Switch course ▾
Session 1. A Small Tested Change: Choosing a Problem Worth Solving
Objective. Tell a problem-first, tested change from a fad, and choose and write a classroom problem worth solving, using evidence from your own school.
Every school remembers an idea that arrived with excitement in September and was forgotten by December. Often the idea was fine; what was missing was a clear reason for it. When you start from a problem you can show, you know what to change, what to measure and when to stop. This session helps you choose a problem that is real, small and yours, so that your first innovation is a tested step and not a passing fashion.
Key ideas
Innovation is not a fad
- In this course, innovation means a small change in your own practice, made on purpose, checked with evidence and kept only if it helps pupils. It does not have to be new to the world; it has to be new to your class.
- A fad starts from a tool or a trend ('let us use this'). Innovation starts from a problem ('my pupils cannot do this'). The first question is never 'what shall I use?' but 'what problem do I have, and how do I know?'
- Three starting points: adopt a tested idea as it is, adapt it to your class, or invent something new. Most good school change is adapting, so look at colleagues and earlier courses before inventing.
Find a problem with evidence
- Look at what you already have: exit tickets (course 5.5), test results by outcome, pupil work, attendance, a short pupil question, a colleague's note from a visit (course 11.9). Evidence from two sources is stronger than one.
- A good problem statement says who has the problem, what it is and how many, and where the evidence comes from: 'In Grade 5B, 9 of 28 pupils cannot retell a text in their own words (exit tickets, four lessons).'
- Leave out blame ('lazy pupils') and leave out the solution ('they need group work'). Both close your mind before you understand the cause.
Five tests for a pilot problem
- MATTERS: it affects important learning in the standard, not a minor detail. EVIDENCE: you can show it with numbers or pupil work. YOURS: you can change it in your own lessons.
- SMALL: a change can be tried in four to six weeks in one or two classes. MEASURABLE: you can tell whether it improved, using something you can collect in a few minutes.
- Score each candidate problem 0, 1 or 2 on each test. A problem outside your control (the school timetable) goes to the principal as a request, not into your pilot.
Pupils are not guinea pigs
- Any change you test must be something you reasonably expect to help, or at least not harm. It must keep the standard's level of demand and must not take learning time from something essential.
- Tell the head of studies before you begin, follow school rules on informing parents, and keep pupils' names and marks out of anything you share (course 5.7). Take extra care with exam-year classes.
- One pilot at a time. A teacher who starts four changes in September has no way to learn from any of them.
From problem to small test
Why good ideas fade
Studies of school improvement keep returning to one point: the idea itself is only part of the story, and how it is introduced matters as much. The EEF guide to implementation describes four phases: explore, prepare, deliver and sustain. Fads usually skip the first. Exploring means asking whether the idea fits a real need in your school, whether you have the time and skills to do it well, and whether something you already use could do the job.
In practice, the question 'is this method good?' becomes 'is this method good for this problem, in this class, with our time?' The same method can be right in one school and wrong in another.
Symptom, cause and change idea
'Pupils are bored' is a symptom. Behind it may sit different causes: the text is too hard, the task is too easy, pupils do not see why it matters, or the lesson never changes pace. Each cause needs a different change, and a tool chosen before the cause is known may miss it. Looking at two or three sources, such as exit tickets, pupil work and one short question to pupils, usually shows which cause is likely.
Keep one question close: 'what would I see if this were solved?' If you cannot describe it in pupils' answers or work, the problem is still too vague to test.
Examples from the classroom
Retelling in Samarkand
Scene. Dilnoza teaches Grade 5 Uzbek reading in Samarkand. After four lessons her exit tickets show that 9 of 28 pupils cannot retell a text in their own words. At a seminar she saw a reading app and wants to try it. Instead she asks the class what is hard. Many say they read only to finish. She writes the problem as retelling, not as lack of an app, and plans a retelling routine.
Insight. The evidence pointed to a skill, so the change idea was built around the skill and not around the tool.
Tablets or tables in Fergana
Scene. Aziz teaches Grade 7 geography in Fergana. The school has bought six tablets and he is told to use them. He looks at the last test first: 14 of 30 pupils could not read a climate table. The problem is reading tables, not lack of devices. He designs a 15-minute printed practice with real temperature and rainfall data from the Fergana valley and decides to consider tablets only afterwards.
Insight. Starting from the pupils' difficulty showed that paper was enough for the first test.
Four candidates in Tashkent
Scene. Malika teaches in a Russian-medium primary school in Tashkent. She lists four possible projects: a language club for the whole school, a new timetable, a speaking routine in her Grade 3 class where 4 of 25 pupils spoke Uzbek in a 10-minute observation, and a reading corner. She scores them on the five tests. The club and the timetable fail 'yours' and 'small'. The speaking routine passes all five.
Insight. A problem outside your control goes to the principal as a request; the one inside your lessons becomes the pilot.
Common pitfalls
- Choosing the tool first and looking for a problem afterwards.
- Writing a blame word or a solution into the problem statement.
- Picking a problem so large that no change in four to six weeks could show anything.
- Using only your impression, with no number or pupil work to show.
Choose a Problem Worth Solving
- Open the Activity Kit. Pairs (10 min): in Table C, sort the six proposals into problem-first or tool-first. Rewrite two tool-first ones as a question about pupils' learning.
- Pairs (15 min): in Tables A and B, score the six sample situations 0 to 2 on the five tests. Choose the best two for a one-teacher pilot, and say why situation E cannot be one.
- Alone, then in trios (15 min): take real evidence from your own class (exit tickets, a test result sheet, pupil work) with names removed. In Table D write your own problem statement: who, what and how many, evidence, no blame, no solution.
- Second activity, pairs (10 min): the why-chain in Table E. Ask 'why?' up to five times about your problem until you reach a cause you can act on in your own lessons. Stop when the cause is outside your control.
- Trios (10 min): challenge round. Each reads their statement; the others ask three questions: How do you know? Is it yours to change? What would you see if it were solved? Revise one line.
- 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 tool-first proposal was hardest to turn into a problem, and what did the rewrite change?
- What did your evidence show that you did not expect?
- Which of the five tests did your first problem fail, and how did you fix it?
Reflect
- Think of the last new idea your school adopted. Did it start from a problem?
- Which pupils' problem do you see every week but have never written down with numbers?
- Who could challenge your problem statement honestly and kindly?
Try it tomorrow
- Tally your last two exit tickets or one test by outcome and write one sentence with who, what and how many.
- Score one problem on the five tests with a colleague and change one line of the statement.
- Ask your pupils one short question: 'what is hardest for you in this topic?' and record the three most common answers.
Further reading and links
Start from a problem you can show, keep it small and yours, and name the pupils' learning, not a tool.