In short
- Six measures cover a reading program: participation, amount read and attendance for engagement, comprehension and level movement for outcomes, and staff time for operations.
- Each measure needs a written definition, a fixed way to collect it and a known trap.
- A measure used to make decisions comes under pressure to bend, so pair each measure with one that would expose its distortion.
- Keep the definitions the same every term, and drop any measure that has not informed a decision in a year.
A platform can produce dozens of numbers. A school leader needs a handful that answer three questions. Are pupils taking part? Is their reading improving? What does it cost the staff? This article sets out six program-level measures in those three groups. For each one it gives a definition, a way to collect it and the trap that comes with it. It then explains why any measure used as a target starts to bend, and how to pair measures so that each one checks another.
What KPIs should a school reading program track?
A school reading program needs six measures, in three groups. Engagement asks whether pupils take part: participation, amount read and attendance. Outcomes ask whether reading improves: comprehension and level movement. Operations asks what the program costs in staff time. Six measures are enough to judge a program and few enough to read on one page.
The list is practice, not a research standard. It follows the logic of the Education Endowment Foundation's guide to implementation. Sharples, Albers, Fraser and Kime (2019) ask schools to define clear implementation outcomes and to monitor them with robust and pragmatic measures. Participation, attendance and staff time tell you how the program was delivered. Comprehension and level movement tell you what it achieved.
Words retained is a good seventh measure if your records give it. How to Write a Term Reading Report for Your Board shows how five of these numbers fit on one page.
Six measures, defined
The table gives each measure a definition, a way to collect it and its main trap. Write your own definitions down, and print them under every report.
| Measure | Definition | How to collect it | The trap |
|---|---|---|---|
| Participation | Pupils who finished at least one story in the period, out of all pupils enrolled at the start | The reading record, checked against the class list | Counting pupils who opened a story but finished nothing |
| Amount read | Stories or words finished by the middle pupil and by the lowest quarter | The reading record, with the length of each story in words | A total or an average that hides pupils who read nothing |
| Attendance | Sessions attended out of sessions offered, by class | The register, or the platform's attendance record | Counting pupils who were present but did not read |
| Comprehension | Share of check questions answered correctly on stories at the pupil's level | Short checks after each story | Rehearsed or easy questions push scores up without more understanding |
| Level movement | Pupils who moved a level, or a step inside a level, out of those checked twice | A level check at the start of the year and the end of each term | Comparing different groups of pupils, or treating one test as final |
| Staff time | Hours per week given by teachers and the coordinator beyond normal lessons | A time log kept for two ordinary weeks each term | Leaving out setup, chasing pupils and writing reports |
Engagement: did the program reach pupils?
Engagement comes first because the other measures depend on it. If half the pupils read nothing, the outcome numbers describe only the other half.
Participation needs a fixed denominator: every pupil enrolled at the start of the period, not only those who logged in. Amount read is best given as the middle pupil and the lowest quarter, because a mean is pulled up by a few keen readers. Attendance tells you whether the planned sessions happened and whether pupils came. How to Track Attendance in Reading Sessions covers the register itself.
The EEF guide also shows how to check delivery. Sharples and colleagues (2019) give the example of reviewing timetables and measuring the frequency of sessions, to see whether an intervention was delivered as intended. For a reading program, that means counting the sessions that actually ran, not the sessions on the plan.
Outcomes: did reading improve?
Comprehension is the weekly outcome and level movement is the termly one. Both need care in how you read them.
Scores on a program's own checks tend to rise faster than scores on general tests. Kraft (2020) reports that effects on researcher-designed tests aligned with a program are often two to four times larger than effects on broad standardised tests. His evidence comes from research trials, but the lesson carries over. A rise in quiz scores on the stories pupils read is not the same as a rise in reading ability.
The EEF guide warns against a related overreach. Sharples and colleagues (2019) note that a survey on reading for pleasure often relates to how well pupils can read, but does not offer an accurate assessment of reading ability. If you run an attitude survey, report it on its own line.
Level moves slowly, so report steps inside a level as well as moves between levels. How to Measure CEFR Progress by Cohort shows how to count them for a fixed group of pupils.
Operations: what does it cost in staff time?
Staff time is the measure most often left out. It is also the one a leader most needs, because time spent on one program is time not spent on something else.
Levin and Belfield (2015) describe the ingredients method for costing an education program. It lists every resource the program uses, including personnel, puts a value on each, and calculates the total and the average cost per participant. Kraft (2020) adds that a full analysis counts non-monetary costs too, such as the opportunity cost of educators' time.
- Time teachers spend in reading sessions beyond their normal lessons.
- Time spent setting up classes, chasing pupils and marking.
- The coordinator's time, including writing reports.
- Training time, counted once, in the term it happened.
EdTech ROI for Schools: How to Judge a Reading Program's Value turns staff time and other costs into a cost per pupil.
When a measure becomes a target
Campbell (1979) stated a law that applies to every measure in this article. He wrote that 'the more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures'. The indicator then tends to distort the very process it was meant to monitor.
He gave an example from an employment office, drawn from earlier research. When staff were rated by the number of people placed in jobs, they concentrated on the easiest cases. The people who most needed the service were neglected.
