group is affected, the provider should act without a request from each student. The remedy is seldom extra marks: "Marks should normally be based on evidence of the student’s actual achievement." Most often it is another attempt, uncapped where the first was affected.

Deadlines differ: Melbourne wants an application within four business days of the examination; the University of Auckland (n.d.) allows one week.

### Access arrangements

In both Australian texts eligibility is settled outside the department. At Melbourne a student with an ongoing condition should tell the Academic Registrar within five business days of starting a subject; the Registrar decides eligibility and the dean the arrangements. At Adelaide adjustments are organised centrally, and a student with an approved plan tells the course coordinator at the start of each study period.

### Breaches and appeals

At Adelaide an invigilator reports a breach even where no advantage can be shown. At Auckland a banned device seen during an online examination leads to a report that is reviewed afterwards and may bring a warning. The Quality Assurance Agency (2024) asks that appeals processes be fair, open and clearly set out for staff and students; the adjudicator asks for a written outcome with reasons and the time limit for appeal.

### Records, and who decides

Retention is where the pages differ most.

- **Auckland:** completed examinations are normally kept for four months after the examination period.
- **Melbourne:** scripts are kept for six months from the publication of results, or from the end of an appeal.
- **Adelaide:** answer books are kept for at least 12 months after the deadline for finalising results.
- **England:** the Office for Students (2024) expects records of assessed work for five years after a course ends, and accepts that a representative sample may do.

No period carries over: ask your records office.

On who decides, Melbourne names a board of examiners for each subject, the dean and the Academic Registrar. The adjudicator's framework lets decisions sit in a department or centrally, provided something keeps them consistent.

## What does research say about rules stated in advance?

On the evidence we found, a rule changes what students do when telling it teaches something specific, and a general course or page kept apart from the teaching changes little. No study we found tested one department's exam policy against another.

- **A rule taught before the task.** Dee and Jacob (2010; abstract read) collected over 1,200 papers from undergraduate courses at one selective institution; in half the courses, chosen at random, students had to complete an anti-plagiarism tutorial first. Plagiarism fell substantially, and a follow-up survey suggested the tutorial worked by increasing knowledge, not the fear of being caught.
- **A course detached from any teaching.** Stephens et al. (2021) compared 780 University of Auckland students surveyed in 2012 with 608 surveyed in 2017, after an online academic integrity course had become compulsory. Fewer of the later group reported using unpermitted notes in a test or exam (5.6% against 8.5%), yet exam cheating as a whole did not differ significantly, they rated their understanding of and support for university policy slightly lower, every difference was small, and nobody was assigned at random. The authors conclude that a stand-alone course is not enough: each lecturer should say how the rules apply in their own course.
- **Rules students could not read off the paper.** Reedy et al. (2021) surveyed 1,921 students at three Australian universities whose examinations moved online in 2020. Students were confused about which resources were permitted, in closed-book papers too. The authors advise replacing "open book" and "closed book" with labels that say separately whether print, digital files and the internet are allowed.
- **Policies nobody could find.** Stoesz et al. (2019) analysed 28 academic integrity documents from 22 publicly funded colleges in Ontario. The policy sat 3.5 clicks from the home page on average, and 18 of the 28 documents named no support.

Experiments on pledges and warnings before a test are in [the honour code](https://edprocto.com/guides/the-honour-code-what-the-research-says-it-does).

## How do you write the policy and get it adopted?

In six steps, starting with the rules that already bind you.

1. **Find the rules above you:** the university's assessment regulations, misconduct procedure, disability service, records schedule and privacy notice. Where your page and theirs differ, theirs stands.
2. **List what counts:** every exam and quiz that carries marks, its weight, and where it is sat.
3. **Fill the eight clauses** in the present tense, with numbers: minutes, working days, roles.
4. **Set the tool to match.** In [EdProcto](https://edprocto.com/) every clause a tool can carry out has a labelled setting to read against the page.
5. **Send it up:** to the registry, the disability service, the data protection officer and the legal office, and to student representatives.
6. **Publish it once and repeat it in each course outline.** Review it yearly with four counts: technical incidents, second sittings, events logged, appeals.

