I85PREP
⌘ K
CSQE study materials
CSQE study resources

CSQE practice questions

One question at a time. Build your CSQE understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer is choosing a model that describes how defects are introduced and detected across requirements, design, coding, and testing phases. Which model best supports this analysis?

  1. The Rayleigh defect-distribution model
  2. The Fagan inspection process model
  3. The Rayleigh model of reliability growth
  4. The V-model of development phases Correct answer.

Explanation

The V-model maps each development phase to a corresponding verification phase, so defects can be traced to the phase where they were introduced and the phase that should detect them. It is the standard framework for phase-containment analysis. Reliability growth models address failure rates over test time rather than phase-level defect introduction.

Question 2Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which metric directly measures the effectiveness of a code review process?

  1. Lines of code reviewed per reviewer-hour
  2. Defect detection rate expressed as defects found per thousand lines reviewed Correct answer.
  3. Number of review meetings held per release
  4. Percentage of code covered by automated unit tests

Explanation

Review effectiveness is measured by how many defects the review actually finds relative to the volume of material examined, typically defects per thousand lines or per inspection hour. This captures yield rather than effort. Coverage and meeting counts measure activity, not detection performance.

Question 3Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A team plots cumulative open defects against cumulative closed defects over a release cycle. The two curves remain widely separated and the gap widens near the release date. What does this pattern most directly indicate?

  1. The defect arrival rate has stabilized at a low level
  2. The inspection process is finding more defects than testing
  3. The defect removal rate is lagging behind the defect arrival rate Correct answer.
  4. The team has achieved a stable, predictable release readiness

Explanation

When the cumulative-open curve stays well above the cumulative-closed curve and the gap grows, defects are arriving faster than they are being removed. This signals a removal-capacity problem and elevated release risk. A converging pair of curves would instead indicate stabilizing readiness.

Question 4Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

In a software quality cost framework, which activity is classified as an appraisal cost?

  1. Training developers in secure coding practices
  2. Regression testing performed before release Correct answer.
  3. Rework performed after a customer-reported failure
  4. Warranty support provided after delivery

Explanation

Appraisal costs are incurred to evaluate a product and detect defects before delivery, and regression testing is a classic example. Prevention costs such as training aim to stop defects from occurring, while internal and external failure costs cover rework and post-delivery support.

Question 5Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A project needs to decide whether to release software whose measured defect density is 2.5 defects per KLOC against a historical release average of 1.8. The quality goal is to release only when density is at or below the historical average. What is the most defensible quality decision?

  1. Release, because 2.5 is within typical industry ranges
  2. Hold the release and investigate the elevated density before deciding Correct answer.
  3. Release, provided the schedule milestone has been met
  4. Release, because defect density is not a release criterion

Explanation

The stated goal is a density at or below the historical average, and 2.5 exceeds 1.8, so the criterion is not met. The disciplined response is to hold and investigate the causes rather than override the goal. Schedule pressure or generic industry ranges do not supersede an explicit quality criterion.

Question 6Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which pairing correctly matches a measurement scale with an operation it legitimately permits?

  1. Interval scale - computing a ratio between two values
  2. Ratio scale - computing a coefficient of variation Correct answer.
  3. Ordinal scale - computing the arithmetic mean of the ranks
  4. Nominal scale - ranking categories by magnitude

Explanation

Ratio scales have a true zero, which permits meaningful ratios and statistics such as the coefficient of variation. Interval scales support differences but not ratios, ordinal scales support order but not arithmetic means, and nominal scales support only classification.

Question 7Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A software quality engineer wants to determine whether a new static analysis tool reduces escaped defects. Which study design provides the strongest evidence?

  1. A vendor benchmark report listing tool features
  2. A survey of developers asking whether they trust the tool
  3. A single team's before-and-after comparison over one sprint
  4. A controlled comparison of similar projects with and without the tool Correct answer.

Explanation

A controlled comparison across similar projects isolates the tool's effect from confounding factors such as team skill and schedule. Before-and-after designs on one team are vulnerable to history and maturation effects. Surveys and vendor reports provide opinion or marketing data rather than outcome evidence.

Question 8Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which statement best describes the purpose of a Pareto analysis in software quality improvement?

