data integrity
noun
Data integrity means that data remains correct, complete and internally consistent, and changes only in permitted ways.
A unique customer identifier helps maintain data integrity by preventing the same identifier from being stored twice.
Data integrity means being able to trust that data will not change unexpectedly, disappear or contradict other data. It covers both errors during input and processing and unauthorized changes. A system with good data integrity does not store an order without a customer or allow an invoice amount to change unnoticed, for example.
How do you maintain data integrity?
The rules in the Glossary · In briefdata modelA data model describes which data a system records, how that data is related and which rules it must satisfy.Read more provide the starting point. A Glossary · In briefdatabaseA database is a structured collection of data that software can store, retrieve and modify. A database management system controls access to that data.Read more can enforce these rules with required fields, unique values and references between tables. A foreign key can prevent an order from referring to a customer who does not exist.
Checks are also needed outside the database. Input validation rejects invalid values, Glossary · In briefauthorizationAuthorization is the process of determining and enforcing which data and functions a user, device or application may access.Read more determines who may change data, and transactions ensure a series of changes is either completed in full or rolled back. Backups, checksums and an Glossary · In briefaudit trailAn audit trail is a chronological record of actions and changes in a system. It shows who or what performed an action, when it happened and what changed.Read more help detect and recover from data loss or unwanted changes.
Data integrity and data quality
Data integrity and data quality overlap, but emphasize different aspects. Integrity concerns the validity and consistency of data within agreed rules. Data quality considers more broadly whether that data is useful for a particular purpose.
An address can be technically valid and correctly stored in the database yet be outdated because someone has moved. The integrity rules have not been broken, but data quality is insufficient. Reliable reporting and processes therefore need both.