Switch course ▾
Session 5. Review, Honest Reporting, and Protection Against Gaming and Blame
Objective. Review progress on a calendar with decision rules, write an honest one-page report for the school board and parents, audit a plan for gaming and use blame-free language.
The last part of a cycle is where good plans are made safe. If the review is run on a calendar with rules set in advance, the school learns even when the target is missed. If the report is honest and short, the board and parents can trust it. And if gaming and blame are named early, the numbers stay a tool for learning instead of becoming a weapon. This session gives you routines for all three.
Key ideas
A review calendar with decision rules
- Fix the review points in advance: at 30 days, at the end of each term and at the end of the year. At each point ask three questions: what happened, why do we think so, what do we do now?
- At 30 days, look at leading indicators only: is the action actually happening? Agree a decision rule before you see the numbers, for example 'if fewer than 4 of 6 teachers set the task by week 4, we fix the timetable'.
- At term and year end, end each review with a stop, keep or change decision for every action, and write down what you learned even if the target was met.
One honest page for the board and parents
- Five boxes: where we started, what we did, what changed (in pupils and percent), what we do not know, and the next step. Use a hand-drawn bar chart and plain words, and keep names out.
- Report good and bad news in the same tone. A missed target with a clear lesson earns more trust than a hidden miss. Say 'from 67 to 70 pupils' and avoid claiming that your plan alone caused the change.
- Reports on school data can help communities ask good questions, but only when they point to action. IIEP's review of school report cards finds that many do not link information to anyone who can act on it.
How a measure gets gamed
- Gaming raises the number without raising learning. Common routes: easier tests or generous marks, coaching only pupils near the pass line, keeping weak pupils away on test day, teaching only what the test asks, and moving the strongest teachers to tested classes.
- The pressure rises when one number carries high stakes (the principle in course 11.6). The more a figure is used to reward or shame, the more it is bent. Targets for learning work best when the stakes for any one number are low.
- Safeguards: use more than one measure, compare school marks with an outside test or a blind-marked sample, count every enrolled pupil, check areas the test does not cover, and give staff a safe way to say 'this looks wrong'.
Blame-free review
- Ask what in the system produced the result, not who is at fault. Use 'we' and 'the class', name the starting point and supports, and keep individual teachers and pupils out of board papers.
- Keep improvement review separate from staff appraisal (course 11.1 explains why coaching is not evaluation). If teachers fear that data will be used against them, they will stop giving honest data.
- Share good news widely and thank the person who reports bad news early. Then close the cycle: the year-end review becomes the first diagnosis of the next cycle.
The review calendar
Decision rules before the numbers
People read numbers through hope. If the result is good, they believe it; if bad, they explain it away. A decision rule written before the data arrive removes much of that: 'if fewer than 4 of 6 teachers set the task by week 4, we fix the timetable'. The review then asks what the rule says, not who is to blame.
Plan three branches. Delivery low: fix the obstacle. Delivery good but pupils not moving: revisit the theory of change and look at the weakest assumption. Everything on track: still check one outside measure, such as a sample of papers marked blind by a teacher from another school, to make sure the good result is real. Record each decision in one line.
An honest page that parents can read
A one-page report works when it uses the same words for the same things, gives numbers in pupils as well as percent, and states what is not known. Say 'from 67 to 70 pupils' and 'two pupils move the percent by about three points'. Avoid league-table language, show the starting point and name the support given, so a class that began lower is not read as a worse class.
Research on school report cards suggests two cautions. Information alone changes little unless it is tied to people who can act, so end with a next step and a name of who will do it. And wider measures than test results give a fairer picture, so include attendance, pupil views and one example of pupils' work next to the test figure.
Examples from the classroom
The day-30 rule in Karshi
Scene. In Karshi, six maths teachers agreed in September: if fewer than four set the weekly explain task by week 4, the head fixes the timetable. In week 4, only three do. The deputy, Aziz, does not discuss anyone's commitment. He asks what stopped them and learns that the task sits in a lesson that is often cancelled for assemblies. The team moves it to another lesson. By week 6, five of six set it.
Insight. A rule set in advance turns a low number into an obstacle to fix, not a person to blame.
Blind marking in Samarkand
Scene. A Samarkand school meets its target: 80 percent of Grade 8 at mark 3 or above. Before reporting, Dilnoza asks a teacher from another school to blind-mark 20 papers. By that marking 14 of 20 reach mark 3, against 17 by the school's own marks. The report says: 'Our marks may be a little generous; with 20 papers this is a warning, not proof. We will agree a common scale and double-mark a sample next term.'
Insight. An outside check keeps good news honest, and reporting it builds trust.
The alarm in Namangan
Scene. In Namangan, a head asks Grade 9 class teachers to focus extra lessons only on pupils close to the pass line. A teacher, Jasur, says quietly, 'The others will be left behind and our figure will not show it.' The head thanks him and changes the plan: extra lessons for all pupils below the line, counting every enrolled pupil in the report, and an anonymous box where staff can raise concerns.
Insight. A safe way to say 'this looks wrong' protects pupils and the honesty of the figure.
Common pitfalls
- Deciding what the numbers mean only after seeing them, so hope and fear decide.
- Reporting only the subjects or classes that met the target and hiding the rest.
- Using a single figure to judge a teacher or class, which invites gaming and silence.
- Treating a met target as proof, without any outside check.
Day-30 Review, Year-End Report and Gaming Audit
- Open the Activity Kit. Pairs (10 min): read Table A, the SAMPLE day-30 leading data, and apply the decision rules agreed in advance. Decide for each action: keep, change or stop, and what you would say to staff.
- Trios (15 min): read Table B, the SAMPLE year-end data (including an outside-marked sample). Write the one-page report in the five-box template, in plain words for the school board and parents.
- Trios (15 min): run a gaming audit of the plan from Session 3 in Table C. List five ways the target could be reached without helping pupils, rate how likely each is, and write a safeguard and an owner for the top three.
- Second short activity (10 min): rewrite the three blame sentences in Table D as system sentences, then in pairs give a two-minute spoken report to 'parents' using your five boxes.
- Whole group (10 min): two trios read their reports. The group checks that each report has a clear 'what we do not know', a number in pupils and no names. Pick one gaming safeguard to adopt in every school plan.
- Alone (10 min): write your school-based task from the Kit and the date of your first review point. Set the date for starting your next cycle's diagnosis.
Your facilitator will guide you through this activity in the session.
Questions to discuss
- What did the outside-marked sample tell you that the school's own marks did not?
- Which sentence in your report was hardest to write honestly, and what made it hard?
- Which gaming route is most tempting in your school, and who could safely raise the alarm?
Reflect
- What would you do if the number your board wants most could rise by a trick that harms pupils?
- Who in your school would be willing to blind-mark a sample of papers with you?
- What is the first small step in your next cycle, and who needs to hear about it first?
Try it tomorrow
- Write three decision rules for your plan before your first review: one for low delivery, one for good delivery without pupil change, one for on track.
- Draft the five-box report for your plan, including one honest sentence on what you do not know.
- Ask one colleague from another school or subject to blind-mark ten papers and compare with your own marks.
Further reading and links
Review on a calendar with rules set in advance, report one honest page with what you do not know, guard the numbers against gaming, and talk about the system, not about blame.