Connecting to and Preparing Data23% of the exam
Relationships versus joins in the data model
The logical layer (relationships, or noodles) versus the physical layer (joins and unions), how a relationship keeps tables at their own level of detail while a join merges them up front, and when a join is still the right choice.
Create and manage the data model: Add relationships to a data source; Add joins and unions; Explain when to use a join versus a relationship.
~8 min read · 5 practice questions ·
What the exam expects
- Add a relationship by dragging a second table onto the Data Source page canvas, and edit its matching fields.
- Open a logical table to reach the physical layer and add joins or unions there.
- Explain why a relationship keeps each table's level of detail while a join can duplicate or drop values.
- Choose a join instead of a relationship when the merged, filtered or duplicated single table is what you want.
- Relate tables whose key fields have different data types by using a calculated field.
- Key terms: logical layer (the Relationships canvas you see first); physical layer (the Join/Union canvas inside a logical table); relationship (a "noodle" that names matching fields but doesn't merge tables); join (merges physical tables into one table using a join clause and join type); level of detail (the row grain a table or view aggregates at).
Key facts
Default canvas
- Detail
- Logical layer (Relationships canvas)
Reach the physical layer
- Detail
- Double-click a logical table, or right-click > Open
Second table on the logical canvas
- Detail
- Forms a relationship (noodle), not a join
Default relationship settings
- Detail
- Many-to-many, Some records match
Relationship join type
- Detail
- Chosen automatically from the fields in the view
Matching fields
- Detail
- Same data type in the underlying database; a Data Source page type change doesn't count
Calculated fields
- Detail
- Relationships can be based on them
Not allowed in relationships
- Detail
- Geographic fields; circular relationships
Join effect
- Detail
- Merges tables before analysis; can duplicate, filter or add NULL rows
- Why can a dimension-only view of related tables hide unmatched values?
- When is a join the better choice than a relationship?
- How do you reach the canvas where joins and unions are created?
- Why don't related tables need FIXED expressions to deduplicate measures?
- What relationship settings does Tableau use when no key constraints are detected?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the Tableau Desktop Foundations pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 26 lessons of the Tableau Desktop Foundations guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Practice exams purchase adds inline bank questions and topic practice (5 questions for this lesson)
- An official source cited for every claim
- An AI tutor that answers from the guide
- Choose guide-only, practice-only, or save $4 with both