Koretz (2008) describes the testing version as score inflation: increases in scores that do not signal a matching increase in what pupils know. The defence is to pair each measure with one that would expose its distortion. The pairs below are practice, not tested rules.
- Participation with amount read by the lowest quarter. Pupils counted as taking part should also be reading.
- Amount read with comprehension. More stories with falling scores suggests skimming.
- Attendance with amount read. Pupils can attend and still not read.
- Comprehension with level movement. Quiz scores can rise on familiar question types alone.
- Level movement with the number of pupils checked twice. A better result from fewer pupils is not better.
- Staff time with attendance. Cutting staff time is no saving if sessions stop running.
Do not tie pay, praise or rankings of staff to any one of these numbers. That is where the pressure Campbell described comes from.
How many measures, how often, and when to drop one
Six measures collected the same way every term are worth more than twenty collected once. Schildkamp (2019) describes data use as a cycle that starts by defining goals and ends with improvement actions and evaluation. A measure that never feeds an action has dropped out of that cycle.
Example: a coordinator's measure sheet for one term (invented numbers)
Grade 7, four classes, 104 pupils enrolled in week 1, twelve reading weeks.
Participation: 97 of 104 finished at least one story. Paired check: the lowest quarter finished 2 stories or fewer.
Amount read: the middle pupil finished 9 stories, about 18,000 words.
Attendance: 88 per cent of offered sessions attended. 46 of 48 planned sessions ran.
Comprehension: 74 per cent of check questions correct on stories at level, against 71 per cent last term.
Level movement: of 99 pupils checked twice, 21 moved a step inside a level and 4 moved a level.
Staff time: about 35 minutes per teacher per week beyond lessons, plus one hour a week for the coordinator.
Decision: keep all six measures and drop the minutes-online figure, which nobody has used in a year.
Review the list once a year. Keep the definitions fixed between reviews, and if one changes, say so next to the number. For the teacher's weekly view of the same record, see How to Measure Reading Progress in a Language Program.
How iRead handles this
In iRead, school leaders see levels reached, words retained and stories completed, by class, grade and term. Attendance at sessions is recorded automatically. Teachers see retention per word, per game and per class. A school can run a free pilot with one class for 4 to 6 weeks, with full dashboards, before any school-wide decision.
The core principle: Define each measure once, collect it the same way every term, and pair it with a measure that would expose its distortion.
See it in iRead: School leaders see levels reached, words retained and stories completed, by class, grade and term, and attendance at sessions is recorded automatically.
See the leadership viewKey takeaway
Track six measures in three groups, each with a written definition and a known trap. Pair them so they check each other, and drop any measure that no longer drives a decision.
Frequently asked questions
What are good KPIs for a school reading program?
Six measures cover most programs: participation, amount read and attendance for engagement, comprehension and level movement for outcomes, and staff time for operations. Define each one in writing, collect it the same way every term, and report the lowest quarter of pupils beside any average.
How many KPIs should a reading program have?
About six. That is enough to cover engagement, outcomes and staff time, and few enough to fit on one page. A long list collected once and rarely read informs fewer decisions than a short list reviewed every term with the same definitions.
What is Campbell's law?
Campbell (1979) wrote that the more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures. The indicator then distorts the work it was meant to monitor. In schools, a number that becomes a target can rise without real improvement.
Should KPI targets differ by grade or starting level?
Yes, if you set targets at all. Younger pupils and pupils who start lower may move at different rates, especially in level. Keep the definitions the same across grades so the numbers compare, and set any expectation for each group rather than one figure for the whole school.
How do you measure staff time for a reading program?
Ask each teacher and the coordinator to keep a simple time log for two ordinary weeks each term. Count time in sessions beyond normal lessons, setup, chasing pupils, marking and reporting. Count training once. Levin and Belfield (2015) treat personnel time as one of a program's costs.
Sources
- Campbell, D. T. (1979). Assessing the impact of planned social change. Evaluation and Program Planning, 2(1), 67-90. doi.org/10.1016/0149-7189(79)90048-X
- Koretz, D. (2008). Measuring Up: What Educational Testing Really Tells Us. Harvard University Press. www.hup.harvard.edu/books/9780674035218
- Schildkamp, K. (2019). Data-based decision-making for school improvement: Research insights and gaps. Educational Research, 61(3), 257-273. doi.org/10.1080/00131881.2019.1625716
- Sharples, J., Albers, B., Fraser, S., & Kime, S. (2019). Putting Evidence to Work: A School's Guide to Implementation. Guidance Report (2nd ed.). Education Endowment Foundation, London. d2tic4wvo1iusb.cloudfront.net/eef-guidance-reports/implementation/EEF_Implementation_Guidance_Report_2019.pdf
- Kraft, M. A. (2020). Interpreting effect sizes of education interventions. Educational Researcher, 49(4), 241-253. doi.org/10.3102/0013189X20912798
- Levin, H. M., & Belfield, C. (2015). Guiding the development and use of cost-effectiveness analysis in education. Journal of Research on Educational Effectiveness, 8(3), 400-418. doi.org/10.1080/19345747.2014.915604