Definitions Flashcards
(30 cards)
Estimates of true values (often derived from the mean of results from a number of laboratories) (target value, consensus value, conventional true values)
Assigned values
Deviation of a test result from the assigned value; reproducible inaccuracies that are consistently in the same direction (systematic error)
Bias
Theoretical distribution used to model events falling into two categories; e.g. motile/nonmotile, viable/nonviable
Binomial distribution
Plot of the difference between a series of paired observations against their mean (x-axis: average; y-axis: difference)
Bland-Altman Plot
Source of natural variation that affects all individual values of the process being studied
Common cause variation
Interval calculated from observed data that includes the true value in 95% of replicates (x-Bar +/- 1.96 x Standard Error or N +/- 1.96 x square root N for counts)
95% confidence interval
A time sequence chart showing a series of individual measurements, together with a central line and control limits
Control chart
Maximum allowable variation of a process due to common causes alone; variation beyond a control limit is evidence that special causes may be affecting the process
Control limits
A plot showing cumulative or added deviation from a target result; a tool for early warning of bias and drift
Cusum plot
Successive small changes in values leading to a change in accuracy with time (bias)
Drift
Comparisons between results from procedures performed locally by different laboratories, undertaken by an external body; useful for detecting, systematic variation, and assessing accuracy
External quality assessment (EQA)
A set of principles that provide a framework within which laboratory studies are planned, performed, monitored, recorded, reported and archived
Good laboratory practice
A process is in control when all values are within expected control limits
In control
Quality test measuring the variability in a procedure that exists within a laboratory; such test evaluate the precision of day-to-day operations; useful for detecting, random variation (assessing precision)
Internal quality control (IQC)
International organization for standardization, a body that sets international standards, including for laboratory quality
ISO
Commercially available samples, manufactured and analyzed according to manufacturing guidelines
Manufactured QC samples
A process is out of control when a measured value exceeds expected control limits or is within control limits, but shows a significant trend in values; a process that is out of control must be evaluated
Out of control
Plan, Do, Check, Act
PDCA or Shewhart Cycle
A theoretical distribution used to model counts
Poisson distribution
Closeness of agreement between replicate measurements; commonly expressed as in precision (drift, within, between, inter-run, batch, assay, or laboratory variation); measurements of precision are not affected by bias. See also sampling error.
Precision
A form of EQA, where an external body compares results from several laboratories and issue certificates when set levels of performance are met
Proficiency test
Also called sampling error or precision error
Random error
A control chart of standard deviations of measured values against time, used to monitor process, uniformity and measurement precision
S chart
The error involved in counting a limited number of specimen (inversely proportional to the square root of the number counted); or sampling error is the standard error of a count expressed as percentage of the count
Sampling error
