NIST Decision Tree for Key Comparisons

Version 1.0.4

About the NIST Decision Tree:

The NIST Decision Tree (NDT) is a web application that implements the Decision Tree for Key Comparisons (Possolo et al. 2021), which is intended for use as an aid for scientists who carry out interlaboratory studies aimed at generating Key Comparison Reference Values (KCRV). The NDT guides users through a series of hypothesis tests intended to help them in deciding upon an appropriate statistical model for their particular data. The NDT then carries out the preferred statistical procedure, and displays the results via plots, tables, and a downloadable report. Results include the KCRV estimate and associated uncertainties, an estiamte of dark uncertainty (where applicable), and the MRA Degrees of Equivalence and their uncertainties, among other results. For detailed instructions regarding the application, please click the button below to download the User Manual.

Download PDF Manual

Data Entry

Directions: Enter your data using the interactive table below. The table functions similarly to common spreadsheet software. For example, you can copy and paste data from a spreadsheet directly into the table below. To manually add/remove rows, right click on the table and select the desired option. Labs in rows with a the option checked will be taken into account for the KCRV computation. Labs in rows left unchecked will not be used in the KCRV computation, but their degrees of equivalence will be computed.

When you have finished entering the data into the table, click the 'Validate Data' button to confirm your selections. Then proceed to the 'Decision Tree' tab.

Alternatively, you may upload a .csv file following the same format as the above table. (Column names should read "Laboratory","MeasuredValues","StdUnc","DegreesOfFreedom".)




Unilateral Degrees of Equivalence Table

Table of Uncertainties for Each Lab