All demonstrationsGRIM + GRIMMER / INTEGER ARITHMETIC

Can that rounded mean exist?

Ask whether a reported mean or standard deviation passes necessary arithmetic checks for bounded integer observations.

Not always. With 10 integer responses, every possible mean is a multiple of 0.1, so a reported 3.48 cannot occur, while 3.50 can. This is the GRIM test (Granularity-Related Inconsistency of Means). An inconsistency is a reason to check the data, not proof of fraud. Read the guide

audit_statisticsTry the real call
Concept artwork · not calculation evidence
01 / UNDERSTAND THE METHOD

See what the tool does.

ILLUSTRATION · NOT A CALCULATION
INTERACTIVE 3D / METHOD STUDYExplore the method in space.
n × reported meanCan rounding reach an integer sum?3233343536

Preparing the 3D exhibit…

A teaching illustration. Run the tool below to obtain a native calculation.Drag to inspect · arrow keys move the camera · no MCP call

02 / ASK THE ACTUAL TOOL

Make it your experiment.

Complete tool arguments

Arrays, coordinates and nested contracts remain editable here. Units and limits are checked by the real tool.

Runs only when you press the button. Input edits discard the previous displayed result. Public compute limits apply.

KNOW WHAT THE RESULT MEANS

Useful evidence needs boundaries.

  • These checks apply only when integer observations and the specified rounding rules describe the data.
  • Passing necessary conditions does not reconstruct a dataset or prove one exists.
  • A flagged inconsistency can reflect transcription or methodological differences; it is not an accusation of fraud.
Choose another demonstrationCall the same tool from your AI