# Reading quiz results as a teacher: what to reteach tomorrow

**How do I read quiz results to decide what to reteach tomorrow?** Reading quiz results as a teacher takes about ten minutes in a fixed order: topics before students, then the question most of the class missed and the wrong answer most of them chose, then names. Reteach a topic under 60% in a different way, regroup when a few students share a gap, and move on at 85%. The cut-offs are working rules.

Published 2026-10-10 by EdProcto. https://edprocto.com/guides/reading-quiz-results-as-a-teacher-what-to-reteach-tomorrow

The quiz closed at the end of the last lesson and tomorrow's plan is still open. Reading quiz results as a teacher is neither marking nor statistics: it is ten minutes that turn a page of scores into one decision for the class, one for a group and one for a few names.

The order matters more than the figures: topics before students, then the question most of the class missed, the wrong answer most of them chose, and only then names. The habit is often called responsive teaching: reteach, regroup or move on.

EdProcto, which publishes this guide, sets its [results](https://edprocto.com/#results) out in that order: see its page [for teachers](https://edprocto.com/solutions/teachers). The cut-offs here are our working rules, and labelled so.

## How do you read quiz results in ten minutes?

Read them in a fixed order, the class before the names, and leave each step once it has given you a decision. The timings are our estimate for a quiz of ten to fifteen questions.

1. **Clear what is waiting (1 minute).** An open answer not yet graded counts as zero in the scores. EdProcto's [results](https://edprocto.com/#results) say how many are waiting.
2. **Topics, weakest first (2 minutes).** One line per topic: its mean and the number of questions behind it. Label each reteach, watch or secure.
3. **The question most of the class missed (2 minutes).** In the weakest topic, find the lowest question. Check its key and wording before you blame the teaching.
4. **The wrong answer most of them chose (2 minutes).** Open the options. One wrong answer far ahead of the rest is a shared idea; misses spread evenly are a gap.
5. **Names (3 minutes).** Only now the students: who is under half on a topic the class holds, who rushed, who got everything right.

Then write three lines on tomorrow's plan: what the whole class meets again, who sits with you for ten minutes, and what you drop because it is secure.

## Why read by topic before by student?

Because a topic mean rests on the whole class's answers and points at a lesson, while one student's score on a topic rests on two or three answers and points at very little.

In a class of 28, a topic with three questions is 84 answers, and one changed answer moves its mean by 1.2 points. One student's score on that topic moves by 33 points on a single answer. A student's one out of three is a reason to ask, not a diagnosis.

Names first also turns the plan into a list of pupils to rescue. Datnow and Hubbard (2015), reviewing a decade of research on teachers' use of assessment data, report that changes made from data often focus on struggling students, and call that an equity concern. Topics first keeps the question on the teaching.

## What does the question most of the class missed tell you?

It tells you one of four things, and only the last two are about your teaching: the key is wrong, the wording is unclear, the content was never taught, or the class shares a wrong idea.

### Check the key and the wording first

A wrong key looks like a disaster in the class and is a slip in the quiz. When your strongest students agree on another option, reread the question before you plan anything: EdProcto prints "Check the key" for that pattern. [What the numbers mean](https://edprocto.com/guides/item-analysis-for-teachers-what-the-numbers-mean) explains the figure behind it, and [when to throw a question out](https://edprocto.com/guides/when-to-throw-a-question-out-after-the-exam) covers a test that counts.

### Then read the wrong answer most of them chose

A common wrong answer is a message about what the class believes. Ask what a student would have to think for that option to look right. In the example below, x² + 9 is what squaring each term of (x + 3)² gives.

The habit this replaces is "they were careless". Oláh et al. (2010) interviewed every third- and fifth-grade teacher in five Philadelphia elementary schools three times in a year about benchmark results in mathematics. The teachers did not use the results to make sense of conceptual understanding: they tended to read errors as procedural missteps and to respond with procedure. Read the option as an idea first and a slip second.

