When we are designing the database we refer to entities and attributes. An entity is a collection of information about a particular object, person or thing. When we create a database an entity will usually become a table and attributes will become fields.

But before we can create the database we need to think about the data that we are going to store, this will allow us to create a data dictionary. This will allow us to design the characteristics that each attribute has so that we can make them into tables and fields.

A data dictionary will look at the following characteristics.

• Entity Name

• Attribute Type and Size

• Key

• Validation

A worked Data Dictionary Example can be seen here but the structure is shown below.

Entity Name

When we design the entity we need to give it a descriptive name. Such as Dog, Customer, Item. It will usually be a singular term as opposed to plural.

When storing text you are required to tell the database how many characters of text you are going to store. This amount of space will usually be reserved for each record, so you want to store enough but not an excessive amount.

As discussed earlier each entity will require a primary key and if it is part of a relationship it may also require a foreign key.

Ever noticed that when you are filling an online form that

• Certain pieces of information can’t be left blank?

• Password has to be a certain length?

• Month has to be in a certain range?

• Or you can only choose your title/gender/country from a list?

These are all forms of validation. This is when we are trying to ensure that data is sensible and in the format that we expect.

Types of Validation

The forms of validation we will look at are:

1. Presence Check

2. Restricted Choice

3. Length Check

4. Range Check