Software Engineering Metrics: What Do They Measure and How Do We Know?

Construct validity is about the question, how we know that we're measuring the attribute that we think we're measuring? This is discussed in formal, theoretical ways in the computing literature (in terms of the representational theory of measurement) but rarely in simpler ways that foster application by practitioners. Construct validity starts with a thorough analysis of the construct, the attribute we are attempting to measure. In the IEEE Standard 1061, direct measures need not be validated. "Direct" measurement of an attribute involves a metric that depends only on the value of the attribute, but few or no software engineering attributes or tasks are so simple that measures of them can be direct. Thus, all metrics should be validated. The paper continues with a framework for evaluating proposed metrics, and applies it to two uses of bug counts. Bug counts capture only a small part of the meaning of the attributes they are being used to measure. Multidimensional analyses of attributes appear promising as a means of capturing the quality of the attribute in question. Analysis fragments run throughout the paper, illustrating the breakdown of an attribute or task of interest into sub-attributes for grouped study.