In short
- An assessment is formative when its results change the next step in teaching or learning, whatever kind of check it is.
- Black and Wiliam (1998) found firm evidence that stronger classroom feedback brings substantial gains; later, stricter estimates of the effect are smaller.
- Exit questions, letter cards, mini whiteboards and one-sentence summaries show you in minutes who understood the story.
- Decide what you will do with each possible answer before you ask the question, and keep the checks ungraded.
Most reading teachers already ask questions about the story. Formative assessment is what happens next. You use the answers to decide what to teach, while there is still time to fix the problem. This article explains what formative assessment is and what the research found, including where the claims have been too large. It then gives quick checks of reading that take from thirty seconds to five minutes, and a plan for what to do with the answers.
What is formative assessment in reading?
Formative assessment is any check of learning whose results you use to decide the next step. Wiliam (2011) defines it by use. In his words, paraphrased, an assessment functions formatively to the extent that evidence of achievement is gathered, interpreted and used to make decisions about next steps in teaching. Those decisions should be better than the ones you would have made without the evidence.
So formative is not a type of test. It is a use of the results. A five-question quiz on a chapter is formative if the next lesson changes because of it. The same quiz is only a record if the scores go into a mark book and nothing changes.
Summative assessment does a different job. It sums up what a pupil has learned at the end of a unit or term, often for a report. Why Short Daily Quizzes Beat Termly Tests compares the daily quiz with the termly test. This article is about the minutes inside a lesson.
Completion is not formative evidence either. Knowing that a pupil finished the story tells you that reading happened. It does not tell you what the pupil understood.
What did the research find?
Black and Wiliam (1998) reviewed about 250 publications on classroom assessment. They found firm evidence that changes designed to strengthen the frequent feedback pupils receive about their learning bring substantial learning gains.
One experiment they describe shows that the form of feedback matters. Pupils who received only comments on their work improved between sessions. Pupils who received grades, or grades with comments, did not. The grades seemed to cancel out the help the comments gave. The tasks were thinking tasks, not reading.
Later reviews are more cautious about the size of the effect. Kingston and Nash (2011) examined more than 300 studies of formative assessment from kindergarten to grade 12. Only 13 gave enough information to calculate an effect. Their weighted mean effect was 0.20, well below the 0.40 to 0.70 often claimed.
A large trial ran in England. Anders and colleagues (2022) tested a light-touch formative assessment program, run by groups of teachers inside each school, in 140 secondary schools. The effect on national exam results was 0.09, or 0.11 in a sensitivity analysis. The program may also have narrowed the gap between pupils with high and low prior attainment.
| Study | What was studied | Finding | Limit for an EFL reading class |
|---|---|---|---|
| Black and Wiliam (1998) | Review of about 250 publications | Firm evidence that stronger feedback brings substantial gains | A broad review; not about foreign language reading |
| Kingston and Nash (2011) | 13 usable studies, from over 300, kindergarten to grade 12 | Weighted mean effect of 0.20; 0.32 in English language arts, 0.17 in mathematics, 0.09 in science | English language arts means pupils' first language |
| Anders et al. (2022) | Randomised trial in 140 secondary schools in England | Effect of 0.09 on national exam results | All subjects, older pupils, not EFL |
What does this mean for a reading teacher? The evidence supports the idea, but the gains in careful studies are modest. None of these studies looked at EFL reading classes. Use quick checks because they make your next lesson better aimed, not because a large effect is guaranteed.
The five strategies behind formative assessment
Leahy and colleagues (2005) set out five strategies that define formative assessment in the classroom. Here they are, each with a reading example:
- Share what success looks like. Before reading, say what pupils should be able to do afterwards: "retell the chapter in three sentences".
- Ask questions that reveal thinking. A why question about a character tells you more than a who question.
- Give feedback that moves pupils forward. Tell a pupil which paragraph to reread, not only the score.
