Alleviate the tension between students and teachers.

Shows whether a student can't start, won't start, or doesn't know enough to start. Also flags students who rush or don't read.

Get the diagnostic What it is for

What it is for

When a student won't start, we usually add pressure. That helps if the problem is willingness. It backfires if the problem is fear or missing knowledge.

The test records how each student begins, how long they stay, and how they respond when the same task is framed differently. The viewer gives each student a starting hypothesis with its evidence, and compares your class across conditions.

Four things a stalled student can look like

  • Can't start

    What you see
    Stalls on open problems, then works once given a first step.
    What to try
    Offer an entry point: a first step or a smaller case.

  • Chooses not to start

    What you see
    Starts some tasks but not others, depending on stakes, context or subject.
    What to try
    Talk about why the work matters. Offer a choice of context.

  • Missing prerequisite knowledge

    What you see
    Misses the basic floor questions despite real effort.
    What to try
    Reteach the skill first.

  • Not reading, or rushing

    What you see
    Fails reading checks, or answers faster than anyone could read.
    What to try
    Treat their other results as provisional. Retest or talk first.

How it is built

Forward only
No going back. A refresh resumes the same screen.
Same task, different start
The same problem, open or with a first step given.
Stakes banners
Not graded, or your teacher will read it. Shows whether stakes change who starts.
Skipped-step answers
Multi-step problems, answer-only and guided. Wrong options match the skipped step.
Floor questions
Basic prerequisites. Separate "doesn't know" from "didn't try".
Reading checks
Three checks, early, middle and late, plus a reading-speed timing check.
Reasoning and a chained problem
Deduction problems and one chained multi-part problem.
Beyond math
Other-subject items and a retry-and-hint ladder, to see persistence.

Using it fairly

Treat each label as a hunch, not a verdict. A student sees only a few items per condition, so one result can be noise. Patterns across the whole class are more reliable.

Say what is recorded. The first screen tells students you can see their answers, their timing and what they type. Leave that line in.

Try it on one class first. The cutoffs that sort students into patterns are reasoned guesses, not yet tested on real students. Check them against what you already know about your class.

Get the diagnostic

Automatic uses both forms across a class.
Letters and numbers only.
Open it

Send students this link.

Version

2.0.0Released October 20, 2026
  • 2.0.0 Scoring or item meaning changes. Don't pool results.
  • 1.1.0 Items or screens added.
  • 1.0.x Wording, layout, fixes.

Results carry their version. The viewer warns when major versions are mixed.

2.0.0 October 20, 2026

  • Removed the "counts toward participation" banner: the test carries no credit. Stakes are now two levels, not graded and teacher reads it, three problems each per form.
  • Added session codes: students send results straight to the viewer. Pulling the cord and wordmark is interactive.
  • Results from 1.x used three stakes levels. Don't pool them with 2.x.

1.0.2 October 20, 2026

  • New "Eraser dust" palette and animated logos, shared by every page.
  • Items, scoring and results files unchanged.

1.0.1 October 20, 2026

  • Progress is now a cord that gains slack. Viewer header shows one bead per student.

1.0.0 October 20, 2026

  • First release: test, viewer and printable booklets.