Data integrity in general is the measure of how well the data is maintained within the data resource after it has been created or captured. Data Structure Integrity is a subset of this data integrity that guides data relations. To say that a data has high integrity means that data has functioned in the way it was intended to be.
A data structure integrity rule defines the specification of a data cardinality for a data relation in a circumstance where there are no exception that apply. This rule make the data structure a lot easier to understand.
A conditional data structure integrity rule is slightly different in that this only applies for a data relations when there are conditions or exceptions that apply. This data structure integrity rule also shows that there is an option for coded data values which are typically difficult to show on an entity relation diagram.
To better illustrate the use and benefits of having achieved data structure integrity, let us say that we have two tables within the database. The first table, let’s call it "Persons" table contains data in a list of names. Let us have another table and call it "PhoneNumbers".
In the real world, people may have one, more than one or no telephone at all. In database terms, the two tables would the have three kinds of relationships: one to one; one to zero; and one to many relationships. This literally means that every person with the "Persons" table may have one, zero or many phones number within the "PhoneNumbers" table.
It is worthy to note that every phone number within the "PhoneNumbers" table has one and only person within the "Persons". By not applying the data structure integrity rule, the two tables may result in data being mixed up in the complicated relationships. It could result in data redundancy which can significantly slow down the whole system and result in data inconsistencies.
Data structure integrity can be achieved by designing a database which is consistent, logical and stable. This can be done by including declared constraints in the overall design of the database which is often referred to as the local schema.
Database normalization is also one of the biggest factors that can help an implementation achieve data structure integrity. In database normalization, the tables are made sure there is no redundant data and some desirable properties of the tables are selected from a logically equivalent set of alternatives.
Referential integrity rules in relational databases make sure that data is always valid. There can be any referential integrity rules depending on the needs or requirements of the data model based on the business rules. The only thing to take careful notice of is that in the end, the data structure integrity is always maintained in the effort of meeting all the requirements.
The use of very precise rules for data integrity greatly solves a lot of problems pertaining to data quality which are very prominent in a lot of data warehousing implementations in both public and private sector organizations.
Precise rules for data integrity reduce the impact of bad information and allow many organizations to make the most use of their limited resources to more value added undertakings. They also help many business organizations in their quest for identifying accountability in the area of data warehouse management which is as equally important as other areas like human resources, finance and sales.
With careful planning – from the data architecture to the physical implementation – data structure integrity can surely be achieved to give the company quality data as basis for sound decisions.