- Make pupils owners of their learning. Pupils say, before the quiz, which part of the story they found hard.
- Use pupils as resources for one another. Pairs compare their one-sentence summaries and agree on the better one.
Quick checks of reading you can run in a lesson
Leahy and colleagues (2005) describe simple techniques that let a teacher see every pupil's answer, not only the answers of pupils with their hands up. The table adapts them to reading. One-sentence summaries and five-question quizzes are added from everyday reading practice.
| Check | How it works | Time | Best for |
|---|---|---|---|
| Exit question | Each pupil answers one question about the story on a card before leaving | 3 minutes | Seeing who understood the main event |
| ABCD cards | Pupils hold up a letter card to answer a multiple-choice question | 1 minute | A whole-class snapshot at a key point |
| Mini whiteboards | Pupils write a word or short answer and hold it up together | 1 to 2 minutes | Checking a word or a fact from the text |
| Traffic lights or thumbs | Pupils show a green or red card, or thumbs up or down, for how well they followed | 30 seconds | How confident pupils feel, not what they understood |
| One-sentence summary | Pupils write the chapter in one sentence of their own | 3 minutes | Whether pupils grasped the main idea |
| Five-question quiz | Questions on the plot, a word and the point of the story | 5 minutes | A fuller picture after each story |
Two cautions. A traffic light shows confidence, which is not understanding, so pair it with a question. And use no-hands-up questioning: Leahy and colleagues (2005) suggest choosing who answers, or picking at random, so that quiet pupils are heard too.
Classroom example: exit questions on an A2 story
Story: Omar finds a small wet dog on his way to football. He reads the address on its collar and walks it home. He misses the start of the match.
Question 1 (plot): Why was Omar late for football?
Question 2 (word): "He read the address on its collar." What is a collar?
Question 3 (point): Was Omar right to be late? Write one reason.
Sorting the cards: most pupils answer questions 1 and 3, but nine of 26 miss "collar". The next lesson opens with a picture of a collar and two new sentences that use the word.
The summary check is a comprehension strategy as well as a check. Reading Comprehension Strategies for EFL Students shows how to teach it. For quiz length, see How Many Questions Should a Reading Quiz Have?.
Hinge questions: one question that decides the next step
Leahy and colleagues (2005) describe a hinge-point question. It comes at the point where the lesson can go in different directions, depending on how pupils answer. In a reading lesson, ask it after a key passage, before you move on.
A good hinge question has wrong options that each reveal a different misreading. Then the letter a pupil holds up tells you what went wrong, not only that something did. This design rule is practice; it has not been tested for reading classes.
Classroom example: a hinge question with telling options
Text: "Omar was late, but he did not mind. The dog was safe."
Question: How did Omar feel at the end? A) Angry, because he missed the match. B) Happy, because the dog was home. C) Sad, because he lost the dog. D) Afraid of the dog.
What the answers tell you: A means the pupil read "late" but missed "did not mind". C means the pupil lost the plot. D points to a problem with the word "safe".
If most pupils hold up B, move on. If many hold up A, reread the two sentences together and talk about "did not mind".
What to do with the answers
A check is formative only if something changes. Decide before the lesson what you will do with each kind of result. The table below is a rule of thumb from practice, not a research finding.
| What you see | Likely cause | What to do next |
|---|---|---|
| Most of the class misses the same question | The teaching or the text, not the pupils | Reteach that point and ask about it again in a new form |
| Pupils get the plot but miss the word questions | Too many new words | Meet the words again in new sentences before reading on |
| A few pupils miss what the rest get right | The text may be too hard for them | Check their level and offer an easier story |
| Pupils get the facts but miss the point | Reading for detail, not meaning | Ask what the story teaches and discuss the answers |
| Everyone gets everything | The check was too easy | Ask a harder inference question next time |
Feedback to pupils matters as much as your next plan. Hattie and Timperley (2007) found that feedback is among the strongest influences on learning, but that its effect can be positive or negative. Good feedback answers three questions: Where am I going? How am I going? Where to next? Feedback about the pupil as a person, such as praise, tends to work least well.
