Problem identification — OCR A-Level Computer Science

Problem identification is the foundational stage of computational thinking and the software development lifecycle. It involves recognising and defining a real-world problem that can potentially be solved using a computer system. This step is critical because a poorly defined problem leads to wasted resources, incorrect solutions, and project failure. In the OCR A-Level specification, problem identification is assessed through both theory questions and the non-exam assessment (NEA), where students must justify their choice of problem and demonstrate a clear understanding of its scope and requirements.

The process begins with identifying a need or opportunity, often through stakeholder consultation, observation, or analysis of existing systems. Students must learn to distinguish between a problem that is suitable for computational solution and one that is not. Key activities include writing a clear problem statement, defining success criteria, and considering constraints such as time, cost, and technical feasibility. This stage directly feeds into analysis and design, making it the cornerstone of any successful project.

In the wider context of computer science, problem identification teaches students to think critically about the world around them and to apply computational thinking to break down complex issues. It also introduces ethical considerations, such as whether a problem should be solved with technology and the potential impact on users. Mastering this topic ensures students can approach both exam questions and their NEA with confidence, as they will be able to articulate the purpose and value of their proposed solution.

• →Problem definition: Writing a concise, unambiguous statement that describes the problem to be solved, including who is affected and what the desired outcome is.
• →Success criteria: Measurable outcomes that define what a successful solution must achieve, such as speed, accuracy, or user satisfaction.
• →Stakeholder identification: Recognising all individuals or groups affected by the problem, including end-users, clients, and developers.
• →Feasibility study: Assessing whether the problem can be solved within given constraints (time, budget, technology, legal/ethical boundaries).
• →Scope definition: Clearly outlining the boundaries of the problem to avoid scope creep and ensure the solution remains manageable.

• Describe and justify features that make the problem solvable by computational methods.
• Explain why the problem is amenable to a computational approach.
• Identify and describe stakeholders, explaining how the solution is appropriate to their needs.
• Research the problem and similar existing solutions to justify the chosen approach.
• Identify and describe essential features of the proposed computational solution.
• Explain and justify limitations of the proposed solution.
• Specify and justify solution requirements, including hardware and software configurations.
• Identify and justify measurable success criteria for the solution.

• 💡Ensure the chosen problem is non-trivial and allows for a substantial coded element.
• 💡Use the command words in the assessment criteria to drive the depth of your evidence.
• 💡Ensure all evidence is authentic and individual to the learner.
• 💡Focus on justifying decisions rather than just describing them.
• 💡Ensure the problem is well-defined and user-driven.
• 💡In the NEA, examiners look for a clear, well-defined problem that is neither too trivial nor too complex. Choose a problem that allows you to demonstrate a range of skills, such as data processing, user interface design, and testing. Avoid over-ambitious projects that cannot be completed in the time available.
• 💡When writing about problem identification in exams, always link the problem to a specific user or stakeholder. Use phrases like 'the problem affects...' and 'the solution must achieve...' to show you understand the real-world context. This demonstrates higher-level thinking.
• 💡Be precise with success criteria. Instead of 'the system should be fast', say 'the system should process a query in under 2 seconds'. Quantifiable criteria are easier to test and show a deeper understanding of requirements.

• Choosing a problem that is too trivial to demonstrate the required range of computational skills.
• Failing to justify why the problem is suitable for a computational approach.
• Neglecting to identify measurable success criteria, making evaluation difficult later.
• Lack of depth in researching existing solutions to similar problems.
• Failing to link the proposed solution features back to the identified stakeholder needs.
• Misconception: Problem identification is just about stating the problem. Correction: It also involves understanding the context, stakeholders, and constraints. A vague problem statement leads to a vague solution.
• Misconception: Any problem can be solved with a computer. Correction: Some problems are not suitable for computational solutions due to ethical concerns, lack of data, or the need for human judgment. Students must evaluate suitability.
• Misconception: Success criteria are optional or can be added later. Correction: Success criteria must be defined early to guide development and testing. Without them, it's impossible to know if the solution is effective.