  1. It predicts the exact number of defects remaining in the product
  2. It replaces root cause analysis with a ranking of symptoms
  3. It identifies the small number of causes responsible for most defects Correct answer.
  4. It proves that the process is statistically stable over time

Explanation

Pareto analysis ranks defect categories by frequency to reveal the few causes that account for the majority of problems, focusing improvement effort. It is descriptive and prioritizes attention but does not estimate remaining defects or establish statistical control. Control charts, not Pareto charts, demonstrate process stability.

Question 9Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A control chart of weekly defect counts shows one point beyond the upper control limit but no other patterns. What is the appropriate interpretation?

  1. The data are invalid and the chart should be discarded
  2. The process is in control because only one point is out of limits
  3. The process shows a special cause that warrants investigation Correct answer.
  4. The control limits should be widened to include the point

Explanation

A point beyond a control limit is a standard signal of a special cause of variation, and the process should be investigated rather than adjusted by changing the limits. Control limits are derived from the process data, not chosen to fit observations. A single out-of-limit point is a meaningful signal, not a reason to discard the chart.

Question 10Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which activity is the primary purpose of a software configuration audit?

  1. Verifying that the product matches its approved configuration documentation Correct answer.
  2. Selecting which defects will be fixed in the current iteration
  3. Estimating the effort required for the next release
  4. Measuring the cyclomatic complexity of the delivered code

Explanation

A configuration audit verifies that what was built and delivered conforms to the approved baseline documentation, including requirements, design, and change records. It is a conformance check on the configuration items, not a code metric, estimation, or triage activity.

Question 11Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

During a formal inspection, a reviewer records a defect that the author believes is not a defect. What is the most appropriate way to handle this disagreement within the inspection process?

  1. The recorder logs only defects that the author accepts during the meeting
  2. The reviewer and author debate the point until consensus is reached in the meeting
  3. The author immediately deletes the record to keep the inspection on schedule
  4. The moderator records it as a defect and the team resolves the dispute after the meeting Correct answer.

Explanation

Inspection practice separates defect logging from defect resolution so that discussion of validity does not consume the meeting or suppress findings. The moderator captures the item and the dispute is resolved afterward. Deleting or filtering records during the meeting distorts the defect data and undermines the inspection's purpose.

Question 12Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which pair of metrics is most useful for monitoring whether a test process is converging toward release readiness?

  1. Defect arrival rate and defect removal rate over time Correct answer.
  2. Test cases written and testers assigned
  3. Build frequency and repository commit count
  4. Requirements count and design document page count

Explanation

Comparing defect arrival and removal rates over time shows whether detection and correction are converging, which is the core evidence of test progress and readiness. Volume metrics such as test cases written or commits made measure activity rather than convergence.

Question 13Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A reliability model is being selected for a product whose failure rate is expected to rise early, peak, and then decline during the test period. Which model family is most consistent with this assumption?

  1. A Rayleigh or Weibull-shaped failure intensity model Correct answer.
  2. A geometric defect-removal model
  3. A constant-failure-rate exponential model
  4. A linear regression of defect counts on calendar date

Explanation

The Rayleigh and Weibull families can represent a failure intensity that rises to a peak and then declines, matching the assumed lifecycle. The exponential model assumes a constant failure rate, which contradicts the stated rise-and-fall pattern. Geometric removal models describe decreasing defect counts per fix, not the shape of failure intensity over test time.

Question 14Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which practice best supports traceability from a requirement to the test that verifies it?

  1. Assigning unique identifiers to requirements and recording them in test cases Correct answer.
  2. Recording traceability only after the product has been released
  3. Relying on developers to remember which tests cover each requirement
  4. Storing test results in a shared spreadsheet without identifiers

Explanation

Traceability requires stable identifiers that link requirements to design elements, code, and test cases. Unique requirement identifiers recorded in test cases make the links explicit and auditable. Informal memory or post-release reconstruction cannot provide reliable bidirectional traceability.

Question 15Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A team wants to estimate how many defects remain undetected before release. They have seeded a known number of artificial defects into the product and measured how many were found in testing. Which technique are they applying?

  1. Root cause analysis with Ishikawa diagrams
  2. Orthogonal defect classification
  3. Statistical process control with p-charts
  4. Fault seeding with capture-recapture estimation Correct answer.

