Strongest Structural Testing Technique: Statement vs Branch vs Multiple Condition Coverage

In structural (white-box) testing, coverage criteria define what parts of a program must be exercised by test inputs. Among the options—statement coverage, branch coverage, and multiple condition coverage—the strongest is typically the one that imposes the most detailed requirements on how conditions inside decisions are evaluated.

We compare them by strength/implication: if a test suite satisfies a stronger criterion, it also satisfies any weaker criterion that is logically implied.

A decision often has a structure like:

1if (C1 op1 C2 op2 C3 ...) S1 else S2

• Statement coverage requires that each statement (line/instruction) is executed at least once.
• Branch coverage requires that each control-flow outgoing branch of each decision is taken at least once (e.g., both sides of each if, and each target of switch cases).
• Multiple condition coverage requires that, within each decision containing multiple atomic conditions, all combinations of truth values of those conditions occur at least once (subject to realizability).

This creates an implication chain in terms of requirements:

Therefore, multiple condition coverage-based testing is the strongest among the three, because it constrains not only which branches execute, but also which truth-value combinations of the individual conditions occur.

Key keyword term relationships:

• Coverage criterion
• Statement coverage
• Branch coverage
• Multiple condition coverage

Answer: Multiple condition coverage-based testing is the strongest, because it subsumes the requirements of branch coverage and statement coverage.

Concrete Justification Using a Decision Example

Consider this decision:

1if (A && B) {
2  S1;
3} else {
4  S2;
5}

What each criterion enforces

• Statement coverage

  • Needs at least one test where S1 executes and at least one where S2 executes.
  • Example tests might be:
    • Test 1: A=true, B=true → executes S1
    • Test 2: A=false, B=false → executes S2
  • Notice: you can achieve statement coverage without ever forcing all combinations of (A, B).

• Branch coverage

  • Needs both branches: the then branch and the else branch must be taken.
  • For if (A && B), the then branch requires A=true and B=true.
  • The else branch requires at least one of the conditions to be false.
  • You still may not explore all (A, B) combinations—only enough to ensure both outcomes occur.

• Multiple condition coverage

  • Requires all combinations of atomic conditions for the decision.
  • For two conditions (A, B), that means:
    • (A=true, B=true )
    • (A=true, B=false)
    • (A=false, B=true )
    • (A=false, B=false)
  • This set necessarily includes inputs that trigger both branches, and it executes the statements accordingly.

Strength implication (why multiple condition is strongest)

If you satisfy multiple condition coverage, then by construction you will also generate inputs that:

1. Make the overall decision evaluate to both true and false (hence both branches get exercised), and
2. Execute statements in both the then and else blocks.

So multiple condition coverage implies branch coverage, which implies statement coverage—making it the strongest among the three.