[Finding misconceptions in a quiz](https://edprocto.com/guides/misconceptions-finding-them-in-a-quiz-and-fixing-them-in-class) tells a shared wrong idea from a slip, a gap and a bad question.

## Who needs a word, and who is ready to move on?

A student needs a word when their result departs from the class's pattern, and is ready to move on when they were right where the class was wrong.

- **Under half on a topic the class holds.** A handful of students with none or one of three right, on a topic at 75%, are a group, not a lesson.
- **Fast and low.** EdProcto notes a student who "Finished in a quarter of the usual time with a low score". That prompts a conversation and proves nothing.
- **Everything right, the weak topic included.** These students need something better tomorrow than sitting through the reteach.

When EdProcto reports that a quiz "measured fairly consistently", it adds that small differences between students are not to be trusted: do not rank the middle. A score is not a grade.

## What does the research say about reteaching, regrouping and moving on?

It says that teaching adjusted to assessment evidence helps a little at scale, that choosing the next step is the hard part, and that a second explanation must differ from the first.

### What it supports

- **Formative assessment, modestly.** Anders et al. (2022) report a randomised trial of a formative assessment programme in 140 English secondary schools: an effect size of 0.09 on national examinations. [Common assessments](https://edprocto.com/guides/common-assessments-one-quiz-across-every-class-in-a-year-group) explains why the larger figures often quoted are disputed.
- **The next step is the hard part.** Heritage et al. (2009), working from measures of teachers' knowledge for teaching mathematics, found teachers better at inferring what students understood from assessment information than at deciding what to teach next.
- **Reteach differently.** Guskey (2010), describing mastery learning, separates corrective instruction from reteaching that restates the first explanation "louder and more slowly". Correctives add roughly 10 to 20% to a unit's time. A second check with different questions follows after a lesson or two. Students who showed mastery get enrichment that broadens the topic without moving them ahead, so the others can rejoin the sequence.

### What it does not say

- **No cut-off, and not one quiz.** No study we read tests 60%, 80% or any other line for reteaching, or measures one teacher reading one quiz.
- **Mastery learning is disputed.** Kulik et al. (1990), across 108 controlled evaluations, found positive effects on examination performance. Slavin (1987), reviewing group-based mastery learning in schools, found no evidence of gains on standardised achievement measures.
- **Regrouping is no guarantee.** Pooled studies give grouping inside a class modest support: [baseline tests](https://edprocto.com/guides/baseline-tests-for-a-new-intake) has the figures. But Faber et al. (2018) used 144 classroom observations to predict the mathematics achievement of 953 pupils: differentiated instruction showed no significant positive effect, and pupils in low-ability groups profited less than the others. Our reading: form a group for one idea and dissolve it after ten minutes.

We read Guskey (2010) in full and, on ERIC, the abstracts only of the other studies cited here.

## What decision rules can a teacher use?

Use three: reteach the class when a topic is under 60%, regroup when a few students are under half on a topic the class holds, and move on at 85%. EdProcto prints the 60% and 85% lines; the rest are our working rules, not findings. Move them for a small class or a topic the next unit stands on.

| What the results show | Decision | Tomorrow |
| --- | --- | --- |
| A topic mean under 60% in EdProcto's [results](https://edprocto.com/#results) | Reteach the class | Fifteen to twenty minutes taught another way, then two new questions |
| A topic at 60% to 85% | Fix the leading wrong answer, or practise if the misses are scattered | Five minutes, then a new question |
| Three to eight students under half on a topic the class holds | Regroup | Ten minutes with you while the rest work on |
| One or two students under half | A word | Two minutes at the desk |
| A topic at 85% or above | Move on | Drop the recap; [ask again in a fortnight](https://edprocto.com/guides/re-tests-that-come-due-how-spacing-is-scheduled) |

Whatever you choose, check it within the week with a new question on the same idea. [Exit tickets](https://edprocto.com/guides/exit-tickets-three-questions-and-a-reflection) has the three-question version of these rules.

