Software Testing Fundamentals Methods Flashcards

Question 1

What is the main purpose of software testing?

Accepted Answer

Software testing is the process of evaluating a system to find defects and to gain confidence that it behaves as its requirements and users expect. Testing cannot prove a program is completely free of faults, but it reduces risk by revealing failures early, when they are cheaper to fix.

Question 2

What does the principle 'testing shows the presence of defects, not their absence' mean?

Accepted Answer

Testing can demonstrate that defects exist, but passing tests cannot prove that a product has no defects left. A test suite only samples the many possible inputs, states, and environments, so unexecuted paths may still hide faults. Confidence grows with good coverage and well-chosen tests, but absolute certainty is not achievable.

Question 3

Why is exhaustive testing usually impossible for real software?

Accepted Answer

Exhaustive testing means running every possible combination of inputs and conditions, and even simple programs have far more combinations than could be executed in any reasonable time. Testers therefore select a manageable set of tests using risk, priority, and design techniques to cover the most important behavior.

Question 4

What is the difference between an error, a defect, and a failure?

Accepted Answer

The three terms form a cause chain. A human error is a mistake made by a person during development. That mistake can leave a defect, a flaw in the code, design, or documentation. When the defect runs under certain conditions, it can produce a failure, an observable incorrect behavior.

Question 5

What is the purpose of the test oracle?

Accepted Answer

A test oracle is the mechanism that decides whether the observed behavior of a test is correct. It may be a specification, a previously validated result, a calculation, or the judgment of a tester who knows the expected behavior. Without a reliable oracle, a test can run successfully yet still miss a failure.

Question 6

Why does testing early in development reduce the cost of fixing defects?

Accepted Answer

Defects found early are usually cheaper to fix because fewer other components depend on the faulty work. When a requirement or design error is discovered during requirements review, correction may require editing a single document. Discovered after release, the same error may require code changes, retesting, redeployment, and support work for affected customers.

Question 7

What is the difference between verification and validation in testing?

Accepted Answer

Verification checks whether the product is built correctly according to its specifications, while validation checks whether the right product is being built for the users' needs. Verification asks, did we build it right? Validation asks, did we build the right thing? Both activities together give a fuller picture of quality than either one alone.

Question 8

Why do defects tend to cluster in a small number of modules?

Accepted Answer

Defects tend to cluster because a small number of modules often carry most of the complexity, change frequency, and new code. However, clustering is an observed tendency rather than a fixed law, so testers should still examine quieter areas that might hide problems.

Question 9

What is black-box testing and how does it differ from white-box testing?

Accepted Answer

Black-box testing designs tests from the specified behavior of the software, such as requirements, inputs, and outputs, without relying on the internal code structure. White-box testing, by contrast, uses knowledge of the code, branches, and logic to design tests. Black-box methods suit user-level and specification-level checks, while white-box methods target internal paths.

Question 10

How does equivalence partitioning help choose test inputs?

Accepted Answer

Equivalence partitioning divides the input domain into groups, called partitions, where the software is expected to behave the same way for every value in the group. Testers then pick one representative value from each valid and invalid partition rather than testing every possible value.

Question 11

What is boundary value analysis, and why does it focus on edges?

Accepted Answer

Boundary value analysis is a technique that selects test values at the edges of each valid range, along with values just outside those edges. Programming errors often occur at boundaries. For a range from 1 to 100, typical tests would use 0, 1, 2, 99, 100, and 101.

Question 12

What does a decision table help a tester do?

Accepted Answer

A decision table helps a tester capture complex business rules by listing every combination of conditions and the actions that should follow from each combination. Testers can then turn each column into a test case, confirming that the system produces the expected action for that combination.

Question 13

How does state transition testing help reveal faults in workflows?

Accepted Answer

State transition testing models the system as a set of states, the events or inputs that trigger changes, and the transitions that result. Tests cover valid transitions, rejected invalid transitions, and unusual sequences. This approach is useful for workflows where the response depends on history.

Question 14

What is the main benefit of pairwise testing when many parameters interact?

Accepted Answer

Pairwise testing creates test sets in which every combination of values for any two parameters appears at least once. Testing all combinations of many parameters is often impractical, yet many failures are triggered by interactions between only two factors. This approach therefore finds many interaction defects with far fewer tests.

