How DP Computer Science Is Assessed: Revision Notes

Condensed for quick recall of how the course is assessed. For the full subject overview, use the IB DP Computer Science subject overview.

Why Paper 2 has no fixed percentage listed separately

Paper 1 and Paper 2 together make up the external assessment share remaining after the IA’s weighting is subtracted (70% at SL, 80% at HL), with Paper 1 taking a larger individual share of that total at both levels. Rather than memorising Paper 2’s exact percentage in isolation, it’s more useful to remember the two totals that matter for planning: internal assessment (30% SL / 20% HL) versus combined external papers (70% SL / 80% HL), since that split drives how revision time should realistically be divided between IA polishing and paper-style practice.

SL vs HL weightings

Component	SL	HL
Paper 1	35%	40%
Paper 2	(remaining external %)	(remaining external %)
Internal assessment (computational solution)	30%	20%

Paper 1 vs Paper 2

• Paper 1: extended-response questions on core concepts, plus questions based on a pre-seen case study released in advance.
• Paper 2: focuses on algorithmic thinking, assessed at the level of pseudocode — not any specific programming language’s syntax. It tests reasoning about an algorithm’s logic.

The computational solution (IA)

• No prescribed programming language — choose whatever suits the solution.
• Marked against criteria covering the whole problem-solving cycle (identify → design → develop → evaluate), not just the finished product.
• Weighted more heavily at SL (30%) than HL (20%).

The computational solution in more depth

The IA is graded against criteria that mirror the full problem-solving cycle: planning (defining the problem clearly and specifying success criteria); system overview and rationale (justifying the choice of tools, techniques and approach); development (implementing the solution, including evidence of the process, not just the final code); and evaluation (assessing the solution against the original success criteria, and identifying genuine strengths and limitations). A common weakness is submitting a technically working program with little documented evidence of the design decisions and iterations that led to it – since development process, not only the finished product, is explicitly assessed, keeping a running log of key decisions and problems solved during development substantially strengthens this component.

Command terms and how the papers use them

Paper 1’s extended-response questions on core concepts use command terms consistent with the DP sciences group: define, identify and outline for lower-tier recall; explain, describe and compare for middle-tier reasoning; evaluate, discuss and justify for top-tier questions requiring a reasoned judgement, often applied to the pre-seen case study. Paper 2’s algorithmic questions instead use command terms specific to computational thinking – trace, construct and write an algorithm – which test the ability to follow, build or design a solution in pseudocode rather than recall factual content, a genuinely different skill from Paper 1’s more discursive style.

Why the pre-seen case study matters more than it first appears

Because Paper 1’s case study is released well in advance of the exam, questions drawing on it are, in effect, partially predictable – research and familiarity built before the exam translate directly into faster, more confident answers on exam day. Students who treat the case study as background reading to skim once, rather than material to actively analyse and annotate in the weeks before the exam, consistently underperform relative to their preparation on the rest of the syllabus, since case-study-based questions typically carry a meaningful share of Paper 1’s marks.

How SL and HL differ beyond the headline weightings

Beyond the shift in Paper 1’s weighting (35% at SL, 40% at HL) and the IA’s weighting (30% at SL, 20% at HL), HL students study additional syllabus content beyond the SL core and are examined on it within the same paper structure, rather than through a separate HL-only paper as in some other DP science and mathematics courses. This means HL revision should be built as “SL content plus HL extension,” integrated together rather than treated as two separate bodies of material, since Paper 1 and Paper 2 at HL draw on both without distinguishing which parts of a question are “SL-level” and which are “HL-level.”

Exam traps

• Preparing Paper 2 revision around a specific language’s syntax, when the exam actually assesses algorithmic logic in pseudocode.
• Underweighting Paper 1 case-study preparation, since it’s released in advance and rewards research done before the exam.
• Treating the computational solution as only about the final working program, when the process (identify, design, develop, evaluate) is what’s assessed.

Self-test

1. What does Paper 2 actually assess algorithms in — a specific language, or pseudocode?
2. What advance material does Paper 1 draw on?
3. How does the IA’s weighting differ between SL and HL?
4. What four stages does the computational solution’s assessment criteria mirror?
5. Why does the IA benefit from a documented log of design decisions during development?
6. Why is the pre-seen case study, in effect, partially predictable exam content?

Answers: 1. Pseudocode — testing reasoning about an algorithm’s logic, not a specific language’s syntax. 2. A pre-seen case study, released in advance of the exam. 3. 30% at SL, 20% at HL. 4. Identify a problem, design a solution, develop it, and evaluate it. 5. Because the assessment criteria explicitly cover the development process, not just the finished product, so evidence of decisions and iterations strengthens this component beyond what the final code alone shows. 6. Because it is released well in advance of the exam, so research and familiarity built beforehand translate directly into faster, more confident answers on questions drawing on it.

Official syllabus

International Baccalaureate Organization, Diploma Programme Subject Brief – Sciences: Computer Science, first assessment 2027, published 2024 – the same source cited by the full syllabus guide and the subject overview.