## What does this look like for one class? A worked example

For this class it comes to a twenty-minute reteach, a group of five and one dropped recap.

The class is invented, no student is real, and the figures come from running EdProcto's analysis code on these answers. Twenty-eight students took a twelve-question multiple-choice algebra check: four topics, three questions each, one point a question. Tomorrow's plan was factorising quadratics.

| Topic | Questions | Class mean | Students under half |
| --- | --- | --- | --- |
| Expanding two brackets | Q7 to Q9 | 47.6% | 16 |
| Solving equations | Q10 to Q12 | 77.4% | 5 |
| Expanding one bracket | Q4 to Q6 | 81% | 2 |
| Collecting like terms | Q1 to Q3 | 91.7% | 1 |

The last column is our own count from the marks file: students with none or one of a topic's three questions right. The page's first three sentences were:

- "Reteach before moving on: Expanding two brackets (47.6%)."
- "Secure, and safe to build on: Collecting like terms."
- "A shared wrong answer on Q8: the class was taught something, and it was not this. Worth a minute in the next lesson."

### The question most of the class missed

Q8 asked for (x + 3)² expanded, and 7 of 28 got it.

| Option on Q8 | Chose it | Their mean |
| --- | --- | --- |
| x² + 9 | 16 (57%) | 65.1% |
| x² + 6x + 9, the key | 7 (25%) | 98.8% |
| x² + 3x + 9 | 3 (11%) | 88.9% |
| 2x + 6 | 2 (7%) | 41.7% |

The key is sound: its takers outscored the sixteen on x² + 9. The same habit, first times first and last times last, gives the leading wrong answer on Q7 (x² + 10, six students) and Q9 (x² − 4, seven), and all thirteen of those choices came from the sixteen.

### The three decisions

1. **Reteach: expanding two brackets, whole class, twenty minutes.** Not the first explanation again. Draw a square of side x + 3 cut into four pieces: x², 3x, 3x and 9. Test both answers with x = 1: (1 + 3)² is 16, x² + 9 gives 10 and x² + 6x + 9 gives 16. Two new questions follow, (x + 5)² and (x − 2)(x + 6). Factorising, the same idea run backwards, takes the rest of the lesson.
2. **Regroup: five students on solving equations, ten minutes.** A topic at 77.4% hid five students with none or one of three right. Four of them answered x = 6.5 to 2(x − 3) = 10, which is what 2x − 3 = 10 gives. They sit with the teacher while the others practise.
3. **Move on: collecting like terms, with an extension for six.** The planned recap is dropped. Six students got all twelve right; during the reteach they investigate why consecutive square numbers differ by an odd number, starting from (n + 1)² − n². Expanding one bracket, at 81%, drew no sentence, but seven students chose −2x − 10 for −2(x − 5): one starter question next week.

One more name: a student scored 25% in 1:50 against a median time of 9:24, and the "Notes" column said so: a two-minute conversation, not a mark.

## How does EdProcto's results page give each reading?

EdProcto is the best tool for this job when the plan has to be written tonight, because its results page follows the order of the ten minutes and writes the reteach decision as a sentence, directly under the five headline figures.

- **By topic.** Each question in the builder has a field, "Topic (for the results by topic)". The page groups questions by the exact wording typed there, capitals included. "By topic" lists each topic's number of questions and mean, weakest first. "Reteach before moving on" names every topic under 60%; "Secure, and safe to build on" names every topic at 85% or above once five students have handed in.
- **The question most missed.** Under "Each question", "Got it" is the share of the points the class earned, and "Heat map" shows every question at once. Under 30%, the row reads "Check the key first, then whether this was taught."
- **The common wrong answer.** Click a multiple-choice row: each option shows "Chose it" and "Their mean", the average quiz score of the students who chose it. When a wrong option with three or more takers draws more students than the key, the row says so.
- **The names.** "Students" lists each score and time, with a note beside a fast finish with a low score, and a row opens that student's answers, each marked right, partly or wrong. For who is under half on one topic, press "Download the marks": a spreadsheet file with a row for each attempt and a column for each question.

