Software Engineering MCQs - Software Testing and Debugging (Set 5)

Q1. A program's control flow graph has 10 edges, 8 nodes and is a single connected component. What is its cyclomatic complexity?

a) 4

b) 2

c) 3

d) 5

Q2. What is the cyclomatic complexity of the following pseudocode?

read(x, y)
while (x != y):
    if (x > y):
        x = x - y
    else:
        y = y - x
print(x)

a) 2

b) 4

c) 3

d) 5

Q3. Counting each simple condition of a compound predicate as a separate decision, what is the cyclomatic complexity of this fragment?

for i in 1..n:
    if (a[i] > 0 and a[i] < 100):
        count = count + 1
    elif (a[i] == 0):
        zeros = zeros + 1

a) 3

b) 5

c) 4

d) 6

Q4. Using standard (non-robust) boundary value analysis with the single-fault assumption, how many test cases are needed for a function of 2 independent input variables?

a) 5

b) 13

c) 25

d) 9

Q5. Robust boundary value testing also uses values just outside the valid range. How many test cases does it need for 3 input variables under the single-fault assumption?

a) 13

b) 15

c) 19

d) 27

Q6. A field accepts an age from 18 to 60 inclusive; all other integers are rejected. Using equivalence partitioning on this rule, how many equivalence classes are there?

a) 2

b) 3

c) 4

d) 6

Q7. Which statement about coverage criteria is correct?

a) 100 percent branch coverage guarantees 100 percent statement coverage

b) 100 percent statement coverage guarantees 100 percent branch coverage

c) Statement coverage is stronger than path coverage

d) Branch coverage ignores the false outcome of decisions

Q8. What is the minimum number of test cases needed to achieve 100 percent branch coverage of this code?

if (x > 5):
    y = 1
if (z < 3):
    y = y + 2

a) 1

b) 3

c) 4

d) 2

Q9. In bottom-up integration testing, the temporary components needed to call the modules under test are called

a) Stubs

b) Drivers

c) Mocks of the database only

d) Oracles

Q10. Beta testing is best described as testing performed

a) By real users in their own environment, without developers present

b) By developers on their own machines

c) By an independent team at the developer's site under supervision

d) Automatically by the CI server on each commit

Q11. After fixing a defect in the billing module, the team re-runs earlier tests on related modules to ensure nothing else broke. This is

a) Smoke testing

b) Acceptance testing

c) Load testing

d) Regression testing

Q12. Acceptance testing is typically carried out to

a) Check each function in isolation

b) Verify interfaces between two modules

c) Let the customer decide whether the system meets their requirements and should be accepted

d) Measure code coverage of unit tests

Q13. In mutation testing, 50 mutants are created. 5 of them are equivalent to the original program and 36 of the rest are killed by the test suite. What is the mutation score?

a) 72 percent

b) 90 percent

c) 64 percent

d) 80 percent

Q14. A debugger starts at the statement where the wrong output appears and traces the code in reverse to find where the value first went wrong. This debugging approach is

a) Backtracking

b) Brute force with memory dumps

c) Cause elimination by binary partitioning

d) Program slicing at compile time

Q15. "Are we building the product right?" is the question answered by

a) Validation

b) Certification

c) Verification

d) Benchmarking

Q16. A quick set of tests is run on each new build to check that the most critical functions work before deeper testing begins. This is called

a) Mutation testing

b) Smoke testing

c) Stress testing

d) Exhaustive testing

Q17. A programmer misreads the specification and writes the wrong formula; when the program runs, the user sees an incorrect total. The incorrect total is best termed a

a) Failure

b) Error

c) Fault

d) Mistake in specification

Q18. Which testing approach designs test cases from the internal structure of the code, such as its paths and branches?

a) Black-box testing

b) Acceptance testing

c) Usability testing

d) White-box testing