Overview

Amal is an Arabic-reading decision system for teachers in the Arab world. For each child in Grades 1 to 4, it shows, reading skill by reading skill, what the child has learned in Arabic, suggests support to try next, tracks whether that support is working, and then re-measures. The teacher decides at every step.

This Platform Guide is the deep version of that story. It walks the whole system from end to end, with the real specifics: how the reading check works, how a status is decided per skill, how support plans are built and chosen, how progress is watched, and how reports reach teachers, parents, and school leaders. It is written for school leaders, partners, and evaluators who want a thorough understanding rather than a quick pitch.

The problem it addresses

A teacher can tell when a child is struggling to read. What is hard, even for an experienced teacher with thirty children, is saying precisely which reading skill is the sticking point and what to try next. A single grade or a rough impression hides the detail. Two children who both read “below level” often need very different help: one may be stuck on decoding letters and short vowels, while the other decodes fine but loses the meaning of a passage. Without a measure built for Arabic, that difference stays invisible, and the support a teacher gives is a guess.

Amal makes the difference visible. It measures Arabic reading the way Arabic actually works, rather than translating tools made for other languages, and it reports each child’s picture in plain growth language that a teacher and a parent can act on. It never reduces a child to one number, and it never attaches a clinical label.

A map of the whole platform

The guide is organized into five areas. Each is a major part of the teaching cycle, and each has its own deep section.

Measurement and decisions

The foundation. A short native-Arabic diagnostic adapts to the child and builds an internal reading-ability estimate per skill. A decision layer then turns that evidence into an honest status for each skill and each reading area, and it stops with a clear “not enough data yet” when the evidence is thin.

Read: How Amal Works for the full cycle, then the diagnostic and practice, skills taxonomy, skill-status engine, student profiles, and standards and benchmarks pages.

Intervention and teaching

What the platform proposes once a child’s needs are clear. A support plan is chosen from a fixed, curated set, built around one main skill plus a small number of linked skills and a strength the child already has. Children who can be taught together are grouped, and the teacher reviews and activates everything.

Read: intervention plans, grouping and activation, task delivery, and content and items.

Monitoring and response

How the platform checks whether the support is actually helping. Short repeated checks track progress, teacher-scored fluency reads are recorded, and a response layer decides whether to continue, adjust, or change the level of support, always asking first whether the plan was really delivered.

Read: progress monitoring, fluency checks, response decisions, and evidence and review.

Reports and dashboards

How the picture reaches people. Teachers see each child’s status per skill and per reading area. Parents receive a short, encouraging summary of their child’s growth. School leaders see reading health across classes and schools, without any public ranking of children.

Read: teacher reports, parent reports, and school and network reports.

Platform foundations

The parts that hold everything up. Who can see and do what, how organizations and classes are structured, the short non-reading warm-up that settles a child before a session, accessibility, the language-safety layer that keeps copy in growth terms, and the determinism that makes every past decision reviewable.

Read: roles and access, organizations and classes, cognitive warm-up, accessibility, language safety, and quality and determinism.

What is different

• Built for Arabic. The measurement is designed around how Arabic reading works, not adapted from a tool made for another language.
• The teacher is in control. Every recommendation is a suggestion the teacher accepts, adjusts, or sets aside. Nothing acts on a child on its own.
• Growth language only. Reports describe what a child can do and what to work on next, never clinical or comparative labels.
• No single overall score. Reading is reported per skill and per reading area, by design, because one number would hide the detail a teacher needs.
• Honest about gaps. When there is not enough evidence to judge a skill, the platform says so and waits, rather than guessing.
• No AI in measurement or in the child’s experience. Statistics measure and people decide. Language models only help teachers draft their own writing, never a child’s task or score.

Start here

If you are new, read in this order:

1. How Amal Works for the full cycle end to end.
2. Trust and Privacy for how children’s data and the teacher’s authority are protected.
3. Roadmap for what is available now versus planned.
4. Then any of the five areas above that matters most to your role.