Sign up free

Extra time in tests: how much, and how to set it per student

What studies and examining bodies say about extra time in tests, how to work out the minutes, and how to give named students a longer clock without fuss.

The EdProcto editorial teamWritten and checked against the research it cites. Last checked . How the guides are made

Guides · Teaching from the results · · 13 min read

How much extra time should a student get in a timed test, and how do I set it for named students?

Extra time in a timed test removes a barrier the test is not meant to measure; it is not a bonus. Your institution decides who has it. Add the agreed percentage, commonly 25%, 50% or 100%, to the class's time, round up to the next whole minute, and give that student a copy of the quiz with the longer limit.

Extra time, or extended time, is the accommodation universities most often provide to students with disabilities in exams (Gelbar & Madaus, 2021). It removes a barrier the test was never meant to measure, such as slow reading in a history test. It is not a bonus, nor a cure for a test too tight for everyone.

This guide is for the teacher, lecturer or exams officer who holds the list of names. Every source was read on 10 October 2026, and "abstract read" marks those where an abstract or an index record was all we opened.

EdProcto, which publishes this guide, gives a student a longer clock as a copy of the quiz with a limit and a code of its own. Time limits per quiz and per question covers the class's limit; its studies of extra time are not repeated here.

What is extra time for, and who decides who gets it?

Extra time is for a student whose condition slows the work in a way the test is not about. Your institution decides who has it, under the law where you teach; this guide gives no legal advice. Two public bodies show how the purpose is written down.

  • The U.S. Department of Justice (2020), in guidance on exams for admission, licensing and certification, says a testing body must run its exam so that the result reflects aptitude or achievement "rather than the individual’s impairment", unless the impaired skill is what the exam measures. Extended time is among its examples.
  • For the national tests in primary schools in England, the Standards and Testing Agency (2026) sets the boundary: "Access arrangements must never advantage the pupil." Support should follow normal classroom practice.

If speed is part of what the test measures, as in a mental arithmetic drill, more time changes the test: tell whoever agrees arrangements.

What does research say about extra time?

Where a test is tight, extra time helps students with and without a disability, and many who have it use little.

Does extra time help only the students who need it?

No. The idea called differential boost is that a fair arrangement lifts students with a disability by more than everyone else; reviews do not find that reliably.

  • Cahan et al. (2016) reanalysed 11 experiments: about the same proportion of students without a disability benefited from extra time as of students with a learning disability (abstract read).
  • Duncan and Purcell (2020), reviewing studies of students with specific learning difficulties, found no consensus; the outcome seemed to turn on how suitable each study's time limit was (abstract read).
  • Lewandowski et al. (2008) gave 64 high school students, half with a reading disability, a reading test under several time conditions. Those without a disability gained more, yet extended time let those with one attempt as many questions as their peers did in standard time (abstract read).

Our reading: time the class generously first. Where nearly everyone finishes, an allowance gives room without giving marks.

How much of their extra time do students use?

Often less than they are given. Gelbar and Madaus (2021) took the records of 596 students with disabilities at one large research university, 3,726 exams in all: extended time was used in about half, and where it was, about as many sittings ran under 50% more time as over it (abstract read). So record the minutes each student uses: whoever reviews the arrangement needs them.

What is known about rest breaks and separate rooms?

Less than their popularity suggests. We found no experiment that set a rest break against added minutes for the same students.

  • Saka et al. (2025) compared people who sat the Psychometric Entrance Test twice, in ten years of records that included 416 test takers with ADHD. Those who moved from no arrangement to breaks in a small group gained significantly more than either comparison group (abstract read; nobody was assigned at random).
  • Pritchard et al. (2016) read the school records of 96 pupils with ADHD in grades 3 to 8 in Maryland. Extended time was on 88% of their support plans, a reduced-distraction room on 77%, more frequent breaks on 45%. None of the five arrangements studied went with better state test scores; whether a listed arrangement was used on the day is not known.
  • Weis and Beauchemin (2020) assigned college students at random to sit a Spanish placement exam in a group or a separate room. In the group everyone did equally well; in the separate room, students with disabilities scored significantly lower than their classmates (abstract read).

