Lesson 4: Structured Analysis and Data Modeling

Structured analysis focuses on representing how data moves through a system and how processes transform that data.

A context diagram shows the system as a single process and identifies external entities, major data inputs, and major data outputs. It gives the highest-level view.

Example for a clearance system:

• External entities: Student, Registrar, Cashier, Department Office
• Inputs: clearance request, payment confirmation, approval decision
• Outputs: status update, clearance result

A Data Flow Diagram (DFD) breaks the system into smaller processes. It usually contains:

Levels of DFD

• Context diagram: whole system as one process
• Level 0 DFD: major subprocesses
• Lower-level DFDs: more detailed decomposition of one process

Example Level 0 processes for a clinic system: Manage patient records; Schedule appointments; Issue medicines; Generate reports.

Rules to remember

• Every process should have at least one input and one output.
• Data should not move directly from one external entity to another inside the model.
• Data should not move directly from one data store to another without a process.
• Lower-level DFDs should remain consistent with higher-level ones. This is called balancing.

DFDs are useful because they force you to think about the actual movement of information, not just screens.

A DFD shows movement, but it does not fully define each data item. That is why analysts also prepare a data dictionary — a description of the meaning of data elements, records, and structures.

Example:

A stronger dictionary may also include type, format, allowed values, source, destination, and validation rule.

Process specification

Some DFD processes are simple enough to understand from the title alone. Others need more explanation. A process specification describes the internal logic of a process, written using structured English, decision tables, decision trees, or simple algorithmic steps.

Example using structured English:

1. Receive clearance request.
2. Verify student identity.
3. Retrieve balance and disciplinary records.
4. If all units are clear, mark as approved.
5. Otherwise, return pending requirements list.

This helps developers and testers understand what the process is supposed to do before code is written.