Question 15

What is cause-effect graphing and what does it produce?

Accepted Answer

Cause-effect graphing is a technique that models how input causes combine to produce effects, using logical relationships such as and, or, and not. The graph is converted into a decision table, and each column becomes a test case. It suits complex logic where many inputs interact.

Question 16

How do tests derived from use cases differ from tests derived from requirements lists?

Accepted Answer

Use case testing builds tests from scenarios showing how an actor interacts with the system to reach a goal, including main success paths and alternate or exception flows. A plain requirements list gives individual statements that may be tested one by one without showing the sequence of steps.

Question 17

What is statement coverage and what does it measure?

Accepted Answer

Statement coverage measures the percentage of executable statements in the code that are run by at least one test. It can reveal code that has never executed. However, a suite can reach full statement coverage while still missing faults in conditions or decisions that were never tested in both outcomes.

Question 18

How does branch coverage differ from statement coverage?

Accepted Answer

Branch coverage requires each decision to be exercised with both its true and false outcomes, while statement coverage only requires each statement to run at least once. An if statement without an else can be fully statement-covered by a single test that takes the true path, yet the false path remains untested. Branch coverage therefore detects more untested decision outcomes than statement coverage does.

Question 19

What is condition coverage, and how is it different from decision coverage?

Accepted Answer

Condition coverage requires each individual Boolean condition inside a decision to take both true and false values at least once. Decision coverage checks only the overall true or false outcome of the whole decision. The two criteria are different, and satisfying one does not automatically satisfy the other, so testers often aim for both.

Question 20

What is path coverage, and why is it often impractical?

Accepted Answer

Path coverage requires that every possible route through a program's control flow be executed at least once. Loops and decision sequences multiply the number of routes, so modest programs can have vast numbers of paths. Because of this explosion, testers usually rely on branch coverage or focus path testing on critical functions.

Question 21

What is a control flow graph used for in white-box testing?

Accepted Answer

A control flow graph represents a program as nodes for blocks of sequential statements and edges for possible transfers of control between them. Testers use it to visualize decisions, loops, and merge points. Each path through the graph corresponds to a possible execution sequence that a test may need to exercise.

Question 22

How is cyclomatic complexity calculated for a single function?

Accepted Answer

One common method starts at one and adds one for each decision point, such as an if, loop, or case branch. In a control flow graph, the value equals edges minus nodes plus two for a single connected graph. A higher value usually means more independent paths and more tests are needed.

Question 23

What is data flow testing and what kind of fault does it target?

Accepted Answer

Data flow testing examines how values are defined, used, and redefined as they move through a program, focusing on pairs of definitions and later uses of variables. It targets faults such as using a variable before it is assigned or using a value that an intervening statement has overwritten.

Question 24

Why is mutation testing used to assess the quality of a test suite?

Accepted Answer

Mutation testing makes small deliberate changes to a program and checks whether existing tests detect each changed version. A mutant that causes a test to fail is called killed, while a surviving mutant reveals a gap in the tests. The proportion of killed mutants indicates how well the suite catches faults.

Question 25

What are the main test levels in the traditional testing model?

Accepted Answer

The commonly taught test levels are unit testing, integration testing, system testing, and acceptance testing. Unit testing checks individual components in isolation, integration testing checks how components work together, system testing evaluates the complete integrated product, and acceptance testing confirms that it meets business and user needs.

Question 26

What is the main purpose of regression testing?

Accepted Answer

Regression testing checks that changes to software, such as bug fixes, new features, or configuration updates, have not broken functionality that previously worked. Testers usually reuse existing test cases and select them based on the areas affected by the change. Because any change can introduce side effects, regression testing is repeated throughout the project.

Question 27

How does smoke testing differ from sanity testing?

Accepted Answer

Smoke testing is a broad, shallow check that a new build is stable enough for further testing, covering the most critical functions. Sanity testing is a narrower check after a specific change, confirming that the fixed behavior works and the build is reasonable to test further. Both are quick, but smoke testing looks across the whole application while sanity testing focuses on a small, targeted area.

Question 28

What is the difference between functional and non-functional testing?

Accepted Answer

Functional testing checks what the system does, verifying that features, calculations, and workflows behave according to their requirements. Non-functional testing checks how well the system performs, covering qualities such as performance, security, usability, reliability, and compatibility. A system can produce correct results yet still be too slow, insecure, or difficult to use.

