Data Modeling Basics (Part 2)

Related Blog: Data Modeling Basics (Part 1)

This is the second part in a 3-part series on the basics of data modeling.
In the first part of this blog series, we talked about data modeling and we
tried to define a data model. In part 2, we will start to look into the three
types of data models, the different notations used in data modeling and then
jump into Conceptual Data Model (CDM).
Data Models
There are three basic styles of data models - these are Conceptual Data
Model (CDM), Logical Data Model (LDM) and Physical Data Model (PDM).
A Conceptual Data Model (CDM) consists of the basic business entities
and the relationships that exist between the entities. It is usually used to
explore and discuss domain concepts with the stakeholders to gain better
understanding of a certain domain (e.g. Banking, Manufacturing, Finance,
etc.).
A Logical Data Model (LDM) is the next data model in the progression of
data models. In addition to the basic business entities and the relationships
between these entities, the LDM now has the attributes and the primary key
defined. A primary key is just a way to identify the unique record in a certain
entity.
A Physical Data Model (PDM) is the data model representation in an
actual database management system (DBMS). The entities, relationships,
attributes, primary keys are now represented as tables, foreign-key
relationships, columns and primary keys in the database management
system. The data types e.g. CHAR (1), VARCHAR (10) or INTEGER is defined
for the data values to be stored in each column or field.
Data Modeling Notations
There are several data modeling notations. In the Agile Data web site (link),
several are listed. These notations include data modeling notation from
Information Engineering (which includes the popular "crow's feet notation"),
the Barker notation, IDEF1X notation, the UML notation and the Entity
Relationship Diagram (ERD) notation.
All throughout my blog, I will use the "crow's feet notation" because of my
familiarity with using this notation. You can choose whatever notation you
prefer when you do data modeling.
Here is a link to a web article from Lucidchart on Entity Relationship Diagram
or ERD. It is a very informative web page on ERD.
Conceptual Data Model (CDM)
To create a conceptual data model, we first identify the entity types.
An entity type is a thing that exists either physically or logically. An entity
may be a physical object such as a house or a car (physical existence), an
event such as a house sale or a car service, or a concept such as a customer
transaction or order (logical existence as a concept). It is usually a noun.
Below are examples of entity types in a school where typically there is a
[Student], [Course] and [Teacher].

Then, we define the relationships between the entity types.

Relationship captures how entity types are related to one another.
Relationships can be thought of as verbs, linking two or more nouns.

As part defining the relationships between entity types is to define the right
cardinality.
Cardinality is the number of occurrences in one entity which are associated
(or linked) to the number of occurrences in another. There are three degrees
of relationship, known as: one-to-one (1:1), one-to-many (1:M) and many-to-
many (M:M).
An example of a many-to-many relationship is show below.
We now have a Conceptual Data Model. We can now explore and discuss
with the stakeholder or a subject matter expert to validate our model and to
add more entity types or relationships in the domain (which is a School in
this case).
In the the third part of this blog series, we will continue to Logical Data Model
(LDM) and Physical Data Model (PDM). We will also talk about concepts on
normalization.
Next Blog: Data Modeling Basics (Part 3)

Data Modeling Basics (Part 3)

Related Blogs:
Data Modeling Basics (Part 1)
Data Modeling Basics (Part 2)
This is the third part in a 3-part series on the basics of data modeling.
In the first part of this blog series, we talked about data modeling and we
tried to define what is a data model.
In part 2, we discussed the three types of data models i.e. Conceptual Data
Model (CDM), Logical Data Model (LDM) and Physical Data Model including
the different notations used in data modeling. We then jumped into creating
a Conceptual Data Model (CDM).
In part 3, we complete our journey in Data Modeling Basics by continuing on
creating the Logical Data Model and the Physical Data Model.
Logical Data Model (LDM)
In part 2 of this blog series, we created a conceptual data model and we will
jump right off of it.
After we have identified the entity types and defined the relationships
between these entity types, in logical data modeling we add the
attributes of each entity type.
Attributes are various properties of an entity type that characterize the
entity. It is something that describes the entity.
For example (refer to the illustration below), [StudentNumber], [FirstName],
[MiddleNm], [LastNm] and [BirthDate] are all attributes of the [Student]
entity type.

We then choose a candidate key to determine unique records. This will

become our primary key when we go to Physical Data Model.
A candidate key is an attribute or combination of attributes that can
determine uniqueness of a certain record. There can be multiple candidate
keys in an entity type.
Next step is to normalize our logical data model.
Database Normalization, or simply normalization, is the process of
organizing the attributes (columns) and entities (tables) to avoid data
redundancy and improve data integrity by removing insertion anomaly,
update anomaly and deletion anomaly.
Here is a link to site explaining each type of anomaly mentioned above. It
also provides a good explanation of the first normal form, the second normal
form and the third normal and how to achieve each normal form.
Physical Data Model (PDM)
From a logical data model, we now go to the last data model type i.e.
Physical Data Model. This is where the data model transforms to something
that can actually store data like in a database management system (DBMS).
As you can recall, while a Conceptual Data Model is a representation of the
concepts of business entities and relationships, the Physical Data Model is
the representation of the actual tables, columns, data types and foreign key
relationships.
In physical data modeling, we define the data types for each column.
A data type is a particular kind of data item, as defined by the values it can
take, the programming language used, or the operations that can be
performed on it.
Summary of Data Modeling Steps
To summarize, here are the steps in data modeling from Conceptual Data
Model to Logical Data Model and finally to Physical Data Model.
1. Identify the entity types.
2. Define the relationships between the entity types.
Output: Conceptual Data Model
3. Add the attributes of each entity type.
4. Choose a candidate key for each entity type.
5. Normalize the logical data model.
Output: Logical Data Model
6. Define the data types for each column.
Output: Physical Data Model
This is depicted in the table below.

Source: https://www.lucidchart.com/pages/ER-diagram-symbols-and-meaning