Most departments already own a quiz library. It is spread over six teachers' accounts, two memory sticks and a folder called "Tests NEW (2)". A department quiz library for a school puts those quizzes in one named collection, each with an owner, a version and a date.
EdProcto, which publishes this guide, is where such quizzes are written, passed on and run: see its page for schools. The steps for handing one quiz to a colleague are in how to share a quiz as a kit; this guide is about the collection.
The research below was read on 10 October 2026, as abstracts where we say so. None of it tested a quiz library, so the rules here are our working rules.
Is a shared quiz library worth the trouble?
It is, where shared time goes on the quizzes themselves: the evidence favours teachers working together and warns that meetings alone add work.
What studies of teacher collaboration found
- Schools. Goddard et al. (2007) surveyed 452 teachers in 47 elementary schools of one large urban district in the American Midwest, with 2,536 fourth-grade pupils. Pupils scored higher in mathematics and reading where teachers reported more collaboration for school improvement. Nothing was assigned at random, and the authors call the support preliminary (abstract read).
- Departments. Lomos et al. (2011) used TIMSS 2003 data on about 3,000 students and 130 schools in the Netherlands. Achievement was higher where the mathematics department was seen as a professional community, one that focuses on reflective dialogue, collaborative activity, a shared vision and achievement. The effect size was above 0.20 (abstract read).
- Assessments written together. Frey and Fisher (2009) describe one American elementary school whose teachers set up common formative assessments, and report higher achievement and teachers better at designing assessments (abstract read).
What teachers and schools say about workload
Hunter and Parkinson (2023) report two surveys by their own institute of more than 7,000 Australian principals and teachers. In 2021, 49% of teachers called collaborative planning meetings a barrier to having enough preparation time. In 2022, nine in ten wanted curriculum materials to share with colleagues, yet 55% said such meetings usually or always drifted to other matters, and three quarters named a lack of effective leadership as a barrier to shared materials.
The Department for Education (n.d.), in its workload reduction toolkit for England, publishes one secondary school's account: King Charles I School keeps departmental resources centrally so that nobody plans from scratch, and accepts "less autonomy in the classroom" for less work at home. Nothing was measured.
Our reading: the gain comes from material made once, checked by a second person and then used. No study we found measured a quiz library itself.
What goes in the library, and what stays personal?
Put in the library what more than one teacher will set, and keep personal what was written for one class or names a pupil.
| In the library | Stays with its teacher |
|---|---|
| End-of-unit tests, and any quiz every class sits | A quiz written to reteach one class's gap |
| Practice quizzes tied to a week of the scheme of work | Drafts and experiments |
| Baseline quizzes for the year group | Class results, names and notes on pupils |
| Retired versions, marked as retired | Anything built on material the department may not pass on |
The last cell covers a publisher's test bank or an exam board's paper: the sharing guide above lists the checks. The library fixes the starting point; each teacher still adapts their own copy.
How do you name and label a quiz so a colleague finds it?
Put the year group, unit, week, purpose, version and date in the title, always in that order, and list every quiz in one index.
A naming convention for the title
Y8 U1 W2 Cells: practice, v1.1, 16 Sep 2026
- Year, unit and week first, so a unit's quizzes sit together.
- Topic, in two or three words.
- Purpose: baseline, practice, test or retest, which says whether answers may be shown.
- Version and its date: the first number changes with a rewrite, the second with a correction.
EdProcto's dashboard lists a teacher's quizzes under "Your quizzes", with a box that reads "Search by name or template". The search matches the run of characters typed, so a fixed order pays: Y8 U1 brings up the unit and Y8 U1 W2 the week. A copy keeps the title it was taken under, which says which version it is.
The index and the labels inside
The index is one sheet on the drive the department already uses, a row a quiz: title, owner, shared address, status (in review, live or retired) and last change.
Inside each quiz, give every question a topic in the scheme of work's words: the full analysis, on the Educator plan and above, reports by topic.
Who may change a shared quiz, and how does a correction reach everyone?
One named owner for each unit changes the library's version; everyone else changes their own copy and reports faults to the owner. It is version control for quizzes, done by hand.
One owner a unit
The owner holds the master of each quiz in the unit, in their own EdProcto account, and alone edits it. A master is never sat: for their own class the owner presses "Duplicate this quiz". An owner is the smallest form of the leadership the teachers in Hunter and Parkinson (2023) missed.
A short review before a quiz joins
Before a quiz becomes v1.0, a second teacher takes a copy and sits it cold on EdProcto's study screen, which checks each answer as it is given. The reviewer checks every key, every wording, and whether each class will have been taught what a question assumes. Studies of question review are in our guides to common assessments and question banks.
