Number Line test online: place numbers from 0 to 1,000

A number appears above a bare line marked 0 at one end and 100 or 1,000 at the other, and you click where it belongs. The score is how far, on average, your clicks land from the right spot.

Start the Number Line test

2 blocks · about 6 min · mouse or trackpad

After block 2 you see your placement error for each range and can keep it with a free account.

2 blocks · about 6 min · mouse or trackpad

This task needs a mouse or trackpad and a larger screen. Open this page on a computer to take it.

Start the Magnitude Comparison test

Magnitude comparison works on a phone: tap the larger of 2 digits.

72 0 100
72 goes about hereBlock 1 · 0 to 100

How it works

4 practice numbers come first and are not scored, then 2 blocks of 44 numbers. Each placement takes one click and nothing is timed, so you set the pace: about 6 minutes.

  1. 1 · Practice

    4 numbers to start

    25 and 75 on a 0 to 100 line, then 500 and 250 on a 0 to 1,000 line. They show how the task works and do not count.

  2. 2 · Each number

    One click

    Click the line where the number belongs. A marker shows your choice for about a third of a second, then the next number appears. A click cannot be undone.

  3. 3 · Block 1

    0 to 100

    44 placements: 22 numbers between 3 and 96, each shown twice, never twice in a row.

  4. 4 · Block 2

    0 to 1,000

    44 more on a line of the same length: 22 numbers between 2 and 938, each twice. Every unit on this line is 10 times smaller.

Nothing tells you how close you were, and the line has no marks between its 2 ends, so every placement is an estimate. A countdown opens block 1, a short screen introduces block 2, and a bar at the top shows how far along you are.

Your click is converted using the line's measured width, so a narrow window does not squeeze your answers. Number line estimation has mostly been studied in children; in a population study of 6,484 ninth-graders, accuracy on the task went with scores on a standardised maths test even after other factors were taken into account (Nuraydin et al., 2023).

What it measures

Each click turns a number into a position. The gap between your position and the true one is what the task measures.

Block 1: 0 to 100

  • How precisely you map familiar 1- and 2-digit numbers onto space.
  • Mean placement error, as a percentage of the line.
  • Whether your placements rise in a straight line or bunch toward one end.

Block 2: 0 to 1,000

  • The same mapping on a scale where each step is finer.
  • Mean placement error for the wider range, the figure the dyscalculia check compares with a published adult value.
  • The same straight-line or bunched check, on the wider scale.

Errors can come from a vague sense of size or from difficulty translating numbers into space. Lafay and colleagues found that children with dyscalculia erred more when placing written numbers, which points to trouble reaching the number line from symbols (Lafay et al., 2017). Huber and colleagues found that adults with dyscalculia also learned new number-to-space layouts less well, and that reference points narrowed the gap (Huber et al., 2015).

A meta-analysis tied number line accuracy to maths competence across many studies (Schneider et al., 2018), and in a study of children most group differences showed up on the 0 to 1,000 line (Gut et al., 2022). This is one of the 4 tasks of the dyscalculia check, where the 0 to 1,000 error is set against a published adult value.

A test, not a diagnosis.

Placement error from a single run describes how that run went. It is not enough to decide whether anyone has a learning difficulty.

What you get

Both blocks are scored the moment the last number is placed.

On the result screen

  • A headline reading and a gauge of your average placement error over both blocks, in percent.
  • Your mean error for 0 to 100 and for 0 to 1,000.
  • A card per block with its error, its linear fit and a sentence on whether your placements looked evenly spaced or compressed.
  • A data panel with linear and logarithmic fits, the slope of your estimates and the better fit for each block, plus a CSV file of every click.

When you keep it

  • Kept in your results with a free account and available on any device after you sign in.
  • Placed among your number tasks, where the 0 to 1,000 error stands next to dot comparison and the rest.
  • One more test is suggested at the end, so your results build up.

How this version is scored

Every click stores the target, the range, the click position, the line width, your estimate and your time. The scores use the 88 main placements.

Estimate
Where you clicked, as a share of the line's width, times the range: halfway along the 0 to 1,000 line reads as 500.
Placement error
The distance between your estimate and the target, divided by the range, often called PAE. Placing 72 at 77 on the 0 to 100 line is a 5% error, and so is placing 722 at 772 on the 0 to 1,000 line.
Mean error
The average placement error over the 44 placements of a block. The headline gauge shows the mean of the 2 blocks.
Linear or compressed
A straight line and a logarithmic curve are fitted to your estimates. If the straight line explains them better by more than 0.02 of R², the block reads as linear; within 0.02, as mixed; otherwise as compressed.
Screen reading
A mean error under 5% gets the first reading, 5% to under 10% the middle one, 10% or more the third. The headline follows the weaker of the 2 blocks.

A published comparison. In 227 college-age adults, Williams and colleagues reported a mean 0 to 1,000 error of 3.9%, with a standard deviation of 1.1 (Williams et al., 2020). Their setup differed from this one and older adults were not included, so use it as context. The result screen itself does not give a percentile.

Why a curved pattern is read with care. A compressed, curved pattern used to be taken as a sign of a less linear mental number line. Huber and colleagues questioned that link and proposed a broader difficulty mapping numbers onto space (Huber et al., 2015).

Time is kept, not scored. Your time for each click is saved with the data but plays no part in the error or the fits.

Before you start

Pointer precision matters here, so a steady setup helps.

For a cleaner attempt

  • Use a mouse or trackpad on a computer, with the window wide enough to show the whole line.
  • Check both ends of the line before you click; the right end changes from 100 to 1,000 in block 2.
  • Take your time: nothing is timed and accuracy is the only score.
  • Keep the same pointer and the same posture for the whole run.

Interpret cautiously when

  • You used screen magnification, a screen reader or a motor-access tool, which change what the placement error measures.
  • Your pointer slipped or a click landed before you meant it to.
  • You were tired or rushed in block 2, the block the dyscalculia check compares with published figures.
  • You have done number line tasks before and know the scales well.

Questions

What does a number line estimation test measure?

It measures how accurately you translate a number into a position on a scale, which researchers treat as a measure of numerical magnitude understanding. The main score is the average distance between where you place each number and where it belongs, as a percentage of the line.

What is a good score on the number line task?

Lower error is better. In a sample of 227 college-age adults, mean error on a 0 to 1,000 line was 3.9%. This version's result screen gives its first reading under 5% and its third at 10% or more. Those limits are fixed in the task and are not adjusted for age.

Can I do the number line test on a phone?

Not on this page: precise placement needs a mouse or trackpad, so the tests hub asks for a computer. The dyscalculia check runs the same numbers with a larger touch area and includes a result taken on a phone with a caution. On a phone you can start with the Magnitude Comparison test.

Why does the 0 to 1,000 line feel harder?

The line keeps the same length but holds 10 times as many values, so each unit is smaller and the 2 ends are further apart in value. Studies of children have found more differences between groups on the 0 to 1,000 line than on 0 to 100.

Sources

By Cristian Dominguez Rein-Loring, Creative Technology Lead · Updated October 12, 2026