Explanation

Fault seeding injects known defects and uses the proportion detected to estimate the total population, often via capture-recapture logic. Orthogonal defect classification tags defects by type and trigger, root cause analysis explains causes, and p-charts monitor proportions over time; none of these estimates the residual defect population.

Question 16Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which characteristic distinguishes a formal software inspection from a walkthrough?

  1. A walkthrough uses defined roles, entry criteria, and recorded defect data
  2. An inspection is performed only by the author of the code
  3. An inspection uses defined roles, entry and exit criteria, and recorded findings Correct answer.
  4. A walkthrough always produces a signed quality certificate

Explanation

Formal inspections are characterized by defined roles such as moderator and recorder, explicit entry and exit criteria, and documented defect data. Walkthroughs are less formal, author-led sessions without those controls. Neither activity is performed solely by the author in the inspection case, and neither issues certificates.

Question 17Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer computes a process capability index for a software metric and obtains a value well below 1.0. What does this most likely indicate?

  1. The specification limits are too wide to be meaningful
  2. The process is producing output that frequently exceeds specification limits Correct answer.
  3. The metric is not being measured correctly
  4. The process variation is small relative to specification limits

Explanation

A capability index below 1.0 means the spread of the process relative to the specification width is large enough that nonconforming output is expected. Higher index values indicate a more capable process. The index itself does not diagnose measurement error or specification width without further analysis.

Question 18Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which risk-based testing strategy allocates the most test effort appropriately?

  1. Testing the highest-risk features most intensively Correct answer.
  2. Testing only the features that are easiest to automate
  3. Testing only the features requested by marketing
  4. Testing all features equally regardless of likelihood of failure

Explanation

Risk-based testing concentrates effort where the combination of failure likelihood and impact is greatest, which maximizes the value of limited test resources. Uniform allocation ignores risk differences, and selecting by marketing request or automation ease is not a risk criterion.

Question 19Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

An organization wants to benchmark its defect density against other organizations. Which condition is most important for a valid comparison?

  1. Using the same defect counting rules and size measure definitions Correct answer.
  2. Restricting the comparison to projects with identical team sizes
  3. Comparing only projects completed in the same calendar year
  4. Ensuring all projects use the same programming language

Explanation

Valid benchmarking requires that defect definitions and size measures be equivalent, because differences in counting rules or normalization produce misleading comparisons. Calendar year, language, and team size may affect context but are not prerequisites for definitional comparability.

Question 20Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which statement correctly describes the relationship between verification and validation?

  1. Validation is performed only after verification has fully succeeded
  2. Verification confirms the product meets user needs; validation confirms it meets specifications
  3. Verification confirms conformance to specifications; validation confirms fitness for intended use Correct answer.
  4. Verification and validation are synonyms in software quality practice

Explanation

Verification asks whether the product was built according to its specifications, while validation asks whether it satisfies the intended use and user needs. They are complementary and are often performed concurrently rather than strictly sequentially. Treating them as synonyms collapses an important distinction.

Question 21Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A team finds that 70 percent of its escaped defects trace to requirements ambiguity. Which corrective action addresses the dominant root cause?

  1. Introducing structured requirements review with defined acceptance criteria Correct answer.
  2. Adding more developers to the coding phase
  3. Increasing the number of regression test cycles
  4. Purchasing a faster build server

Explanation

If most escapes originate in ambiguous requirements, the corrective action must target requirements quality through structured review and explicit acceptance criteria. More testing or staffing addresses downstream symptoms, and infrastructure changes are unrelated to the root cause.

Question 22Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which set of elements is essential to a functioning configuration management system?

  1. Inspection, walkthrough, review, and audit
  2. Estimation, scheduling, staffing, and budgeting
  3. Identification, control, status accounting, and audit Correct answer.
  4. Design, coding, testing, and deployment

Explanation

Configuration management comprises identification of configuration items, control of changes, status accounting of those items, and configuration audits. The other lists describe project management, lifecycle phases, or review types rather than configuration management functions.

Question 23Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A control chart for defect counts shows seven consecutive points all above the center line. No point exceeds a control limit. How should this be interpreted?