On a reading check, that means pointing to the text: "Reread the second sentence. What did Omar see on the collar?" It does not mean "Good try" or a mark out of ten. Keep the checks ungraded. In the experiment Black and Wiliam (1998) describe, grades cancelled out the value of comments.
Making it a routine without extra marking
Formative assessment should not turn into extra marking. These steps keep it inside the lesson. They are practice, not a tested program:
- Plan one check per lesson, at the point where the lesson could go two ways.
- Write the question, the right answer and what each wrong answer would tell you.
- Look at the answers in the lesson, or sort exit cards into piles in two minutes afterwards.
- Change the start of the next lesson because of what you saw.
- Once a week, look back: which questions did most pupils miss, and did the reteaching work?
Between lessons, practice questions with feedback extend the same idea. Practice Questions Before a Quiz: What Helps explains what makes that practice useful.
How iRead handles this
In iRead, a quiz runs after every session, and practice questions are available between sessions. The questions are generated from the story itself: its real sentences, its grammar, the plot and the moral. Teachers see retention per word, per game and per class. School leaders see levels reached, words retained and stories completed, by class, grade and term.
The core principle: A check is formative only when its answers change what you teach next. Plan the change before you ask the question.
See it in iRead: Teachers see retention per word, per game and per class. School leaders see levels reached, words retained and stories completed, by class, grade and term.
See the leadership viewKey takeaway
Formative assessment is a habit, not a test: ask a quick question that every pupil answers, look at the answers, and change the next lesson. The research supports it, with gains that are real but modest.
Frequently asked questions
What is an example of formative assessment in reading?
An exit question is the simplest. Before pupils leave, each one answers a question about the story on a card, such as why a character acted as she did. You sort the cards in two minutes and start the next lesson with whatever most pupils missed.
What is the difference between formative and summative assessment?
Formative assessment checks learning while it is happening, so you can change your teaching. Summative assessment sums up learning at the end of a unit or term, usually for a grade or a report. The same quiz can be either, depending on what you do with the results.
Should formative checks be graded?
Mostly not. In an experiment described by Black and Wiliam (1998), pupils given comments improved, while pupils given grades, with or without comments, did not. Keep exit questions and hinge questions ungraded, and save grades for tests designed to measure.
How big is the effect of formative assessment?
Smaller than often claimed. Kingston and Nash (2011) found a weighted mean effect of 0.20 in the studies they could use, higher in English language arts. A trial in 140 secondary schools in England found 0.09 on national exams (Anders et al., 2022). Modest, from a program with only light-touch training.
How much lesson time does formative assessment take?
Very little. Most checks in this article take between thirty seconds and five minutes. The time that matters comes afterwards: a few minutes to look at the answers and decide what to change. A check you never act on is time wasted.
Sources
- Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education: Principles, Policy & Practice, 5(1), 7-74. doi.org/10.1080/0969595980050102
- Wiliam, D. (2011). Embedded formative assessment. Solution Tree Press. cloudfront-s3.solutiontree.com/pdfs/Reproducibles_EFA/The-Main-Idea-Embedded-Formative-Assessment-March-2013.pdf
- Leahy, S., Lyon, C., Thompson, M., & Wiliam, D. (2005). Classroom assessment: Minute by minute, day by day. Educational Leadership, 63(3). eric.ed.gov/?id=EJ745452
- Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81-112. doi.org/10.3102/003465430298487
- Kingston, N., & Nash, B. (2011). Formative assessment: A meta-analysis and a call for research. Educational Measurement: Issues and Practice, 30(4), 28-37. doi.org/10.1111/j.1745-3992.2011.00220.x
- 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. doi.org/10.1080/19345747.2021.2018746