A log of changes
Every change to a master gets one line: date, quiz, versions, what changed. A copy is its holder's own quiz and stays as it was taken, which gives the working rule:
- The owner corrects the master in EdProcto and raises the version in its title. The address stays the same and serves the corrected quiz from then on.
- The owner writes the log line and tells the department.
- A teacher whose class has yet to sit the quiz takes a new copy.
- A teacher whose class has sat it corrects their own copy and presses "Re-mark this quiz" on its results page, which marks the handed-in attempts again from the keys as they stand.
How do you retire a quiz, and why keep old versions?
Retire a quiz by taking it out of circulation without deleting it, because every result already held was earned on one particular version.
A score of 14 out of 20 means something only beside the twenty questions that produced it: a parent's query or a re-mark needs the paper as it was sat.
- To retire a master, the owner presses "Shared as a kit" again, and its address then answers "Nothing is shared here". The owner adds "retired" to the title and marks the index. Copies colleagues hold stay with them, results included.
- For next year's edition, the owner presses "Duplicate this quiz" at the foot of the builder. The duplicate is a draft with the same questions, keys and settings, a new code and no submissions, its title ending "(copy)". Renamed v2.0 and reviewed, it is the new master. The same button gives a teacher a copy for a second class.
- Sat copies are closed and kept. "Delete this quiz" removes a quiz together with its submissions, and its confirmation ends "This cannot be undone." Delete one only when the school's period for keeping results has run out.
How do you keep questions secure when classes and later years reuse a quiz?
Sort the library by stakes: a practice quiz is reused openly, and a test that counts gets a second version, a short window and an address that is open only while colleagues copy.
What studies of reused questions found
- Wood (2009) studied an examination that repeated 36 multiple-choice questions at two sittings. For the 130 candidates who sat both, ability estimates rose on the reused questions and equally on the rest, so reuse did not appear to help them (abstract read).
- Wagner-Menghin et al. (2013) had 671 students sit a new examination in four shifts, on forms mixing published, unused and reused questions. The reused ones did not become significantly easier, and the authors conclude that reusing a low ratio of randomly chosen questions need not spoil an examination (abstract read).
In both, only part of a paper returned. A cohort study in the question banks guide points the other way, and no school study we found covers a whole quiz reused yearly.
Working rules for a department
| Where | Rule | In EdProcto |
|---|---|---|
| A practice quiz | Reuse it freely: showing the answers is its purpose | A copy arrives as a study quiz |
| A test that counts | Versions A and B, written to one blueprint and alternated by year; every class inside a few days | Each teacher sets "Show the score" to "When released" until the last class has sat |
| A master's address | It shows the questions, without their answers, to anyone who has it, and a copy carries the key: post it only where staff read | Press "Shared as a kit" again once colleagues have their copies |
| In the room | The supervision the stakes need | During a sitting EdProcto keeps the answer key on its server and applies the measures the teacher chose. No setting makes copying impossible |
What should a department agree about students' results?
Agree where results sit, what is shared, how long they are kept and what they are never used for. This is practice, not legal advice: your school's data protection lead has the last word.
- Where they sit. Each class's results are in the account of the teacher whose copy the class sat. For one results page across a year group, one teacher runs the quiz for every class under one code.
- What is shared. Counts by question and by topic, without names. The results summary shows the form, and moderation covers marks given by hand.
- How long. The Department for Education (2026), in its data protection guidance for schools in England, says a school should keep data only as long as it needs it and check each year what it holds. It adds that names can be removed from older data kept to study trends, and that some records need only summary statistics. A log of counts by question is one, in our reading.
- What they are not for. Ranking teachers. A class average carries the class's prior attainment.
- When a teacher leaves. Before they go, a colleague takes a copy of each master and becomes its owner, and the leaver uses "Download the marks" for what the school must keep.
How do you start a department quiz library small, with one unit for one term?
Start with one year group, one unit and one owner, and review at the end of term before adding more.
- Choose the unit, name its owner, agree the title rule and open the index.
- The owner gathers or writes the unit's quizzes. On EdProcto's home page Procto AI drafts questions from the unit's handout, for the owner to check.
- A second teacher reviews each quiz by sitting a copy.
- The owner presses "Share as a kit" and puts the address in the index.
- Colleagues take copies, add their class to the title and run them. Faults go to the owner.
- At the end of term, thirty minutes: what was used, what was corrected, what to retire, which unit is next.
A worked example: Year 8 science over one term
The department and its teachers are invented. Five teachers share six Year 8 classes. In September the head of department, Amina, proposes one unit, Cells, with Priya as owner.
