Data integrity

Integrity refers to data that are accurate, reliable and consistent.

In a database;

• data is committed to the database if correct, otherwise rejected. This is the assurance that the data you have in the database is correct.
• the DBMS must provide facilities for the user or programmer to define transaction boundaries - the logical beginning and end to a transaction.

Example:

• On entering a customers data it is found that essential data is missing. The DBMS will reject the partially completed record.
• Field entries are checked with original data source
• Use simple numbers to check formula are correct
• Totals are checked such as the number of records entered equals the number of records to be entered
• Data is checked (validated);
• for the correct type checking (numeric/text)
• whether mandatory /optional data
• whether within a specified range (M/F, mth =1-12)
• whether a member of list
• for a pattern (eg. Check digit)
• whether field dependencies exist.
• whether entered in the correct sequence
• for referential integrity - you cannot delete an employee if there are still timesheet entries.

[Rev 27/5/98] 30/4/97 © 1997-98 V/2-Com (Verhaart), P O Box 8415, Havelock North, New Zealand.