Data Model II: Types of Data Modelling – SS2 Data Processing Lesson Note (First Term)
Lesson note at a glance
- Class
- SS2
- Subject
- Data Processing
- Term
- First Term
- Reading time
- 5 min
A data model is the way a database organises its data and the links between the data.
Behavioural objectives
By the end of the lesson, students should be able to:
- name the six types of data modelling;
- describe the flat, hierarchical and network models;
- explain the relational and object-relational models;
- state what a star schema is and what it consists of;
- compare the models using a simple school example.
What are the types of data modelling?
There are six types of data modelling: the flat model, hierarchical model, network model, relational model, object-relational model and star schema.
- Flat model: data sits in a single table.
- Hierarchical model: data is arranged in a tree.
- Network model: records are linked by sets.
- Relational model: data is held in tables linked by keys.
- Object-relational model: a relational database that also supports objects.
- Star schema: fact tables linked to dimension tables.
What is the flat model?
The flat model (or table model) is a single, two-dimensional array of data elements.
- Columns: all members of a given column are assumed to be similar in values.
- Rows: all members of a row are assumed to be related to one another.
- Example: one table holding a school register with name, class, age and address.
- Strength: it is simple to set up and read.
- Weakness: the same data may be repeated many times in the table.
What is the hierarchical model?
The hierarchical model organises data into an upside-down, tree-like structure.
- Single parent link: each record has a single upward link to describe the nesting.
- Sort field: it keeps the records in a particular order in each same-level list.
- Use: hierarchical structures were widely used in the early mainframe database management systems.
- Example: a school has many classes, and each class has many students.
- Weakness: a record cannot easily have more than one parent.

What is the network model?
The network model organises data using two fundamental constructs, called records and sets.
- Records: they contain fields.
- Sets: they define one-to-many relationships between records: one owner, and many members.
- Many links: a record can belong to more than one set, so it can have more than one owner.
- Example: a student record linked to several teacher records, and each teacher linked to many students.
- Weakness: the many links make the structure harder to design and manage.

What is the relational model?
The relational model is a database model based on first-order predicate logic, in which data is stored in tables.
- Core idea: it describes a database as a collection of predicates over a finite set of predicate variables.
- Constraints: these describe the possible values and combinations of values.
- Table (relation): a set of rows and columns holding data about one subject.
- Row (record): one complete entry, such as one student.
- Column (field): one item of information, such as surname.
- Primary key: a field that identifies each row, such as admission number.
- Foreign key: a field that links one table to another table’s primary key.
- Example: a Students table and a Subjects table linked by admission number.
What is the object-relational model?
The object-relational model is similar to the relational model, but it directly supports objects, classes and inheritance.
- Where supported: in the database schemas and in the query language.
- Middle ground: it sits between relational databases and object-oriented databases (OODBMS).
- What it keeps: tables, rows and columns from the relational model.
- What it adds: objects, classes and inheritance.
What is a star schema?
The star schema is the simplest style of data warehouse schema.
- Fact tables: a few tables (possibly only one, justifying the name) that hold the measurable figures, such as amount sold.
- Dimension tables: any number of tables that the fact table references, such as product, date, store and customer.
- Shape: the fact table sits at the centre and the dimension tables surround it like the points of a star.
- Link to snowflake: the star schema is considered an important special case of the snowflake schema.
How do the six data models compare?
The table below shows how each model arranges data.
| Model | How data is arranged | Simple example |
|---|---|---|
| Flat | One two-dimensional table | A school register |
| Hierarchical | Upside-down tree, one parent per record | School, classes, students |
| Network | Records linked by sets, one owner and many members | Students linked to teachers |
| Relational | Tables linked by keys | Students and Subjects tables |
| Object-relational | Relational tables plus objects, classes and inheritance | A relational database with object support |
| Star schema | Fact table linked to dimension tables | Sales fact with product and date tables |
What is the difference between the hierarchical and network models?
The hierarchical model arranges records in a tree where each record has one parent.
The network model links records through sets, so a record can belong to more than one set.
| Point | Hierarchical model | Network model |
|---|---|---|
| Structure | Tree | Records and sets |
| Owners of a record | One parent only | One or more owners |
| Relationship | One-to-many | One-to-many sets that can combine into many-to-many links |
| Complexity | Simpler | More complex |
Evaluation
- List the six types of data modelling.
- Describe how data is arranged in a flat model.
- In the network model, what are records and sets?
- How is the object-relational model similar to the relational model, and what does it add?
- What does a star schema consist of?
Assignment
Explain the hierarchical model and the network model in your own words. State one difference between them.
This term's Data Processing topics
SS2 First Term · 8 topics