What is asking for an arrangement like?

Hard enough that some go without. Marshak et al. (2010) interviewed 16 college students with disabilities at one state university, and Toutain (2019) reviewed 23 studies: both list negative reactions, feared or met, from other students and teaching staff among the barriers to using an arrangement (abstracts read).

Where do 25%, 50% and 100% come from?

From the rule books of examining bodies; neither page we read says how its figure was arrived at.

  • 25%. The Standards and Testing Agency (2026) allows up to 25% additional time to a pupil with an education, health and care plan, and to others on application. The unit test online has the share of GCSE and A level students with 25%, and a review of its basis.
  • 50% and 100%. The College Board (n.d.), which runs the SAT, names time and one-half and double time: its 64 minutes of reading and writing and 70 of mathematics become 96 and 106, or 128 and 140. The agency allows up to 100% with braille or modified large print papers.

Each figure is a ceiling: the College Board lets a student with extended time move on once the standard time has passed.

How do you work out the minutes?

Add the percentage to the class's limit and round up, never down.

  1. Start from the base time, the limit the class has.
  2. Add the percentage. 45 minutes with 25% is 45 × 1.25 = 56.25.
  3. Round up to the next whole minute: 57. Rounding down gives less than was agreed.
  4. Treat each timed part by itself. Where a question has its own clock, 45 seconds becomes 57 at 25% and 68 at 50%.
  5. Add a rest break in minutes. A supervised break stops the clock: "Break time does not count as testing time", in the College Board's words. The Standards and Testing Agency (2026) asks schools to weigh supervised breaks against additional time.

How is each arrangement set?

By a copy of the quiz for each allowance. The labels are those of EdProcto.

Arrangement What changes In EdProcto
25%, 50% or 100% extra time The overall limit A copy for each allowance, with a longer "Minutes overall"
A clock on each question Each question's seconds The copy's "Seconds a question"
A supervised rest break The clock stops The break's minutes added on that student's copy
No time limit The limit goes A copy with "Time limit" at "None"
A separate room The place The same link, on any device

Students with the same allowance share a copy. Supervise a rest break yourself and note its start and end.

How do you run it without singling a student out?

Let nothing a classmate can see differ: how the link arrives, the name on the screen, the moment of finishing. These are our working rules; we read no study comparing ways of doing it.

  • One way of sending. If three students get their link by private message, everyone does.
  • One title. The start screen shows the quiz's name, so give each copy the class's title.
  • One finishing time. Each clock starts when its student presses begin, so a longer limit can start earlier. Or set the whole class a silent task after the test, so nobody is watched working on.
  • A private word beforehand. Tell each student their minutes, when to begin, and to open only the link you sent.
  • No announcement. Never ask in front of the class whether the time is wanted.

Record it on a sheet of your own: the student, the allowance and who agreed it, the class's limit, the limit you set, the quiz's code and the minutes used. It describes a student's needs, so keep it where your institution keeps such records.

What do you do when the request comes on the day?

Check the list, then act by the stakes: entitlement is not yours to settle at the door.

  1. It is on the list and you missed it. Make the copy now, which in EdProcto is a duplicate, a new number and a link, and have the student begin there.
  2. The student is already inside the class's quiz. Leave its minutes alone while anyone is in it. Let them finish, note what was asked and when, and pass it on; a fresh sitting on a parallel paper is one remedy (retakes and make-ups).
  3. Nothing is on file. On a weekly quiz, our working rule is to give the time on a copy and report it that day: where the limit is generous, unneeded minutes add little. On an exam that counts, it is your exams officer's decision. For its own tests the Standards and Testing Agency (2026) accepts a late application only in exceptional circumstances, not because a school forgot.

What does one class of twenty-eight look like?

Three quizzes and one sheet. The class and every figure are invented: a Year 9 history test, a lesson from 10:00 to 11:00, a limit of 45 minutes, and a school list giving two students 25% and one 50%.

