Switch course ▾
Session 4. Looking at Pupil Work and Inquiry Cycles
Objective. Run a 30-minute protocol for looking at pupil work, and plan a 4 to 6 week inquiry cycle in which the team tries one change and checks pupils' work again.
Looking together at what pupils wrote or drew is the quickest way for a team to learn something new about teaching, because the evidence is concrete. It is also where teams go wrong, slipping into judging pupils or the teacher who set the task. This lesson goes beyond the steps of the protocol: it shows how to change the language of the conversation from labels to observations, and how to design an inquiry cycle that ends in a decision even when the result is disappointing.
Key ideas
Why look at pupils' work
- Pupils' work is the most direct evidence of what they learned. Without it, teachers talk about impressions: 'They did not understand' or 'They just do not read'. With it on the table, the group can say what pupils actually wrote or drew.
- Bring three to five pieces on the same task, names removed. A mix is more useful than only the best work: one strong, two middle, one that struggled. Bring the task as pupils saw it, with the instructions.
- The aim is to understand thinking, not to give marks again. Say this at the start: 'We are not grading. We are learning from this work.'
A 30-minute protocol: Notice, Wonder, Next
- Presenter introduces the task and the question (3 minutes). Everyone reads the work in silence (3). NOTICE (6): colleagues say only what they see, 'I see Pupil B drew two parts but wrote only one number', with no words like good or weak.
- WONDER (6): colleagues ask questions about the pupils' thinking: 'I wonder whether Pupil B knew what the 8 means.' The presenter listens and answers briefly (3). NEXT (6): the group suggests one thing to try in the next lessons. Close (3): the presenter says what they will do.
- The 'Notice and Wonder' idea appears in the ASCD article on protocols by Venables. Larger, more formal protocols exist, such as the Tuning Protocol of the School Reform Initiative, first tried in 1992 among five high schools. A short one is enough to start.
A collective inquiry cycle
- One meeting of looking at work gives ideas; a cycle turns ideas into learning. Steps: (1) a question from the pupils' work, (2) one change to try, (3) try it for 3 to 4 weeks, (4) collect the same kind of work again, (5) look at it together, (6) decide: keep, adjust or drop.
- Keep it small and honest: change one thing, use the same kind of task before and after, and compare what pupils write or draw, not only marks. If nothing changed, that is also a finding.
- A cycle of 4 to 6 weeks fits a quarter: two cycles per term at most. Use the shared goal from Session 1 to choose the question, so the PLC does not scatter.
Keep it honest, safe and light
- Confidentiality: pupils' names are removed, copies stay in the room or are returned, and nothing from the session is used to compare teachers. If the work shows that a lesson did not work, the problem is in the lesson, and the group helps to redesign it.
- Keep a one-page inquiry log: question, the change, the evidence collected, what we noticed, our decision. Paper is enough. It becomes the PLC's memory when members change.
- Do not collect more than you can read. Five pieces of work and 30 minutes beat 30 pieces and no time.
A collective inquiry cycle
The language of looking
Judgement comes fast: 'lazy', 'careless', 'weak', 'did not listen'. These words feel like explanations but they close the question. A helpful habit is to ask, 'What do you see that makes you say that?' and to rewrite the sentence in terms of the work. 'Careless' becomes 'two of the five answers leave out the units'. 'Lazy' becomes 'the answer is one sentence copied from the textbook'. 'Weak' becomes 'the pupil draws the correct picture but writes the wrong number'.
The presenter helps by bringing a question, not a request for praise: 'What do these answers tell me about how pupils decide that something is a reason?' A real question lets the group focus its noticing, and makes the Wonder step richer. The facilitator keeps a card with judgement words heard and reads them out gently at the end.
Cycles that finish
A cycle often fails not because the idea is bad, but because it never reaches a decision. Fix three things at the start: the work you will collect (same type of task, five answers per teacher), the dates (one before, one after, one meeting to look), and what you will count (for example, answers with a stated reason). Without a baseline there is nothing to compare.
If nothing changed, ask whether the change was really used: was the frame used twice a week, or only once? If it got worse, look for what pupils did differently. If it improved, resist claiming too much: five answers in two classes are a good sign, not a proof. Each ending leads to the same step: keep, adjust or drop, and write one line in the inquiry log so the learning is not lost.
Examples from the classroom
A question instead of praise
Scene. A Jizzakh Grade 3 teacher, Gulnora, brought six retellings of a short story. Instead of asking, 'Are these good?' she said, 'I want to understand why some pupils tell what happened but nobody tells why it happened.' The group noticed that every retelling used 'and then', and wondered whether pupils had ever been asked 'why'. The next step was a small change: one 'why' question in each reading lesson.
Insight. A real question from the presenter focuses the group's attention and leads straight to a small change.
From 'careless' to a pattern
Scene. Grade 8 physics teachers in Khorezm brought lab write-ups. First comment: 'They are careless.' The facilitator asked, 'What do you see?' After a minute the group had counted that all 10 write-ups listed the steps, 9 had a table of results, and none stated a prediction before the experiment. The wonder: 'Do pupils know what a prediction is?' The cycle tested a one-line prediction written before each experiment.
Insight. Describing the work turns a vague complaint into a specific, testable pattern.
When nothing changed
Scene. A Samarkand team used a sentence frame for four weeks, but the second set of answers looked like the first: 3 of 10 had a reason. Instead of dropping the idea, they checked the log. The frame had been used once a week in two classes, not twice as planned. They decided to adjust and repeat: two uses a week for four more weeks, and a note on how pupils reacted.
Insight. A result of 'no change' is information; check whether the change was really used before deciding.
Common pitfalls
- Bringing only the best pupils' work so that the group learns nothing about difficulty.
- Allowing 'lazy', 'weak' or 'careless' to stand instead of asking what is seen in the work.
- Starting a cycle without agreeing what will be collected and counted at the end.
- Claiming that a method 'works' after five answers from two classes.
Look at Work, Then Plan a Cycle
- Open the Activity Kit: a SAMPLE task with four anonymous pupil answers, the protocol card, the record sheet and the inquiry cycle plan. If you brought five work samples from Session 3, use them for the second round. (5 min)
- Alone (5 min): read the four answers in silence and write three things you NOTICE, using only facts.
- Groups of four (30 min): run the protocol with roles (presenter, facilitator, timekeeper, note-taker). Fill the record sheet. The facilitator stops any 'good' or 'weak' label and asks 'What do you see that makes you say that?'
- Pairs (10 min): second short activity, Wonder to question. Turn one Wonder into an inquiry question, one change to try and the evidence you would collect again, using the cycle plan.
- Alone (5 min): write your school-based task from the Kit.
Your facilitator will guide you through this activity in the session.
Questions to discuss
- Which statement did the group have to change from a judgement to something seen?
- What did the pupils' work show that the teacher's memory of the lesson did not?
- Which change in your cycle plan is small enough to try in the next four weeks?
Reflect
- When did you last look at a pupil's work without grading it? What did you notice?
- What would make a colleague nervous about bringing work to the group, and how can you reduce it?
- Which question about your pupils have you wanted answered for a long time?
Try it tomorrow
- Choose a task whose answers you can read in five minutes and photocopy five pieces with names covered.
- Write your own question as a presenter, in the form 'I want to understand why...'.
- Agree what you will count after the cycle (for example, answers with a reason) and when the second collection takes place.
Further reading and links
Put pupils' work on the table, describe before you judge, then try one change for a few weeks and look at the same kind of work again.