A measure is a tool that is used to quantify data and gain actionable insights. Its usefulness depends on its accuracy and relevance. Metrics must also be clear, simple, and understandable.
Metrics can help software engineering teams prioritize improvement efforts. They also provide a way for teams to communicate and align their work with customer expectations.
Measurement theory
Measurement theory is a diverse body of philosophical work that spans over several centuries and embraces a wide range of views on the metaphysics, epistemology, and semantics of measurement. Nevertheless, it can be divided into four broad strands: axiomatization, characterization of scale type, axiomatic foundations, and representational theories. Each strand has its own set of axioms, and the underlying assumptions about ontology and physics differ widely.
The first strand of measurement theory considers the adequacy of numbers for expressing magnitudes. Early measurement theorists like Helmholtz and Holder argued that qualitative empirical structures exhibited by magnitudes share structural features with algebraic operations among numbers. For example, the qualitative relation between two lengths of a rigid rod shares structural features with the numerical operation of addition.
Measurement units
A measurement unit is a standard definition of a quantity, such as length, volume or weight. These units are agreed upon by scientists and are regulated internationally. Examples include the metric system, the British imperial system and US customary systems.
For example, a meter is the same length everywhere because it is defined by scientists. The same is true of a kilogram, which measures the mass of an object. Similarly, a gallon is a standardized measure of volume because scientists decided to use the same unit for this quantity.
When referencing a measurement unit, it is important to spell it out in full. This is especially important for people who may use screen readers. When writing for a global audience, it is also important to use a non-breaking space between the number and its unit. This makes the unit easier to read and understand.
Measurement axioms
An axiom is an assertion that forms the basis of a mathematical theory. It may be a logical axiom or non-logical axiom. Logical axioms are symbolic statements that express relationships between other symbols, and non-logical axioms form substantive assertions about the elements of a domain, such as a + b in integer arithmetic or the Zermelo–Fraenkel axioms for set theory. Axioms are not to be confused with scientific postulates, which establish a scientific conceptual framework and must be verified experimentally. For example, Euclid’s four axioms of plane geometry were not proved experimentally until the 19th century. In mathematics, a set of axioms fixes a mathematical universe; in scientific theories, a set of postulates sets a conceptual framework and is verified through experiment.
Measurement systems
A measurement system is a collection of units that are linked to each other. This allows for a consistent and regulated way to measure physical quantities. Examples of measurement systems include the International System of Units (the modern form of the metric system), the British imperial system, and the United States customary system.
In an accurate measurement system, there is a linear relationship between the input x and the measured output y. A non-linear measurement system can be corrected by calibrating the instrument and establishing a new linear relationship. However, systematic errors may be unavoidable and may result from human error, parallax inaccuracies, or environmental changes.
Metrological traceability is the ability to relate a measurement result to a reference through a documented, unbroken chain of calibrations. This is essential for ensuring the quality of laboratory measurements.
Measurement practices
Measurement best practices involve measuring the right things, tracking and reporting results on a regular basis, and involving stakeholders in the measurement process. This improves transparency and accountability, as well as helps to uncover insights and identify areas for improvement.
Organizations should track both leading and lagging indicators. In addition, the indicators should be connected, so that performance in leading indicators leads to improved performance in lagging measures.
It’s important to get employees involved in measuring, tracking and analyzing performance. This improves buy-in and inspires faster action. It’s also helpful to collect comparative data on key measures. This helps organizations gauge their relative performance, identify goals and close performance gaps. Comparative data can be in the form of nominal, ordinal, interval, or ratio scales.