Measurement Accuracy and Precision
Central Value of a Set
Definition of central value: The central value of a set should be determined to understand the data accurately.
Accuracy
Definition: Accuracy indicates the closeness of the measurement to the true or accepted value.
Expression of Accuracy: It is expressed by the difficulty to determine because the true value is usually unknown.
Precision
Definition: Precision refers to the agreement among several results obtained from repeated measurements. It indicates consistency across those measurements.
Example: Two measurements may have the same way of accuracy but varying precision.
Measurements: 4.13, 13.1, 12.9, 13.0, 13.0 show good precision but varying accuracy.
In contrast, measurements like 11, 10, 9, 9, 11, 1, 13 showcase good accuracy but poor precision.
Notation: Measurement 'A' has poor precision but good accuracy, whereas measurement 'B' has good precision but poor accuracy.
True value is considered to be 11.
Determinate Errors vs. Indeterminate Errors
Determinate Errors (Systematic Errors)
Characteristics:
Have a definite value.
Possess an assignable cause.
Exhibit the same magnitude for replicate measurements made under identical conditions.
Lead to bias in measurement results.
Example of Determinate Errors:
Data example: 13, 13.1, 12.9, 13.0, 13.0 indicates that errors are consistent in one direction.
Indeterminate Errors (Random Errors)
Characteristics:
Unpredictable nature.
Non-correctable through standard correction methods.
Can be reduced by averaging multiple measurements.
Can be treated mathematically using statistical methods.
Always present in measurements.
Question: Why can't indeterminate errors be eliminated?
Answer: Because they do not have an assignable cause.
Types of Determinate Errors
Categories:
Instrumental Errors: Errors resulting from the measuring instrument itself.
Method Errors: Errors arising from the method of measurement used.
Personal Errors: Errors that occur due to individual actions or subjective interpretation while measuring.
Figures of Merit (Performance Characteristics of Instruments)
Definition: A number derived from measurements used to evaluate an instrument or analytical technique.
Figures of Merit include:
Accuracy
Precision
Bias
Sensitivity
Detection Limit
Dynamic Range
Minimal Selectivity
Linearity
Measurement and Errors
The concept of absolute measurement:
No analytical instrument measurement is absolute, meaning that:
There are two types of errors: determinate and indeterminate (random).
Systematic errors have an assignable cause that can be traced and corrected, whereas random errors have causes that cannot be traced and corrected.
Error Definition: Refers to the difference between a measured value and the "true" or known value.
Estimated uncertainty: Error often denotes the estimated uncertainty associated with a measurement or experiment in implied form.
Realization: It is impossible to perform a measurement without error, emphasizing that errors are inevitable.
Improving Measurement Reliability
Recommendations for reliability and variability in measurements:
Measurements must be carried out in replicates (multiple instances of the same measurement under identical conditions).
Replicates should celebrate samples of the same kind and be processed through analysis in an exact manner.