  1. The sums. 45 × 1.25 = 56.25, so 57 minutes. 45 × 1.5 = 67.5, so 68.
  2. The copies. "Duplicate this quiz" twice: one copy at 57 minutes, one at 68, each given the class's title.
  3. The timetable. The class begins at 10:00, stops by 10:45 and reads in silence until the bell. The two with 57 minutes begin with the class and stop by 10:57. The student with 68 begins at 9:50, before the lesson, and stops by 10:58.
  4. The screens. Twenty-five start screens say "Time limit: 45 minutes, submitted by itself when it runs out." Two say 57 and one 68.
  5. The record. "Download the marks" from each quiz: 25 rows, 2 and 1, pasted into one sheet of 28. "Minutes" shows 38.4 and 55.9 for the two and 51.2 for the third: one finished inside the class's time.

On Free this is three of the month's ten quizzes.

How does EdProcto give a student a longer clock?

EdProcto is the best tool for giving named students a longer clock, because each quiz's limit is held on its server from the moment an attempt begins, and each student is shown their own limit before the first question.

Making the copy

  1. In the builder of the class's quiz, press "Duplicate this quiz" at the foot. The copy opens at once: a draft titled "… (copy)" with the same questions, keys and settings, a code of its own and no submissions.
  2. Type the class's title over it.
  3. In its settings, under "Time", raise "Minutes overall", which takes 1 to 600. If "Time limit for each question" is at "Set", raise "Seconds a question" too (5 to 3,600). The copy keeps the class's "Warn the student with this many minutes left" and its hand-in at zero.
  4. Press "Publish & share" and give the copy's link to its students only.
  5. When they have handed in, press "Close the quiz".

What the student sees and the server holds

Each start screen states its own quiz's limit and nothing of any other: "57 min" under "Time", and under "House rules", "Time limit: 57 minutes, submitted by itself when it runs out." A clock on each question adds a line that begins "57 seconds for each question."

The questions arrive only once an attempt has begun (why it is built this way). The server records that moment and the clock on the page counts down from it by the server's own clock, so a device whose clock is wrong shows the same time as every other; a question's own seconds are counted by the page and kept with the attempt. Where "Submit by itself when time runs out" is on, the page hands in at zero, and anything reaching the server over two minutes past the limit changes no answer: the server hands the attempt in on what was saved in time, as of the moment the time ran out, and the attempt's record says so. A page that never hands in, because the tab was closed or the device slept, is handed in by the server the same way. Where the switch is off, so late a hand-in is marked in full and logged as "Handed in after the time ran out".

The marks

Each copy has its own results page. "Download the marks" gives a spreadsheet file, one row to a sitting handed in: "Name", "Email", "Sitting", "Started (your time)", "Handed in (your time)", "Score %", "Points", "Out of", "Reduced by %", "To grade", "Minutes", "Security events" and "Notes on the record", then a column for each question. "Minutes" is the time spent on the questions. Paste the copies' rows under the class's in your own sheet and work out the class's figures there.

Every plan has all of this. Free takes thirty students a quiz and ten quizzes a month, and each copy is one of the ten; Tutor takes 100 students a quiz; Educator takes 500 and adds the full analysis. Pricing has the figures; see also the pages for teachers and universities. A limit records the time given, not who sat; no setting makes cheating impossible, and every score is yours to release.

Quick questions

How much extra time should a student get in a test?

The amount your institution has agreed for that student. Examining bodies commonly state up to 25%, 50% or 100% (Standards and Testing Agency, 2026; College Board, n.d.).

Is extra time an unfair advantage?

Not on a test nearly everyone finishes. On a tight one, extra time helps students with and without a disability in similar proportions (Cahan et al., 2016).

How do you calculate 25% extra time?

Multiply the class's minutes by 1.25 and round up to the next whole minute: 45 becomes 57, and 60 becomes 75.

How do I give one student extra time in EdProcto?

In EdProcto, press "Duplicate this quiz", set a longer "Minutes overall" on the copy, and give its link to that student alone.

References

The settings

Time, attempts, feedback, grading, access, identity, security, tools, accessibility: every rule of a quiz, set by itself.

On every plan

See the settings

Read next