Which of the following is used to determine the specificity of requirements?…
2016
Which of the following is used to determine the specificity of requirements?
Where \(n_1\) is the number of requirements for which all reviewers have identical interpretations, \(n_2\) is the number of requirements in a specification.
- A.
\(\frac{n_1}{n_2}\) - B.
\(\frac{n_2}{n_1}\) - C.
\(n_1 + n_2\) - D.
\(n_1 - n_2\)
Attempted by 272 students.
Show answer & explanation
Correct answer: A
Concept: Specificity (also called lack of ambiguity) is a software requirements specification quality metric. It measures the fraction of requirements that every reviewer interprets in the same way. Formally it is a normalised ratio - the count of unambiguous (unanimously interpreted) requirements divided by the total count of requirements - so its value always lies between 0 and 1.
Applying it here: let n1 be the requirements all reviewers interpret identically and n2 the total number of requirements. The specificity metric is therefore n1/n2.
n1 counts the requirements with a single shared interpretation (the unambiguous ones).
n2 counts every requirement in the specification (the whole basis for comparison).
Dividing n1 by n2 expresses the unambiguous requirements as a proportion of the whole, giving a score in [0, 1].
Precondition: n2 must be greater than 0; if n2 = 0 the ratio is undefined.
Cross-check by interpretation: a value of 0 means no requirement is interpreted unanimously (maximum ambiguity), while 1 means every requirement is interpreted identically by all reviewers (no ambiguity). The score rises as the specification becomes clearer, which is the behaviour a specificity measure must have.
Example: if n1 = 18 and n2 = 20, specificity = 18/20 = 0.9, indicating a highly specific (low-ambiguity) specification.