What are some common mistakes and best practices when drawing ERD?

2 Mistake 2: Missing or wrong cardinality

Another common mistake when drawing ERDs is missing or wrong cardinality for the relationships between entities. Cardinality indicates how many instances of one entity can be associated with one or more instances of another entity. For example, one-to-one, one-to-many, or many-to-many. If you omit or misrepresent the cardinality of a relationship, you will lose important information about the data model and its constraints. To avoid this mistake, you should use the appropriate symbols and notation to show the cardinality of each relationship, such as crow's feet, min-max, or Chen's notation.

4 Best practice 1: Use subtypes and supertypes

One of the best practices when drawing ERDs is using subtypes and supertypes to model hierarchical or inheritance relationships between entities. Subtypes and supertypes are entities that share some common attributes and relationships, but also have some specific attributes and relationships that differentiate them. For example, you can have a supertype entity called Employee, and two subtype entities called Manager and Technician, that inherit some attributes and relationships from Employee, but also have some unique attributes and relationships of their own. To apply this best practice, you should use the ISA symbol and notation to show the subtype and supertype relationship, and specify the constraints and rules that govern it.

5 Best practice 2: Use associative entities

Another best practice when drawing ERDs is using associative entities to model many-to-many relationships between entities. Associative entities are entities that represent the association or link between two or more entities, and that can have their own attributes and relationships. For example, you can have an associative entity called Order, that links the entities Customer and Product, and that has attributes such as date, quantity, and price. To apply this best practice, you should use a diamond-shaped symbol and notation to show the associative entity, and include the primary keys of the related entities as foreign keys in the associative entity.

6 Best practice 3: Use normalization

A third best practice when drawing ERDs is using normalization to optimize your data model and reduce redundancy, inconsistency, and anomalies. Normalization is a process of applying rules and principles to organize your entities and attributes into different levels or forms of normal forms. For example, the first normal form (1NF) requires that each attribute has a single value and that there are no repeating groups of attributes. The second normal form (2NF) requires that each attribute depends on the whole primary key and not on part of it. The third normal form (3NF) requires that each attribute depends only on the primary key and not on any other attribute. To apply this best practice, you should check your data model against the normal forms and make the necessary changes to achieve the highest level of normalization possible.