Accuracy experiment


Practical implementation of the definitions of trueness and precision in accuracy experiments

ISO 5725-1 (1994)

Accuracy (trueness and precision) of a measurement or test method is to be determined with a series of test results reported by a number of participating laboratories organised by a project management.

Samples of a certain material or product are sent from a central point to a number of laboratories at different places.

  • the samples shall be identical when sent out to the laboratories
  • the samples shall remain identical during transport and between measurements

Statistical model of accuracy experiment

When estimating accuracy, it is useful to assume that each measurement result, y, is the sum of three components:

y = m + b + e

where, for a certain test object:

m is the accepted reference (or “true”) value

b is the bias

e is the random error in each measurement


Bias

The bias, b, of a laboratory

is estimated as

labbias

while

The bias, B, in the measurement method

biasmetod

where u ref the accepted reference (or “true”) value.


Variance

The variance in the measurement result, Var(y), is thus the sum of the variances in the bias and the error.

The variance in the bias between laboratories

Var(b) = sL2

The variance in the error for each laboratory:

Var(e) = sW2

can vary from lab to lab.

Repeatability variance

repetervarians

is the average of the own variances of the different laboratories.

Reproducibility variance

reprodvarians

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: