Data Analysis & Reporting Flashcards

A test conductor is building a predictive model to estimate remaining defects using reliability growth models. Which input data is essential?

Answer: Cumulative defects found over time or test cycles

Reliability growth models (e.g., Rayleigh, Gompertz) fit curves to cumulative defect discovery over time to project remaining latent defects.

When a test summary report identifies a variance between planned and actual test execution hours, what must the report include to be actionable?

Answer: Root cause analysis of the variance and its impact on schedule and quality

Identifying variance is only the first step; stakeholders need to understand why it occurred and what decisions it drives to take corrective action.

Which technique is used to segment test results data into groups based on input type, user class, or environment to identify patterns invisible in aggregate data?

Answer: Equivalence partitioning analysis

Equivalence partitioning groups data by shared characteristics, and analyzing results by partition can reveal that defects concentrate in specific input classes.

A test conductor's report shows 92% test pass rate, but post-release customer-reported defects are high. Which metric would have better predicted this outcome?

Answer: Defect detection percentage (DDP) benchmarked against post-release defects

DDP compares testing-phase finds to total defect population including post-release finds, directly measuring whether the test pass rate correlates with real quality.

In a test data traceability matrix, rows typically represent requirements and columns represent test cases. What does a cell marked 'X' indicate?

Answer: The test case in that column covers the requirement in that row

An 'X' in the traceability matrix maps a test case to the requirement it verifies, establishing bidirectional coverage evidence.

A test conductor wants to determine whether a new test automation framework reduced test execution time significantly. Which statistical test is most appropriate?

Answer: Paired t-test comparing execution times before and after framework adoption

A paired t-test compares the means of two related groups (same tests, two conditions) to determine if the difference is statistically significant.

When a test conductor reports 'exit criteria not met' at the end of a test phase, which information must accompany this finding?

Answer: Specific unmet criteria, associated risks, and recommended decision options for stakeholders

Unmet exit criteria require context—what is missing, what risk that poses, and what options exist—so stakeholders can make an informed go/no-go decision.