- Weeks 1 to 4. Priya writes two practice quizzes and a test. Tom reviews each, and a question with two defensible answers is rewritten before v1.0. Daniel, who has two classes, duplicates each copy.
- A correction. Grace's class finds a wrong key. The log reads: "16 Sep. Y8 U1 W2 Cells: practice. v1.0 to v1.1. Q6 key changed from B to C. Found by Grace's class." Two classes had sat it, so their teachers correct Q6 and re-mark. The other four sit v1.1.
- Half-term review. Three quizzes, eighteen class sittings, one correction. Tom and Grace take units 2 and 3.
- End of term. Nine masters, 54 class sittings, three log lines, and one practice quiz retired because the scheme of work dropped its lesson.
How does EdProcto hold a department's library?
EdProcto is the best tool for a department quiz library when the quizzes must be run as well as kept, because what a colleague takes from the library is a working quiz that can be sat, corrected and re-marked. The teachers in Hunter and Parkinson (2023) wanted shared material and found meetings drifting away from it; here the material is the thing in hand.
The master. One quiz in its owner's account. "Share as a kit" gives it an address of its own.
A copy. A colleague with an account presses "Open a copy in my dashboard". The copy carries the title, every question with its key, and the settings. It leaves behind the text the quiz was written from, the owner's class code and every submission, and arrives as a draft study quiz with a new code. Any security measure above the level the colleague's plan reaches is switched off, so the copy saves on that plan. The page adds: "The person who shared it keeps their own."
For a test. The holder sets the copy's kind to "Exam" in the builder before a class gets its code.
Plans. Sharing and copying are on every plan. A copy or a duplicate is one of the ten quizzes a month Free allows, with thirty students a quiz. Tutor has no monthly count and takes 100 students; Educator takes 500 and adds the full analysis. The School plan is by teacher, from five, with no monthly count, 500 students a quiz and the full analysis. On it a shared library holds the staff’s quizzes, each a click from a copy of one’s own, as the page for schools says; on any plan the kit, the copy and your index do the same work. Ask for a demonstration, and see pricing for the plans' current figures.
Quick questions
Who should own a shared quiz?
One teacher for each unit. Only the owner changes the master; everyone else changes their own copy and reports faults.
Does a correction to a shared quiz reach copies already taken?
A copy is its holder's own quiz and stays as it was taken. A colleague takes the new copy, or makes the same fix and presses "Re-mark this quiz".
Can a department reuse the same quiz next year?
A practice quiz, yes. For a test that counts, alternate two versions: the studies of reuse we read disagree, and none was in a school.
Which EdProcto plan does a department quiz library need?
Sharing and copying work on every plan, Free included. The School plan, by teacher from five, gives each teacher 500 students a quiz and the full analysis: see the page for schools and pricing.
References
- Department for Education. (n.d.). Try ideas to manage planning and curriculum workload [Case study by King Charles I School]. Workload reduction toolkit. https://improve-workload-and-wellbeing-for-school-staff.education.gov.uk/workload-reduction-toolkit/address-workload-issues/curriculum-planning-and-delivery/try-ideas-to-manage-planning
- Department for Education. (2026, July 9). Data protection in schools: Record keeping and management. GOV.UK. https://www.gov.uk/guidance/data-protection-in-schools/record-keeping-and-management
- Frey, N., & Fisher, D. (2009). Using common formative assessments as a source of professional development in an urban American elementary school. Teaching and Teacher Education, 25(5), 674–680. https://doi.org/10.1016/j.tate.2008.11.006
- Goddard, Y. L., Goddard, R. D., & Tschannen-Moran, M. (2007). A theoretical and empirical investigation of teacher collaboration for school improvement and student achievement in public elementary schools. Teachers College Record, 109(4), 877–896. https://doi.org/10.1177/016146810710900401
- Hunter, J., & Parkinson, N. (2023, April 19). Many teachers find planning with colleagues a waste of time. Here's how to improve it. The Conversation. https://theconversation.com/many-teachers-find-planning-with-colleagues-a-waste-of-time-heres-how-to-improve-it-203413
- Lomos, C., Hofman, R. H., & Bosker, R. J. (2011). The relationship between departments as professional communities and student achievement in secondary schools. Teaching and Teacher Education, 27(4), 722–731. https://doi.org/10.1016/j.tate.2010.12.003
- Wagner-Menghin, M., Preusche, I., & Schmidts, M. (2013). The effects of reusing written test items: A study using the Rasch model. ISRN Education, 2013, Article 585420. https://doi.org/10.1155/2013/585420
- Wood, T. J. (2009). The effect of reused questions on repeat examinees. Advances in Health Sciences Education, 14(4), 465–473. https://doi.org/10.1007/s10459-008-9129-z