Question 29

What is the difference between alpha and beta testing?

Accepted Answer

Alpha testing is performed internally, usually by the development organization or a controlled group of internal users, before the product is released outside the organization. Beta testing is performed by real external users in their own environments, providing feedback on usability and unexpected issues. Alpha testing typically occurs earlier, when the product is less mature, and beta testing occurs closer to release. Feedback from both phases helps teams decide which defects to fix before the final release.

Question 30

What does security testing try to uncover in a system?

Accepted Answer

Security testing looks for weaknesses that could let unauthorized people access, change, or destroy data or disrupt service. Typical areas include authentication, authorization, input handling that could allow injection attacks, and how sensitive information is stored and transmitted. Testers often combine manual review with automated scanning, since neither approach finds every weakness alone.

Question 31

What is performance testing and what does load testing measure?

Accepted Answer

Performance testing evaluates speed, responsiveness, stability, and resource use under defined workloads. Load testing applies expected levels of concurrent users or transactions to see whether response times and error rates remain acceptable. Stress testing goes further by pushing the system beyond expected limits to observe how it fails and recovers.

Question 32

Why do testers perform retesting after a defect is fixed?

Accepted Answer

Retesting, also called confirmation testing, is performed to verify that a specific defect has actually been resolved. The tester repeats the test that originally failed and checks that the expected result now occurs. Retesting is separate from regression testing, which looks for new problems that the fix may have caused elsewhere.

Question 33

What are the main activities in a typical software test process?

Accepted Answer

A typical test process includes planning, analysis and design, implementation, execution, evaluation of exit criteria, and closure. Planning defines scope, resources, schedule, and risks, while analysis and design derive test conditions and cases from the test basis. Implementation prepares data and environments, execution runs tests, and closure archives results.

Question 34

What is a test case, and what should it include?

Accepted Answer

A test case is a set of preconditions, inputs, actions, and expected results designed to check a specific behavior. A good test case also has a unique identifier, a clear purpose, and steps that another tester could follow to get the same result. Writing expected results before execution helps testers judge the outcome objectively.

Question 35

What is a test plan, and why is it useful?

Accepted Answer

A test plan describes the scope, objectives, approach, resources, schedule, environments, and risks for testing a product or release. It gives stakeholders a shared understanding of what will be tested, how success will be judged, and what might prevent progress. Because plans change as projects evolve, they should be updated as conditions change.

Question 36

What are exit criteria, and how are they used?

Accepted Answer

Exit criteria are conditions that must be met before testing is considered complete, such as coverage targets, pass rates, or the number of open critical defects. They give managers an objective basis for deciding whether a release is ready. Criteria should be agreed upon early so that everyone understands what completion means.

Question 37

What information should a good defect report contain?

Accepted Answer

A good defect report includes a clear title, the steps needed to reproduce the problem, the test data and environment, the expected result, and the actual result. It should also note severity and priority, attach evidence such as logs or screenshots, and state the build or version tested.

Question 38

What is defect severity, and how does it differ from priority?

Accepted Answer

Defect severity describes the technical impact of a defect on the system, such as a crash, data loss, or a minor cosmetic flaw. Priority describes how urgently the defect should be fixed, based on business needs, release timing, and customer impact. The two can differ, since a rare crash may have high severity but low priority, while a visible typo on a launch page may have low severity but high priority.

Question 39

What stages does a defect typically move through during its life cycle?

Accepted Answer

A defect typically moves through stages such as new or reported, assigned, open or in progress, fixed or resolved, retested, and closed. If retesting shows the problem remains, the defect is reopened and returned to development. Some organizations also use rejected, deferred, or duplicate states.

Question 40

Why is root cause analysis useful after important defects are found?

Accepted Answer

Root cause analysis investigates why a defect entered the product, rather than only what the defect is. By identifying process gaps, such as unclear requirements, missing reviews, or insufficient test coverage, teams can change practices so similar defects are prevented or detected earlier. The goal is improvement of the process, not blaming individuals.

Software Testing Fundamentals Methods Flashcards

1Testing Principles and Fundamentals

2Black-Box Test Design Techniques

3White-Box Test Design Techniques

4Test Levels and Test Types

5Test Process and Defect Management

Learning Resources