You are testing a machine that scores exam papers and assigns grades. Based on the score achieved the grades are as follows: 1-49 = F, 50-59 = D-…

Boundary Value Analysis (BVA) is a test design technique that focuses on the values at the edges or boundaries of equivalence partitions. The problem defines six valid partitions for scores (1-49, 50-59, etc.) and two invalid partitions (scores 100). This creates seven distinct boundaries where the system's behavior changes: at the minimum/maximum valid scores (1 and 100) and between each adjacent grade range (e.g., between 49 and 50).

Standard two-point BVA requires testing the value on each side of every boundary. The seven boundaries are 0/1, 49/50, 59/60, 69/70, 79/80, 89/90, and 100/101. Each boundary requires two test cases, resulting in a total of 7 2 = 14 test cases.

A. 8: This count likely represents testing only one value from each of the eight partitions (six valid, two invalid), which is equivalence partitioning, not BVA.

C. 10: This number does not correspond to a standard application of BVA for the given ranges and would result in incomplete boundary coverage.

D. 12: This may represent testing only the six boundaries between the valid partitions (6 2 = 12), incorrectly omitting the tests for the outer boundaries (0/1 and 100/101).

1. International Software Testing Qualifications Board (ISTQB). (2018). Certified Tester Foundation Level Syllabus 2018 V3.1.1. Section 4.2.2, "Boundary Value Analysis," describes BVA as a technique for testing at the boundaries of equivalence partitions.

2. Black, R., van Veenendaal, E., & Graham, D. (2019). Foundations of Software Testing: ISTQB Certification (4th ed.). Cengage Learning. In Chapter 4, Section 4.3.3, the authors explain that for each boundary, two tests are typically created: one on the boundary and one just off the boundary in the adjacent partition.

3. Myers, G. J., Sandler, C., & Badgett, T. (2011). The Art of Software Testing (3rd ed.). John Wiley & Sons. Chapter 4, "Test-Case Design Strategies," pp. 56-61, details BVA, emphasizing that errors tend to cluster around boundaries, justifying the selection of test cases on and around these boundary values.