Full vs Partial Functional Dependencies
You'll learn how to distinguish between functional dependencies that require all their determinant attributes versus those that rely on only a subset.
What You Already Know
You've learned that a functional dependency (FD) means one set of attributes determines another. For example, StudentID → StudentName means knowing the ID uniquely determines the name. But what if the determinant has multiple attributes?
The Core Distinction
A full functional dependency exists when the dependent attribute requires every attribute in the determinant to be uniquely determined. Remove any part of the determinant, and the dependency breaks.
A partial functional dependency occurs when the dependent attribute can be determined by only a subset of the determinant's attributes—you don't actually need all of them.
Concrete Example
Imagine a table tracking student course enrollments:
-
(StudentID, CourseID) → Grade— Full dependency. You need both the student and the course to know the grade. Knowing just StudentID or just CourseID isn't enough. -
(StudentID, CourseID) → StudentName— Partial dependency. StudentName depends only on StudentID; you don't need CourseID at all. The dependency works with just part of the composite key.
Why This Matters
Partial dependencies signal redundancy and update anomalies. In the example above, StudentName gets repeated for every course a student takes. Full dependencies are "tighter" and more appropriate for well-normalized tables.
Quick Test
Ask yourself: "Can I remove an attribute from the determinant and still determine the dependent?" If yes, it's partial. If no, it's full.
Key Takeaway: Full functional dependencies require the entire determinant; partial dependencies work with just a subset, indicating potential normalization issues.