With fewer than five submissions, the page says the numbers describe those students and not yet the class. To read before anyone sees a result, set "Show the score" to "When released" and press "Release the scores" when you are ready.

A class of twenty-eight fits the Free plan: ten quizzes a month, thirty students a quiz, marking by rule and the results question by question. Educator is the plan that lists the full analysis: discrimination, distractors, reliability and what to reteach. See [pricing](https://edprocto.com/pricing), and start on the [home page](https://edprocto.com/) in [multiple choice](https://edprocto.com/templates/mcq), where every wrong answer can be read.

## Quick questions

### What score means I should reteach?

No study we read tests a cut-off. EdProcto prints "Reteach before moving on" for a topic whose mean is under 60%, and we use that line for the whole class and 85% for moving on.

### Should I reteach the whole class or a small group?

The whole class when a topic is under 60% or one wrong answer leads on a question; a group when three to eight students are under half on a topic the rest hold. Both are working rules.

### What do I do with the students who got everything right?

Give them an extension on the same topic during the reteach: Guskey (2010) describes enrichment that broadens a topic without moving students ahead.

### Where does EdProcto show what to reteach?

At the top of a quiz's [results](https://edprocto.com/#results), in a sentence that names each topic under 60%. EdProcto's page [for teachers](https://edprocto.com/solutions/teachers) shows how the quiz is written and shared first.

## References

- Anders, J., Foliano, F., Bursnall, M., Dorsett, R., Hudson, N., Runge, J., & Speckesser, S. (2022). The effect of embedding formative assessment on pupil attainment. *Journal of Research on Educational Effectiveness, 15*(4), 748–779. [https://doi.org/10.1080/19345747.2021.2018746](https://doi.org/10.1080/19345747.2021.2018746)
- Datnow, A., & Hubbard, L. (2015). Teachers' use of assessment data to inform instruction: Lessons from the past and prospects for the future. *Teachers College Record, 117*(4), 1–26. [https://doi.org/10.1177/016146811511700408](https://doi.org/10.1177/016146811511700408)
- Faber, J. M., Glas, C. A. W., & Visscher, A. J. (2018). Differentiated instruction in a data-based decision-making context. *School Effectiveness and School Improvement, 29*(1), 43–63. [https://doi.org/10.1080/09243453.2017.1366342](https://doi.org/10.1080/09243453.2017.1366342)
- Guskey, T. R. (2010). Lessons of mastery learning. *Educational Leadership, 68*(2), 52–57. [https://www.ascd.org/el/articles/lessons-of-mastery-learning](https://www.ascd.org/el/articles/lessons-of-mastery-learning)
- Heritage, M., Kim, J., Vendlinski, T., & Herman, J. (2009). From evidence to action: A seamless process in formative assessment? *Educational Measurement: Issues and Practice, 28*(3), 24–31. [https://doi.org/10.1111/j.1745-3992.2009.00151.x](https://doi.org/10.1111/j.1745-3992.2009.00151.x)
- Kulik, C.-L. C., Kulik, J. A., & Bangert-Drowns, R. L. (1990). Effectiveness of mastery learning programs: A meta-analysis. *Review of Educational Research, 60*(2), 265–299. [https://doi.org/10.3102/00346543060002265](https://doi.org/10.3102/00346543060002265)
- Oláh, L. N., Lawrence, N. R., & Riggan, M. (2010). Learning to learn from benchmark assessment data: How teachers analyze results. *Peabody Journal of Education, 85*(2), 226–245. [https://doi.org/10.1080/01619561003688688](https://doi.org/10.1080/01619561003688688)
- Slavin, R. E. (1987). Mastery learning reconsidered. *Review of Educational Research, 57*(2), 175–213. [https://doi.org/10.3102/00346543057002175](https://doi.org/10.3102/00346543057002175)