Project Rubric for Bachelor's Computer Science: Relational Database Design

Key Indicators

• •Extracts distinct entities and relevant attributes directly from functional requirements
• •Specifies accurate cardinality and participation constraints for all relationships
• •Decomposes many-to-many relationships into associative entities or linking tables
• •Distinguishes correctly between strong and weak entities based on identifying relationships
• •Adheres to standard ER diagrammatic notation consistently

Grading Guidance

To progress from Level 1 to Level 2, the student must move beyond a chaotic or purely illustrative sketch to a structured diagram that attempts standard notation. While Level 1 work often misses core entities or uses arbitrary shapes, Level 2 identifies the primary entities and attributes, though it may struggle with relationship mechanics (e.g., inverted cardinalities or unresolved many-to-many links). Crossing into Level 3 (Competence) requires semantic soundness; the student must successfully resolve all many-to-many relationships into associative entities and ensure that the primary keys and foreign keys (implied by relationships) support the basic data integrity required by the scenario. The transition from Level 3 to Level 4 marks a shift from basic functionality to precise semantic mapping. A Level 4 model correctly handles advanced constructs—such as weak entities, recursive relationships, or complex participation constraints (mandatory vs. optional)—that Level 3 might oversimplify or ignore. Finally, achieving Level 5 requires an elegant, optimized abstraction. At this level, the student employs sophisticated modeling techniques (e.g., generalization/specialization hierarchies) to reduce redundancy and anticipates future scalability, producing a professional-grade schema that captures subtle business rules without over-complicating the design.

Proficiency Levels

Key Indicators

• •Structures DDL scripts with precise data types and primary/foreign key constraints
• •Translates logical models into functional physical schemas without normalization violations
• •Constructs complex DML queries involving multi-table joins, aggregations, and subqueries
• •Optimizes query performance by avoiding Cartesian products and redundant logic
• •Validates query results against functional requirements to ensure data accuracy

Grading Guidance

To advance from Level 1 to Level 2, the student must provide SQL scripts that execute without syntax errors, moving beyond pseudo-code or broken fragments to functional table creation. The transition to Level 3 marks the competence threshold where referential integrity is strictly enforced; the code must not only run but also correctly implement Primary and Foreign Keys, select appropriate data types (e.g., avoiding universal VARCHAR usage), and produce accurate result sets for standard join operations. Moving from Level 3 to Level 4 involves a shift from basic functional correctness to structural quality and complexity. While Level 3 queries return the right data using simple logic, Level 4 queries demonstrate intermediate proficiency, such as effectively using nested subqueries, complex aggregations, or HAVING clauses, while maintaining clean formatting. The student avoids common inefficiencies, such as selecting unnecessary columns or poor join conditions. To reach Level 5, the work must exhibit professional-grade efficiency and sophistication. The SQL implementation handles edge cases robustly, utilizes advanced features (like views, triggers, indexes, or stored procedures) where appropriate to solve specific problems, and is formatted for maximum maintainability. The distinction lies in the elegance of the logic—solving complex data retrieval problems with concise, highly performant code rather than brute-force or procedural methods.

Proficiency Levels

Key Indicators

• •Articulates technical rationale for architectural and design decisions.
• •Applies standard diagramming notations (e.g., UML, Crow's Foot) with syntactic accuracy.
• •Structures the narrative to guide the reader logically through technical complexities.
• •Integrates visual aids and text seamlessly to reinforce technical concepts.
• •Maintains professional formatting and grammatical precision throughout the report.

Grading Guidance

To advance from Level 1 to Level 2, the report must transition from fragmented notes or missing artifacts to a recognizable technical document. The student attempts to use diagrams and headings, though the notation may be inventive rather than standard, and the reasoning may be superficial. The threshold for Level 3 (Competence) requires the strict application of standard conventions; diagrams must use correct syntax (e.g., proper cardinality symbols in ERDs) rather than ad-hoc shapes, and the text must clearly describe the system's components without significant grammatical friction. The leap from Level 3 to Level 4 marks the shift from description to justification. While a Level 3 report accurately states *what* was implemented, a Level 4 report explicitly argues *why* specific technologies or patterns were chosen, referencing trade-offs or requirements. At this stage, diagrams are not just syntactically correct but are optimized for readability, avoiding clutter to effectively support the narrative. The text moves beyond a chronological log of actions to a structured explanation of the solution. Achieving Level 5 requires a professional polish that rivals industry whitepapers. The narrative flows seamlessly, anticipating reader questions and synthesizing complex technical details into an accessible format. At this level, visual aids and text are inextricably linked, with the formatting enhancing the hierarchy of information. The rationale demonstrates deep critical thinking, weighing constraints and alternatives persuasively, distinguishing the work as a comprehensive professional artifact rather than a mere homework submission.

Proficiency Levels