Data-driven instruction means teaching from evidence of what students have learned, in a short cycle: assess with exam-style questions, analyse results by question and topic, reteach what was missed, then reassess. In secondary schools it works when data is small, frequent and used within days. School-wide data collections should stay at two or three a year.
What is data-driven instruction?
A cycle, not a spreadsheet. The term was popularised by Paul Bambrick-Santoyo’s Driven by Data (2010). In a Cambridge or Pearson Edexcel classroom it has four steps, as we recommend them:
| Step | In an IGCSE classroom | Timescale |
|---|---|---|
| 1. Assess | Short exam-style questions on the topic just taught, using the syllabus’s command words | Every one to two weeks |
| 2. Analyse | Results by question and topic: which items did most of the class miss, and why? | Within two days |
| 3. Act | Reteach the most-missed items to the class; target small groups for the rest | Next lesson |
| 4. Reassess | Similar questions a week or two later, to check the reteaching worked | One to two weeks |
The key difference from ordinary testing is step 3. Data only counts as “driving” instruction if it changes the next lesson.
How often should schools collect data?
Often in classrooms, rarely at school level. The DfE’s Teacher Workload Advisory Group was blunt. Unless data “can be collected with no marking or data inputting time outside teachers’ lesson times”, it saw “no reason why a school should have more than two or three attainment data collection points a year”. It added that “increasing assessment frequency is not inherently likely to improve outcomes” (Making data work).
| Level | Frequency | Purpose |
|---|---|---|
| Classroom | Weekly or fortnightly | Decide the next lesson |
| Department | After each mock | Find topic gaps across classes |
| School | Two or three times a year | Report progress and plan intervention |
The classroom cycle is compatible with the DfE’s advice when it runs in lesson time, for example with auto-marked questions or quick checks.
What principles should school data follow?
The DfE report sets four:
- “The purpose and use of data is clear, is relevant to the intended audience and is in line with school values and aims.”
- “The precision and limitations of data, and what can be inferred from it, are well understood.”
- The amount and frequency of data collected is proportionate.
- “School and trust leaders review processes for both collecting data and for making use of the data once gathered.”
It also says “pay progression should never be dependent on quantitative assessment metrics, such as test outcomes”. In our view, data used to judge teachers tends to distort the data.
What does the evidence say about acting on data?
That the feedback step matters most. The EEF’s guidance Teacher Feedback to Improve Pupil Learning groups its six recommendations into principles, methods and implementation (EEF):
| Group | EEF recommendation |
|---|---|
| Principles | ”Lay the foundations for effective feedback” |
| Principles | ”Deliver appropriately timed feedback that focuses on moving learning forward” |
| Principles | ”Plan for how pupils will receive and use feedback” |
| Methods | ”Carefully consider how to use purposeful, and time-efficient, written feedback” |
| Methods | ”Carefully consider how to use purposeful verbal feedback” |
| Implementation | ”Design a school feedback policy that prioritises and exemplifies the principles of effective feedback” |
In the cycle, step 3 is the feedback: to the class on common errors, and to individuals on their gaps.
What does the cycle look like in practice?
An example from IGCSE Chemistry, as an illustration:
- Assess: after teaching moles, a ten-question exam-style quiz, including two “calculate” and two “explain” items.
- Analyse: most students get the calculations right, but most drop marks on the “explain” questions about limiting reagents.
- Act: next lesson opens with a model answer for the explain question and a worked example on limiting reagents; four students who missed the calculations get a short group session.
- Reassess: two weeks later, three similar questions in the starter quiz show whether the gap has closed.
The same pattern works for any syllabus. For mocks, use question-level analysis. For the classroom techniques, see formative assessment strategies for IGCSE.
What goes wrong?
Common problems, in our view:
- Collecting without acting: data drops that change nothing.
- Grades instead of gaps: a predicted grade says nothing about what to reteach.
- Too slow: analysis a month after the test, when the class has moved on.
- Too much manual marking: the cycle stops when teachers can’t keep up.
- Using data to judge teachers, which the DfE warns against.
How schools do this with AI Buddy
AI Buddy runs steps 1 and 2 of the cycle automatically. Students answer past-paper-style questions, marking is instant, and teachers see results by question and topic across the class the same day. That leaves teacher time for step 3, reteaching, and makes step 4 easy, because similar questions can be set again. The instructional decisions stay with the teacher.
Frequently asked questions
What is data-driven instruction?
Teaching from evidence of what students have learned, in a cycle: assess, analyse by question and topic, reteach what was missed, and reassess.
What is the data-driven instruction cycle?
Four steps: assess with short exam-style questions, analyse results within days, act by reteaching in the next lesson, and reassess a week or two later.
How many data collection points should a school have?
The DfE’s Teacher Workload Advisory Group saw no reason for more than two or three attainment data collection points a year, unless data can be collected without extra teacher time.
What is an example of data-driven instruction?
A teacher gives a short quiz, finds most students missed one question type, reteaches it next lesson with a model answer, and checks with similar questions two weeks later.
Discover how AI Buddy helps schools strengthen teaching, learning and evidence-informed school improvement. Or start a short consultation with our schools team using the form below — we will get back to you directly.
Sources
- Department for Education, Making data work: report of the Teacher Workload Advisory Group (November 2018)
- Education Endowment Foundation, Teacher Feedback to Improve Pupil Learning (2021)
- Bambrick-Santoyo, P., Driven by Data (Jossey-Bass, 2010)