Opening the book…
A linear calibration model describes a response that changes approximately in proportion to concentration, allowing for an intercept. A method may instead require an appropriate nonlinear model. Many immunoassays use sigmoidal calibration models, so a standard curve is not necessarily a straight line.
The reportable range is the interval of results over which the procedure provides suitable performance for its intended use. Relevant dilution or concentration steps can connect a sample’s original level to the instrumental working range. Validate or verify those steps as appropriate.
A protein assay can lose proportionality at higher levels. Do not use a straight-line equation in a saturated region. Depending on the approved method, dilute into a validated working range or use a suitable, validated nonlinear model. Do not choose a model merely because it gives a favorable value for an unknown.

This page titled 15.11: Calibration Model Linearity and Range was last modified on Mon, 09 Nov 2026 00:20:17 GMT and is shared under a CC BY-SA 4.0 license and was authored, remixed, and/or curated by Emalee MacKenzie.
The book could not be loaded just now.