  1. The center line was computed incorrectly and must be recalculated
  2. The chart should be replaced with a Pareto diagram
  3. The run signals a shift in the process mean that should be investigated Correct answer.
  4. The process is in control because no point exceeds the limits

Explanation

A run of seven or more consecutive points on one side of the center line is a recognized signal of a sustained shift in the process mean, even without points beyond the limits. Control charts use both limit violations and pattern rules. Recomputing the center line or switching chart types would conceal the signal.

Question 24Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

Which metric best quantifies the containment of defects to the phase in which they were introduced?

  1. Phase containment effectiveness Correct answer.
  2. Test case execution rate
  3. Defect density per module
  4. Mean time to failure

Explanation

Phase containment effectiveness measures the proportion of defects detected in the phase where they were introduced versus those that escaped to later phases. Defect density measures concentration, mean time to failure measures reliability, and execution rate measures test throughput.

Question 25Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A project's defect data show that defects found in system testing are increasingly traced to design-phase origins. Which action is most likely to improve quality outcomes?

  1. Adding more system testers to find the same design defects faster
  2. Strengthening design reviews and design verification criteria Correct answer.
  3. Shortening the system test schedule to reduce cost
  4. Deferring design documentation until after coding

Explanation

When defects originate in design and escape to system testing, the leverage point is improving design-phase verification so those defects are caught earlier. Adding downstream testers treats the symptom at higher cost, and shortening testing or deferring design documentation would worsen the problem.

Question 26Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A software quality engineer is asked to distinguish between quality control and quality assurance activities within a development organization. Which activity is properly classified as quality assurance rather than quality control?

  1. Defining and auditing the process used to review requirements specifications Correct answer.
  2. Inspecting a module for coding standard violations before check-in
  3. Executing the regression test suite against a release candidate
  4. Recording and triaging customer-reported defects after deployment

Explanation

Quality assurance is process-oriented: it establishes and audits the processes by which work is done so that defects are prevented. Quality control is product-oriented and evaluates the actual work products or delivered software against requirements. Auditing how requirements reviews are conducted is therefore assurance, while testing, inspection, and defect triage are control.

Question 27Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

An organization is establishing a software quality program and wants a recognized framework for defining process areas, maturity levels, and appraisals. Which framework is designed specifically for that purpose?

  1. IEEE 829 test documentation standard
  2. COCOMO II cost estimation model
  3. CMMI process maturity model Correct answer.
  4. ISO/IEC 25010 product quality model

Explanation

CMMI defines process areas organized into maturity levels and provides appraisal methods for assessing an organization's process capability. The other references serve different purposes: product quality characteristics, test documentation structure, and effort estimation. Confusing these frameworks is a common source of program design errors.

Question 28Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer is writing a requirement for a medical device user interface and wants it to be verifiable. Which phrasing best satisfies that attribute?

  1. The interface shall respond to any user action within 2 seconds on the reference hardware. Correct answer.
  2. The interface shall be user-friendly and responsive at all times.
  3. The interface shall be designed in accordance with good usability practice.
  4. The interface shall be intuitive for clinical users.

Explanation

A verifiable requirement states a measurable condition that a test can confirm or refute, such as a response time bound on defined hardware. Terms like intuitive, user-friendly, and good practice cannot be objectively evaluated, so they fail the verifiability attribute even when they sound professional.

Question 29Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A project maintains a requirements traceability matrix. Which item should appear as a column in that matrix to support impact analysis when a requirement changes?

  1. The date each requirement was first proposed
  2. The estimated cost of each requirement in staff hours
  3. The names of the developers who wrote each requirement
  4. The design elements, code units, and test cases linked to each requirement Correct answer.

Explanation

Traceability matrices link requirements to downstream artifacts such as design elements, code units, and test cases, which is what enables impact analysis when a requirement changes. Author names, cost estimates, and proposal dates are project metadata that do not reveal which artifacts must be revisited.

Question 30Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A software quality engineer is selecting a review technique for a small, well-defined module and wants reviewers to prepare individually in advance, then meet with a moderator and recorder to log issues systematically. Which technique matches this description?

  1. Formal inspection Correct answer.
  2. Desk checking by the author alone
  3. Pair programming
  4. Ad hoc review

Explanation

Formal inspection is characterized by defined roles such as moderator and recorder, individual preparation before a structured meeting, and systematic logging of defects. Ad hoc review lacks preparation and defined roles, desk checking is a solitary author activity, and pair programming is a continuous construction practice rather than a scheduled review.

