Computer Science Exam Tips

The 5-Minute Habit That Saves a Grade: Scanning and Parsing the Script

Top-performing candidates don't just dive headfirst into writing code or answering recall questions; they start with a disciplined 5-minute scan. In Paper 1 (Principles of Computer Science), where calculators are strictly prohibited, you must spot computational questions early. Look out for image file size estimations, binary conversions, and transmission calculations. Identifying these allows you to carefully write out your working spaces. Remember, examiners award marks for the correct mathematical expression even if you make an arithmetic slip. For instance, expressing a file size as \( \frac{1920 \times 1080 \times 16}{8 \times 1024 \times 1024} \) MiB secures full calculation marks, whereas wasting precious time attempting complex long division can lead to compounding mistakes.

Where the Marks Really Hide: The Trap of Relative Comparisons

When comparing technologies—such as LANs and WANs—candidates frequently lose easy marks by stating the same point twice from opposite poles. If a question asks for two differences, writing "WAN has higher latency" and "LAN has lower latency" will only earn you one mark. Examiners treat these as two sides of a single comparative coin. To secure both marks, you must target distinct parameters: one point for geographic range (or ownership) and another for latency (or bandwidth). Additionally, avoid vague, single-word answers. Simply writing "clock speed" when asked what affects CPU performance is insufficient; you must state that a higher clock speed increases the number of fetch-decode-execute cycles processed per second.

The Architecture of a Top-Score Algorithm: Flowcharts and Boundaries

Flowcharts are a 6-mark goldmine on Paper 1, but they are often undermined by minor structural omissions. Every line connecting your process, input/output, and decision blocks must have directed arrows indicating the flow of execution. Without these direction indicators, the flowchart fails to show a functional program path. When drawing decision diamonds, you must also label every exit path clearly with "Yes"/"No" or "True"/"False".

Furthermore, pay close attention to boundary values in validation algorithms. Substituting strict inequalities like > 31 instead of >= 31 is a classic error that will cost you the boundary test mark. Always test your logic with boundary limits (e.g., if the limit is inclusive of 0 and 255, ensure your code checks >= 0 and <= 255).

The C# / Python / Java Examiner Secrets: Closing Files and Tracking State

For Paper 2 (Application of Computational Thinking), the code you write must interact safely with system resources. Two areas consistently trip up candidates in the practical examination:

• File Handling: When writing output streams to text files (e.g., grades.txt), you must explicitly call the close method (e.g., writer.Close() or theFile.close()) at the end of your program. Failing to close file handles leads to resource leaks and can prevent your outputs from saving correctly, resulting in a loss of functionality marks.
• Variable Initialization: Do not rely on implicit default values. Always initialize your counters and accumulators to standard base types (like setting myAge = 0 or total = 0) before entry into any loop structure. This prevents garbage data from corrupting your calculations.
• Dynamic Array Processing: Never hardcode array boundaries (e.g., using a fixed index like 10 to overwrite a blank element). Your code must dynamically calculate and use the array's actual size (e.g., using tblWords.GetLength(0) in C# or len(tblWords) in Python) so that the program functions correctly even if the size of the dataset changes.

Metaphors for Mastering Complex Theory

When revising tricky theoretical concepts, use these simple mental pictures to keep your facts straight: