Tutoring system: structure, accountability, data

Tutoring works when three things are locked in: a fixed structure, students who are held to it, and data that changes what happens next. Without the data loop, progress checks become paperwork. A review of progress-monitoring research found achievement improved when teachers used the data to change instruction, and teachers often collect it without acting on it (Stecker, Fuchs & Fuchs 2005, via Leiden(opens in a new tab)).

1. Structure (the same every session)

Fixed rules: at least three sessions a week, during the school day, groups of one to three, one structured program, the same tutor where possible (Nickow et al.(opens in a new tab)).

2. Student accountability

• Entry conference. Student, tutor, and family agree on one goal in plain words ("Read 60 words a minute by March"; "All multiplication facts in 3 seconds").
• The student keeps the graph. Each week the student plots their own probe score against the goal line. They should be able to say where they are and what's next.
• Attendance is expected, not optional. Target 3 of 3 sessions. Missed sessions are made up within the week.
• Mastery before moving on. A skill is "done" when the mastery check is passed on two separate days, not when the lesson is taught.
• Family update every two weeks. The graph, the next goal, and one thing to practice at home.

3. Data tracking

Decision rules (common practice in progress monitoring; the site grades the rules themselves as Promising):

• After at least six weekly probes, compare the trend with the goal line.
• Four probes in a row below the goal line: change something. Check attendance first, then the skill level, then the program.
• Four in a row above: raise the goal.
• At or above benchmark for three weeks plus mastery of the unit: exit, then check monthly.

The tracker (one row per student, reviewed in a 20-minute weekly data meeting)

What the evidence says about this system

• Proven: frequent, in-school, small-group tutoring with a structured program (principle 12).
• Strong: using progress data to change instruction, rather than just collecting it (Stecker et al.).
• Promising: specific decision rules and student self-graphing. They are widely used and sensible, but less directly tested.
• The scale warning: Nashville's large program showed small effects (Kraft et al.(opens in a new tab)). Its schedule put many sessions before or after school and it was still building its program, which is exactly what this structure is designed to prevent.