Question 31Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer is planning a sampling approach for auditing a large set of configuration items. Which condition is most important for the sample to support valid inferences about the whole population?

  1. The sample consists of the items the auditor finds most interesting
  2. The sample size equals at least half the population
  3. The sample is drawn randomly or systematically from the full population Correct answer.
  4. The sample is chosen by the item owners themselves

Explanation

Statistical inference requires that every item have a known chance of selection, which random or systematic sampling provides. Judgmental or self-selected samples introduce bias and cannot support conclusions about the population regardless of size. A large but biased sample remains unrepresentative.

Question 32Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

During a quality function deployment exercise, which artifact is used to translate customer requirements into design characteristics and to compare them against competitor offerings?

  1. A Pareto chart of defect categories
  2. The house of quality matrix Correct answer.
  3. A fishbone diagram of a single defect
  4. A control chart of process variation

Explanation

The house of quality is the central QFD matrix that maps customer needs to engineering characteristics, records relationships and correlations, and benchmarks competitors. Control charts, Pareto charts, and fishbone diagrams are quality tools used later for process monitoring and root cause analysis, not for translating customer needs into design.

Question 33Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer categorizes defects by severity and needs a definition that separates severity from priority. Which statement correctly captures the distinction?

  1. Severity is the degree of impact a defect has on the product, and priority is the order in which it is addressed. Correct answer.
  2. Severity applies only to safety defects, while priority applies only to cosmetic defects.
  3. Severity and priority are synonyms and are recorded in a single field.
  4. Severity is the order in which defects are fixed, and priority is the degree of harm to the user.

Explanation

Severity describes the technical or business impact of a defect on the product, while priority reflects scheduling decisions about when it will be fixed. A high-severity defect can have low priority if it occurs in an unused feature, and a low-severity defect can be urgent for a key customer. Keeping the two attributes distinct preserves both technical and business information.

Question 34Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A team plots the number of open defects found per week and observes that the count has dropped sharply while the number of defects still being reported by customers has remained constant. Which interpretation is most defensible?

  1. The product quality has improved because fewer defects are being found internally.
  2. Internal detection effectiveness may have declined, so the drop should be investigated before concluding quality improved. Correct answer.
  3. The defect tracking tool is underreporting and should be replaced.
  4. The customer reports are duplicates and can be disregarded.

Explanation

A falling internal defect discovery rate alongside a stable escaped-defect rate suggests the organization is finding fewer of the defects that exist rather than producing fewer of them. Detection effectiveness, test coverage, and reporting behavior should be examined before treating the trend as improvement. Assuming tool failure or dismissing customer data skips that analysis.

Question 35Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A software quality engineer is constructing a checklist for peer reviews of design documents. Which property makes a checklist item most effective?

  1. It describes the history of the design standard
  2. It states a specific, observable condition that a reviewer can confirm or refute Correct answer.
  3. It reminds reviewers to do a thorough job
  4. It lists the names of the design authors

Explanation

Effective review checklist items are specific and observable so that different reviewers reach consistent conclusions about conformance. General exhortations, author lists, and historical background do not direct attention to a checkable property of the document and therefore produce inconsistent reviews.

Question 36Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

An organization wants to reduce the cost of poor quality. Which cost category represents rework performed on defective software discovered before release?

  1. Prevention cost
  2. Appraisal cost
  3. External failure cost
  4. Internal failure cost Correct answer.

Explanation

Internal failure costs arise from defects detected before the product reaches the customer, such as rework and retesting during development. External failure costs occur after delivery, appraisal costs are incurred to detect defects, and prevention costs are incurred to stop defects from being introduced. Correct classification directs investment toward prevention and appraisal.

Question 37Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer wants to determine whether two categorical variables, development team and defect type, are associated. Which technique is appropriate?

  1. Linear regression of defect count on calendar date
  2. A paired t-test on defect counts
  3. A chi-square test of independence on the contingency table Correct answer.
  4. A moving range control chart

Explanation

A chi-square test of independence evaluates whether the distribution of one categorical variable differs across levels of another using a contingency table. A t-test requires a continuous response and defined pairs, control charts address process stability over time, and regression requires a continuous predictor and response relationship.

