Test: Sum of squares regression
Add one or several tests for the sum of squares regression to your assay.
Overview
To add the tests as assay suitability tests, add them as children of the Assay suitability tests element. To add the tests as sample suitability tests, add them as children of the Sample suitability tests element.
Important:
Test definitions are ignored without notification if you select a
type of test that is not valid for the model you use or the scope you select.
To pass the test, the sum of squares regression (from the ANOVA table) has to be less than or equal to the upper margin and greater than or equal to the lower margin set in the test definition.
Note:
We do not recommend this test if you have statistical or technical
outliers. The number of observations should be equal across all assays, that is, assays
for which you use this test and historical assays used in margin development.
Supported models
Linear parallel-line, 4-parameter logistic, 5-parameter logistic, 3-parameter logistic with fixed lower asymptote, 3-parameter logistic with fixed upper asymptote, Slope ratio
Scope of the test
- The Standard sample is always included in the calculation.
- 'Test samples only' or 'Control samples only' without simultaneous regression: The test is computed for each Test-vs-Standard or Control-vs-Standard comparison.
- 'All assay elements' without simultaneous regression: The test is computed for each Test-vs-Standard and for each Control-vs-Standard comparison.
Note:
With simultaneous regression (formerly called multiplex assay), all
samples are regarded as one single dataset and are tested together (not as individual
samples or pairwise comparisons).