This statement accurately describes the fundamental relationship between reliability and validity in measurement theory, which is a core concept in software quality assurance metrics.

Reliability refers to the consistency and repeatability of a measure. A reliable measure will produce the same results under the same conditions.

Validity refers to the accuracy of a measure—whether it truly measures the attribute it is intended to measure.

The statement consists of two correct assertions:

1. "A measure can be reliable, but invalid." This is true. A measure can consistently produce the same result, but that result can be consistently wrong. For example, a test script that always checks for an outdated requirement will reliably fail every time, but it is invalid because it is not measuring the current quality of the product against the correct specification.

2. "An unreliable measure cannot be valid." This is also true. If a measure produces inconsistent, random, or unpredictable results, it cannot be considered accurate. For a measure to be valid (accurate), it must first be reliable (consistent). If the results are not repeatable, they cannot be trusted to represent the true value of the attribute being measured. Therefore, reliability is a necessary precondition for validity.

This is a "TRUE or FALSE" question. The option "False" is incorrect because it would imply that the relationship described is wrong. This would mean either that a reliable measure must always be valid (which is untrue) or that an unreliable measure could somehow be considered valid (which is a logical contradiction in measurement theory). Both of these implications are incorrect.

1. The Certified Software Quality Analyst Body of Knowledge (CSQA BOK), QAI Global Institute. The BOK outlines the principles of software metrics and measurement in Skill Category 4: "Quantitative Methods." It explains that for a metric to be useful, it must be both valid (measures what it is supposed to measure) and reliable (is consistent). The BOK supports the principle that reliability is a prerequisite for validity. A common analogy used is a target:

Reliable but Invalid: Shots are tightly clustered but are off the bullseye.

Unreliable and Invalid: Shots are scattered all over the target.

Reliable and Valid: Shots are tightly clustered in the bullseye.

This analogy clearly illustrates that a measure can be consistent without being accurate, but it cannot be accurate without first being consistent.

2. Fenton, N. E., & Bieman, J. M. (2014). Software Metrics: A Rigorous and Practical Approach (3rd ed.). CRC Press. In Chapter 2, "The Basics of Measurement," the authors discuss the principles of measurement theory as applied to software engineering. They state, "Reliability is a prerequisite for validity. An unreliable measure cannot be valid." (Fenton & Bieman, 2014, Section 2.3). This foundational text directly confirms the statement in the question.

3. University of Toronto, Department of Computer Science. (2018). CSC444H1F: Software Engineering I, Lecture 5: Software Measurement. Course materials from reputable computer science programs frequently cover this topic. Lecture notes explain that reliability (consistency) is a necessary but not sufficient condition for validity (accuracy). An unreliable measurement process, by definition, cannot be valid because its outputs are not stable enough to be considered an accurate representation of the attribute.