Data types — AQA A-Level Computer Science

Data types specify the attributes of data items, instructing the compiler or interpreter how bit patterns stored in registers and memory should be interpreted and manipulated. They enforce boundaries on variable values, define valid operations, and dictate physical memory allocation.

Mastery of data types is central to computer science, bridging low-level hardware architecture with high-level software engineering. In the AQA specification, a rigorous understanding prevents execution errors, eliminates vulnerabilities like buffer overflows, and ensures optimal space and time complexity across Paper 1 programming tasks and Paper 2 systems theory.

• →Primitive data types (Integer, Real/Float, Character, Boolean) serve as the fundamental building blocks, while Composite types (String, Array, Record) are formed by combining primitives.
• →Internal representation dictates precision and range: two's complement for signed integers, IEEE 754 mantissa and exponent for floating-point reals, and ASCII/Unicode encoding tables for characters.
• →Type systems categorize programming languages as strongly or weakly typed, and statically or dynamically typed, impacting compile-time safety and runtime performance.
• →Type coercion refers to automatic type conversion performed by the interpreter/compiler, whereas type casting denotes an explicit conversion directed by the programmer.

• Defines a data type as specifying the kind of value, the permitted operations and the storage representation.
• Gives appropriate examples such as integer, real, character, string and Boolean.
• Explains that the type determines which operations are valid for a value.
• Shows that the same numerical value may behave differently under different types, for example integer versus real division.
• Links the data type to how the value is stored in memory.

• 💡Use a concrete example, such as integer versus real division, to show understanding rather than only stating a definition.
• 💡Define the concept in terms of value category, permitted operations and storage, since all three are credited ideas.
• 💡Keep definitions precise and avoid vague phrases such as 'type of data' without explaining what the type controls.
• 💡When asked to choose a data type in an exam question, always state the exact type name and follow with two distinct justifications: the nature of the data (range/characters) and the operations required.
• 💡Be precise with terminology: distinguish clearly between 'overflow' (value exceeds the maximum representable limit) and 'underflow' or 'truncation error' (value is too small to be differentiated from zero or loses decimal precision).

• Treats a data type as only the visible value rather than the category governing storage and operations; state that the type defines the category and its rules.
• Confuses data type with variable name or identifier; state that the name labels a value while the type defines what kind of value it is.
• Assumes all numbers behave identically regardless of type; explain that integer and real division produce different results.
• Assuming all decimal fractions can be represented exactly in binary floating-point; fractions whose denominators are not powers of two result in infinite recurring binary representations.
• Believing Boolean variables always occupy strictly one bit of physical RAM during program execution; architectures usually allocate a full byte (8 bits) for byte-addressability and memory alignment.
• Treating strings as numeric values when they contain digits, ignoring that arithmetic operations cannot be directly applied without explicit type casting.

1. 1Day 1-2: Memorise the definitions, standard bit-widths, and typical value ranges for all five primitive data types.
2. 2Day 3-4: Practise identifying data types for ambiguous real-world data (dates, postcodes, currency, phone numbers) with written justifications.
3. 3Day 5-6: Explore binary floating-point representation, focusing on how mantissa and exponent widths affect range versus precision.
4. 4Day 7: Complete past AQA Paper 1 and Paper 2 questions testing data type selection, memory size calculations, and type casting.

• 📋Attribute mapping: Given a table of real-world entity records, choose the most appropriate data type and field length for each field.
• 📋Memory footprint calculation: Calculating total memory usage in bits or bytes for a given data structure containing mixed primitive types.
• 📋Comparative explanation: Explaining the architectural trade-offs between fixed-point and floating-point representations for real numbers.