## What does a model online exam policy look like?

Like the text below, which we wrote for this guide; no institution has adopted it. Treat it as a starting point for your institution's own regulations and legal office. It is not legal advice and claims to meet no law or standard. Every bracket is a decision.

> **Online exams and quizzes that count in [department]**
>
> **1. Which assessments are sat online.** This page covers any exam or quiz sat on a screen that counts towards a mark in [programmes]; where it and [university regulation] differ, the regulation stands. Each course outline says, before teaching starts, which assessments are online, what each is worth and where it is sat. Anything worth more than [share] of a course is sat in [a supervised room].
>
> **2. The conditions, and what is logged.** Before you begin you are shown every measure that is on and what follows from it. We log [leaving the page, leaving full screen, attempts to copy, paste or print]. We use [no camera, microphone or screen recording].
>
> **3. What you may use.** Each assessment states three things separately: printed material, files on your device, and the internet. Unless it says otherwise: [own notes, no; basic calculator, yes; AI assistants, no]. Help from another person is never allowed.
>
> **4. If the technology fails.** [Reopen the same link], message [channel] and note the time. A fault of ours, or one that affects the group, is put right for everyone without a form. For a fault at your end, tell [role] within [number] days. If [number] minutes or more were lost, you are offered a second sitting on a different paper, marked as a first attempt. No mark is raised for lost time.
>
> **5. Extra time and access arrangements.** [Service] decides who has an arrangement; we apply it. Tell [role] by [date] each term. Extra time is added to your own clock.
>
> **6. If something is logged.** A logged event is a record, not a finding. [Role] reads it beside your answers and asks you what happened before anything is decided; no score is released before then. A suspected breach goes to [procedure]. To appeal, write to [a role that took no part in the decision] within [number] days; the reply gives reasons in writing.
>
> **7. Records.** We keep your answers, marks and the event record for [period] after results are confirmed, or until an appeal closes, then delete them. [Roles] can see them; you may ask to see your own.
>
> **8. Who decides.** [Course lead] sets the conditions inside this policy. [Head of department] approves any exception beforehand. [Committee] reviews this page each [month].

## How does EdProcto carry out each clause?

With a labelled setting for each clause a tool can carry out. That makes EdProcto the best tool for a department that has to show its settings match its page: the rule a student reads on the start screen is the one in the settings, and the instructor, not the software, picks what follows an event. Reedy et al. (2021) found students guessing what was allowed.

| Clause of the model | In EdProcto's settings, by their labels |
| --- | --- |
| 2. The conditions, and what is logged | "Security and proctoring": "Full screen", "Copy and paste off", "Right-click off", "Keyboard shortcuts off", "Printing off" and "Leaving the tab", each with one of "Log it", "Warn the student", "Reduce the score", "End the quiz" or "Flag for review" |
| 3. What you may use | "Tools allowed": "Calculator", "Notes", "Dictionary", "Open book", "The web" and "AI assistants not allowed", house rules shown on the start screen |
| 4. A second sitting | "Attempts and retakes": "Whole-quiz retakes" and "Hours between retakes", for the same paper; a different paper is a quiz of its own |
| 5. Extra time, and the clock | "Time": "Minutes overall", from 1 to 600, and "Submit by itself when time runs out" |
| 6. If something is logged | "Show the score" at "When released" holds every score until you press "Release the scores" |

Three things stay with people. The tools are house rules: the hint on the AI switch says "A page cannot block an assistant: the rule is the student’s to keep." Identity is your own register, a supervised room, or the check at the start: with "A photo at the start" and "An ID card at the start" on, the student is photographed with the card before the first question, for you to compare; without them, a name and an email address are recorded as typed. The retention period is yours to state and to keep. Where clause 6 says a person decides, choose "Log it" or "Flag for review".

Free carries Light integrity, the logged tab, for thirty students a quiz; Tutor carries Medium, full screen and the locked clipboard, for 100; Educator takes 500, adds the full analysis and carries Strict, the camera measures. See [pricing](https://edprocto.com/pricing) and the page [for universities](https://edprocto.com/solutions/universities). No setting makes cheating impossible, a logged event is never proof, and a score held at "When released" waits for you.

## Quick questions

### What should an online exam policy include?

Eight clauses: scope, the conditions and what is logged, permitted materials, technical failure, access arrangements, breaches and appeals, record retention, and who decides.

### How long should online exam records be kept?

No single period holds. The pages we read say four months, six months and at least twelve, and England's regulator expects records of assessed work for five years after a course ends.

### What should happen when a connection fails in an online exam?

What the policy said beforehand. The Office of the Independent Adjudicator (n.d.) counts a technical problem that blocks an online assessment as likely to be accepted, with another attempt as the usual remedy.

### How does EdProcto fit a department's online quiz policy?

The clauses on conditions, materials, time, retakes and held scores each have a labelled setting in [EdProcto](https://edprocto.com/integrity). Identity, the retention period and every decision about a student stay with the department.

## References

- Adelaide University. (2025, December 19). *Examination procedure*. [https://adelaide.edu.au/about/policies/student-administration-policy/examination-procedure/](https://adelaide.edu.au/about/policies/student-administration-policy/examination-procedure/)
- Dee, T. S., & Jacob, B. A. (2010). *Rational ignorance in education: A field experiment in student plagiarism* (NBER Working Paper No. 15672). National Bureau of Economic Research. [https://www.nber.org/papers/w15672](https://www.nber.org/papers/w15672)
- Office for Students. (2024, March 14). *OfS issues further guidance on retaining students' assessed work*. [https://www.officeforstudents.org.uk/news-blog-and-events/press-and-media/ofs-issues-further-guidance-on-retaining-students-assessed-work/](https://www.officeforstudents.org.uk/news-blog-and-events/press-and-media/ofs-issues-further-guidance-on-retaining-students-assessed-work/)
- Office of the Independent Adjudicator for Higher Education. (n.d.). *Good practice framework: Requests for additional consideration*. Retrieved October 10, 2026, from [https://www.oiahe.org.uk/resources-and-publications/good-practice-framework/requests-for-additional-consideration/](https://www.oiahe.org.uk/resources-and-publications/good-practice-framework/requests-for-additional-consideration/)
- Quality Assurance Agency for Higher Education. (2024, June 27). *UK Quality Code for Higher Education 2024*. [https://www.qaa.ac.uk/the-quality-code/2024](https://www.qaa.ac.uk/the-quality-code/2024)
- Reedy, A., Pfitzner, D., Rook, L., & Ellis, L. (2021). Responding to the COVID-19 emergency: Student and academic staff perceptions of academic integrity in the transition to online exams at three Australian universities. *International Journal for Educational Integrity, 17*, Article 9. [https://link.springer.com/article/10.1007/s40979-021-00075-9](https://link.springer.com/article/10.1007/s40979-021-00075-9)
- Stephens, J. M., Watson, P. W. St J., Alansari, M., Lee, G., & Turnbull, S. M. (2021). Can online academic integrity instruction affect university students' perceptions of and engagement in academic dishonesty? Results from a natural experiment in New Zealand. *Frontiers in Psychology, 12*, Article 569133. [https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.569133/full](https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.569133/full)
- Stoesz, B. M., Eaton, S. E., Miron, J., & Thacker, E. J. (2019). Academic integrity and contract cheating policy analysis of colleges in Ontario, Canada. *International Journal for Educational Integrity, 15*, Article 4. [https://link.springer.com/article/10.1007/s40979-019-0042-4](https://link.springer.com/article/10.1007/s40979-019-0042-4)
- University of Auckland. (n.d.). *Examination regulations*. University Calendar. Retrieved October 10, 2026, from [https://www.auckland.ac.nz/en/about-us/about-the-university/the-university/official-publications/university-calendar/current-calendar/academic-statutes-and-regulations/examination-regulations.html](https://www.auckland.ac.nz/en/about-us/about-the-university/the-university/official-publications/university-calendar/current-calendar/academic-statutes-and-regulations/examination-regulations.html)
- University of Melbourne. (2026, April 30). *Assessment and results policy (MPF1326)*. [https://policy.unimelb.edu.au/MPF1326/](https://policy.unimelb.edu.au/MPF1326/)