IB DP Computer Science B.2: Programming

This guide covers B.2 Programming, the largest single sub-topic in the entire DP Computer Science syllabus at SL (40 hours), remaining large at HL (42 hours), as set out in the full syllabus guide, first assessment 2027.

Where this fits in the syllabus

B.1 Computational Thinking establishes a specify-decompose-abstract- test process that is meant to be applied while programming, not learned separately from it. Revising B.2 in isolation — purely as syntax recall — misses what the syllabus actually rewards: being able to decompose a described problem into smaller, programmable steps, and to test a solution systematically rather than only running it once and checking the output looks plausible. Paper 2 draws on Theme B, including B.2, and (for HL) the OOP and abstract-data-type extensions in B.3 and B.4.

Syllabus coverage

IB DP COMPUTER SCIENCE — B.2: PROGRAMMING

Whichever language a course uses (Python or Java), the underlying constructs examined are the same:

• Sequence, selection, iteration — the three basic control structures every program is built from: executing instructions in order, branching based on a condition (if/else), and repeating a block of instructions (for/while loops)
• Data types and structures — primitive types (integer, real/float, boolean, character/string) and simple structures such as one-dimensional and two-dimensional arrays/lists, and choosing an appropriate structure for a given problem
• Modularity — breaking a program into functions/procedures with defined inputs and outputs, the direct programming expression of B.1’s “decompose” step
• File and data handling — reading from and writing to files, and validating input, which matters for any solution that needs to persist or check data rather than just process it in memory once
• Pseudocode and program design — representing a solution’s logic in pseudocode or a flowchart-style design before writing code
• Testing and debugging — normal, boundary and erroneous test data; tracing through code to find where actual behaviour diverges from expected behaviour

How to approach it

Practise converting a described real-world problem directly into pseudocode steps, then only afterwards translating that pseudocode into actual Python or Java syntax — this two-step habit is exactly what B.1’s computational thinking process asks for, and is a more reliable way to avoid logic errors than writing code directly from a problem description. A program that runs without crashing is not the same as a program that is correct: a well-designed test plan checks normal (typical, expected), boundary (values at the very edge of what is valid), and erroneous (invalid values the program should reject or handle gracefully) test data against a defined set of expected outcomes before a solution is considered complete. Because code- reading and code-writing draw on the same underlying fluency, practise both — predicting what a given piece of pseudocode or code outputs, as well as producing a short program or function to a specification — not just writing your own programs from scratch.

Worked example: designing a test plan

A program is meant to accept a student’s exam score (0–100) and output a grade. A test plan for this program should check:

Normal:      score = 75  -- a typical, valid mid-range value

Boundary:    score = 0 and score = 100  -- the extreme valid values
             at each edge of the accepted range

Erroneous:   score = -5 and score = 150  -- invalid values outside
             the accepted range, which the program should reject or
             handle gracefully, for example with an error message
             rather than crashing or silently producing a wrong
             grade

A test plan that only checks a normal value (score = 75) has not actually confirmed the program handles the edges of its valid range or rejects invalid input correctly — all three categories are required before a solution can be considered properly tested.

Connecting to the computational solution internal assessment

The 35-hour computational solution internal assessment is where B.2’s skills are applied at full scale: designing, building and documenting a solution to a real-world problem the student defines. Because the IA is scored partly on the quality of design and testing documentation, not just on whether the final program works, practising writing a clear test plan and recording actual test results (not just “it worked”) during any programming practice across the course builds a habit that pays off directly when the IA itself is due – a student who only ever runs a program once and moves on, without documenting what was tested and what the result was, arrives at the IA without the documentation habit it specifically rewards.

Common mistakes

Writing code directly from a problem description without a pseudocode or design step first, which increases the risk of logic errors. Testing a program only with typical, expected values and skipping boundary and erroneous test data. Treating “the program ran without crashing” as equivalent to “the program is correct.” Practising only code-writing and neglecting code-reading, despite Paper 2 testing both skills. Treating modularity as a stylistic preference rather than the direct programming expression of decomposing a problem.

Quick revision checklist

• Practise the pseudocode-then-code two-step habit for unfamiliar problems.
• Build a test plan with normal, boundary and erroneous data for every program written during revision.
• Practise both code-reading (predicting output) and code-writing questions, not just one or the other.
• Know how modularity connects directly to B.1’s “decompose” step.
• Carry good testing and documentation habits from B.2 practice forward into the computational solution internal assessment.

Official syllabus

International Baccalaureate Organization, Diploma Programme Subject Brief – Sciences: Computer Science, first assessment 2027, © 2024 — the same source already cited by the full syllabus guide, which first reproduced B.2’s teaching hours and its place within Theme B. Verified 2026-09-06.