Flat file databases

A flat file database consists of one table. As a result, a large database will most likely have data which is unnecessarily repeated several times in the same table.

The table below holds data on teachers in a school. It also contains information about the department they work in.

ID	Forename	Surname	Department	Department ID	Phone number
1	Colin	Arthur	ICT	001	300
2	Laura	Brown	ICT	001	300
3	Stephen	MacLeod	ICT	001	300
4	Scott	Sinclair	English	002	301
5	Michelle	Wie	English	002	301
6	Ross	Dyett	PE	003	302
7	Ian	Anderson	PE	003	302
8	Betty	Flood	Geography	004	303

ID	1
Forename	Colin
Surname	Arthur
Department	ICT
Department ID	001
Phone number	300

ID	2
Forename	Laura
Surname	Brown
Department	ICT
Department ID	001
Phone number	300

ID	3
Forename	Stephen
Surname	MacLeod
Department	ICT
Department ID	001
Phone number	300

ID	4
Forename	Scott
Surname	Sinclair
Department	English
Department ID	002
Phone number	301

ID	5
Forename	Michelle
Surname	Wie
Department	English
Department ID	002
Phone number	301

ID	6
Forename	Ross
Surname	Dyett
Department	PE
Department ID	003
Phone number	302

ID	7
Forename	Ian
Surname	Anderson
Department	PE
Department ID	003
Phone number	302

ID	8
Forename	Betty
Surname	Flood
Department	Geography
Department ID	004
Phone number	303

In this flat file database, the data on each member of staff is held together with all data for each department.

This is unnecessary as some information is stored more than once (eg the same phone number). This also results in an increased file size for the database. Using flat file databases can lead to three very specific problems:

insert anomaly
delete anomaly
update anomaly

Anomalies

��˿��

Introducing databasesFlat file databases