Question 38Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A project uses an incremental delivery model and performs a quality gate at the end of each increment. Which criterion best defines a meaningful gate?

  1. The project manager signs a status report.
  2. The increment has been demonstrated to at least one stakeholder.
  3. Predefined exit criteria such as test completion, defect thresholds, and traceability coverage are met. Correct answer.
  4. All planned work for the increment has been started.

Explanation

A quality gate is meaningful only when it has predefined, objective exit criteria that can be evaluated as met or not met, such as required test completion, severity-based defect thresholds, and traceability coverage. Starting work, signing a status report, or a single demonstration does not establish that quality conditions were satisfied.

Question 39Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A software quality engineer must choose between inspections and testing for a set of requirements. Which consideration most strongly favors inspection over dynamic testing?

  1. The requirement concerns the throughput of a database under production data volumes
  2. The requirement concerns the behavior of third-party binary components
  3. The requirement concerns timing behavior under heavy concurrent load
  4. The requirement concerns adherence to a coding standard and consistent use of an interface convention Correct answer.

Explanation

Inspection examines the artifact directly and is well suited to properties such as standards conformance, interface consistency, and maintainability that are hard to observe through execution. Timing, load, and throughput behavior emerge only when the system runs, and third-party binaries may not be inspectable, so those cases favor testing.

Question 40Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer is reviewing a test plan and finds that entry criteria are not defined. What risk does this omission create?

  1. Test cases will be executed in the wrong order.
  2. The test plan will fail to specify the project budget.
  3. The defect tracking tool will not accept new entries.
  4. Testing may begin before prerequisites such as stable builds and approved requirements exist, wasting effort on invalid results. Correct answer.

Explanation

Entry criteria define the conditions that must hold before testing starts, such as a stable build, approved requirements, and available environments. Without them, testing can begin prematurely and produce results that must be discarded. Budget, execution order, and tool configuration are governed by other planning elements.

Question 41Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

An organization wants to reduce the number of defects that escape to customers. Which measure of review effectiveness should it track?

  1. The number of reviewers invited to each review
  2. The number of review meetings held per month
  3. The average duration of each review meeting
  4. The percentage of total defects that are detected in reviews before later phases Correct answer.

Explanation

Detection effectiveness, the proportion of defects caught in reviews relative to all defects found, directly measures how well reviews remove defects before they escape. Meeting counts, durations, and attendance measure effort and activity but not whether the reviews actually found the defects that mattered.

Question 42Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer observes that a metric used to evaluate individual developer productivity has caused developers to avoid complex assignments. What is the most appropriate response?

  1. Replace or supplement the individual metric with team-level measures that reflect delivered quality Correct answer.
  2. Increase the metric's weight in performance reviews
  3. Publish individual rankings more widely to increase motivation
  4. Keep the metric but add a penalty for refusing assignments

Explanation

When a metric drives dysfunctional behavior, the measurement system itself is the problem; people optimize what is measured. Shifting to team-level measures that include quality outcomes reduces the incentive to game the number. Penalties or greater publicity intensify the same distortion rather than correcting it.

Question 43Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A software quality engineer is asked to verify that a delivered system meets its specified performance requirements under expected workloads. Which activity does this describe?

  1. A process appraisal, because it assesses the development process
  2. A configuration audit, because it checks delivered items
  3. Verification, because it confirms the system conforms to specified requirements Correct answer.
  4. Validation, because it confirms the system satisfies user needs

Explanation

Confirming conformance to specified requirements, such as stated performance targets under defined workloads, is verification. Validation instead asks whether the system satisfies the intended use in its operational context. Configuration audits check item identification and completeness, and process appraisals assess organizational processes.

Question 44Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer wants to apply statistical process control to a metric that is measured as the proportion of units failing inspection in samples of varying size. Which chart type is appropriate?

  1. An X-bar and R chart
  2. A p chart Correct answer.
  3. An individual and moving range chart
  4. A c chart

Explanation

A p chart monitors the proportion of nonconforming units and accommodates varying subgroup sizes through per-subgroup control limits. Individual and X-bar charts require continuous measurements, and a c chart counts defects per unit of constant opportunity rather than the fraction failing.

Question 45Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A project's quality plan states that all code must be reviewed before integration. A quality engineer finds that 30 percent of integrated modules had no recorded review. Which action best addresses the finding?

  1. Add more reviewers to every review team
  2. Revise the quality plan to remove the review requirement
  3. Analyze why the requirement was bypassed and strengthen enforcement and evidence capture at the integration gate Correct answer.
  4. Retrain the entire engineering staff on the review process

Explanation

A gap between a stated requirement and observed practice calls for root cause analysis of why the control was bypassed, followed by enforcement and evidence capture at the point where integration occurs. Weakening the requirement hides the risk, and blanket retraining or adding reviewers assumes a cause that has not been established.

Question 46Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer compares two test teams and finds that team A reports many more defects than team B on comparable products, while escaped-defect rates are similar. Which explanation is most plausible before further investigation?

  1. Team A has higher detection effectiveness and reports defects that team B misses Correct answer.
  2. Team A produces lower-quality code than team B
  3. Team B's products are inherently simpler
  4. Team A's defect tracking tool duplicates entries

Explanation

Higher reported defect counts with similar escape rates are consistent with greater detection effectiveness: team A finds and records defects that team B misses, so fewer remain hidden. Attributing the difference to code quality, product simplicity, or tool duplication assumes causes not supported by the data. The comparison should be confirmed by examining defects found per unit of test effort and review coverage.

Question 47Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer must decide whether a process change actually reduced mean defect density. Before-and-after samples exist, but the process variance is unknown and sample sizes are small. Which approach is most appropriate?

  1. Compare the two sample means visually and declare a reduction if the after mean is lower
  2. Assume the variances are equal and apply a two-sample z-test
  3. Apply a two-sample t-test that does not assume equal variances, after checking distributional assumptions Correct answer.
  4. Apply a chi-square test to the defect counts

Explanation

With unknown variances and small samples, a two-sample t-test is appropriate, and the version that does not assume equal variances avoids an unsupported equality assumption. A z-test requires known variances or large samples, visual comparison of means ignores sampling variability, and a chi-square test applies to categorical association rather than comparing two means.

Question 48Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A software quality engineer is designing measures for a safety-related system and must distinguish between reliability and availability. Which pair of definitions is correct?

  1. Reliability is the mean time to repair; availability is the mean time between failures.
  2. Reliability is the probability of failure-free operation for a stated time; availability is the proportion of time the system is operable. Correct answer.
  3. Reliability is the proportion of time the system is operable; availability is the probability of failure-free operation for a stated time.
  4. Reliability and availability both measure only the time between failures.

Explanation

Reliability is a probability statement about failure-free operation over a specified interval under stated conditions, while availability is the fraction of time the system is capable of performing its function. Availability depends on both failure frequency and repair duration, so a highly available system can still be unreliable if it recovers quickly.

Question 49Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer finds that a control chart of a software metric shows a stable process, but customer satisfaction scores have declined steadily. Which conclusion is most sound?

  1. The control chart must be miscalculated because satisfaction declined
  2. Customer satisfaction should be removed from the quality dashboard
  3. Process stability in the metric does not establish that the metric captures what customers value, so the measurement system should be revisited Correct answer.
  4. The metric is adequate because the process is in control, so satisfaction decline is unrelated

Explanation

Statistical control indicates only that a metric's variation is consistent over time; it says nothing about whether the metric measures the right thing. A stable metric with declining customer satisfaction signals a validity gap in the measurement system, prompting review of what is measured and whether additional customer-facing measures are needed.

Question 50Software Quality Engineering (Body of Knowledge content not enumerated on the source page)

A quality engineer is planning to apply orthogonal defect classification to improve analysis of defect data. Which information does this classification primarily add beyond defect counts?

  1. The cost of each defect in labor hours
  2. Defect type, trigger, and impact attributes that support causal analysis Correct answer.
  3. The build number in which each defect was found
  4. The identity of the tester who found each defect

Explanation

Orthogonal defect classification assigns defects attributes in distinct dimensions such as type, trigger, and impact so that patterns can be analyzed by cause rather than by raw count. Cost, finder identity, and build number are useful metadata but do not provide the orthogonal causal attributes that support targeted process improvement.