Any information stored in a computer is referred to a data. This data can then, with a specific instruction, be processed and searched. The Data is organised into its parent: individual records which is made up of elements called fields. A field consists of one or more data elements. Each element contains words which is composed of digits of characters or numeric. When we group records together we produce a file; or in another word, a file will contain a number of records. Files are then organised into a database. In a library, a file might  be a catalogue. In this case, the record then becomes each individual entry in the catalogue, and the fields are the elements within entry, eg.: author, title, subject, call number, imprint, ISBN, etc.

The component parts of a record should be planned carefully before designing a file since each record represents a document; thus the record content can be manipulated and retrieved. The powerful of a record retrieved will depend on the characteristic of the elements, eg. : field and subfield.


To discuss record structure, we might not separate the elements which contribute the exixtence of a record, e.g. field. A field is a logical unit of information within a record. Each field contains a specific type of information or an attribute. The characteristics of a field is stated in the following. First, each field should be identified by giving a name. A field name will distinguish from one field to another. It will be helpful during the data input.

Second, the type of data to be included in each field must be decided whether it is alphabetic, numeric, or special character data. Data type specification may be used by search program to check accuracy of the data input. For example, year field is defined as a numeric field, thus any alphabetic input in this field, an error message will be issued. And whenever searcher retrieve the year field, numeric information will come up.

Next, a field may be fixed or variable in length; eg.: a field for date would be fixed if the information were always represented as a series of six digits for day, month, and year (081159). A field may be variable if the field permitted values such as Winter-1993 or 08 November 1993.

Fourth, a field may be a singly occurring or multiply occurring. For example, a document is likely have only one information, eg. ISBN, so the ISBN field would be singly occurring. However, a document might have more than subject content, so the subject field would be multiply occurring. Similarly name for multiply occurring is repeatable, for example, a document has author 1, author 2, author 3, so this field can be used as many names as are required for the document. Fifth, some fields may have subfields, for example, publication field: place, publisher, and date; use of subfield code to break down subfield.

Similarly with fields, records may be fixed or variable in length. If all records for a particular document have the same number of fields and the data stored in the fields is always less or equal to maximum determined length, such as telephone directory, eg.: name, address, phone number;  then the records will have fixed length. Records will be variable in length if any of the followings is true :
– some fields may occur more times in one record than another
– some fields may be absent from a record
– the data in the fields have no maximum length
Other considerations in designing record structure are ambiguity, that it must be uniquely identify fields and what they are usually done by subfields. In semantic considerations that record model should be expressiveness. The expressiveness of a record structure refers to the ability of parties to a communication to convey ideas using the structure. Other requirements for well designed record structure are :
– accommodating all data required for the system
– providing the data reduction necessary for data manipulation and processing
– presenting the data in the form required for computer processing
– flexibility, capable of adapting to changed requirements
– simplicity, does not place unnecessary demands on data preparation


MARC (Machine Readable Cataloguing) format was designed and developed by Library of Congress over three decades ago for library application. This was to meet the demand of the increasing library materials to be stored and providing more accessible for them. The MARC format becomes universally, though every country develops it in individual version for locally use, such as UKMARC in British, USMARC in USA, and AUSMARC in Australia. The main advantages use of MARC format is that libraries could centralise the records from which they could share acquisition, cataloguing, and searching from the remote control. In order to meet a such database, a standardised method of organising data in the bibliographic record and it is what the MARC record structure attempts to do. Obviously, the bibliographic information entered is still identical between the manual and computerised systems; the cataloguer write in a work sheet before input in the system.

Each catalogue record consists of four elements :
1. The leader or the label for the record, it provides information about the length, bibliographic level and type of material being catalogued. The leader is fixed in length.
2. The directory for the record, is a series of twelve character numeric entries, each of which contains the tag, the length of the field and the starting position of the relative to the first character. This information is about addresses of data in the main body of the entry and facilitates searching, reducing positional ambiguity.
3. Control fields contain alpha-numeric data elements, e.g. tag 001 – 009.
001 – record control, in AUSMARC is ISBN
002 – relates only to records with subrecords
008 – information codes, such as data entered on file, type of publication code, place of publication code, illustration code, form code, language and cataloguing code.
4. Data fields contain variable length alpha-numeric data, and contain all the bibliographic information. Each data field is identified by a tag ( a three-character numeric code : from 010 to 999), for example : 245 for title, 300 for description, 500 for general note, 504 for bibliographic note and 650 for subject heading.

After the tag, each data field is introduced by indicator (two numeric characters) which provide descriptive information about the filed. Each data field might be divided into subfields which is introduced by a delimiter (@) followed by a lower alphabetic character indicating what type of data the subfield contains, such as in tag 245 : @athe main title, @bsubtitle and @cstatement responsibility; tag 260 : @aplace, @bpublisher and @cdate. The following is an example of record entry using USMARC format :

Rec Status: n    Legend: am$$    Encoding: $        Descripti: a    Link:$
File Date:200793        DType: s        Date 1:1993    Date 2:$$$$
Country: mnu    Illus: $$$$    Intell: $        Repro:$    Content:$$$$
Govt: $    Confer: 0    Fest: 0            Index: 0    ME/Body:1
Fiction: 0    Biography:$    Language: eng    Mod:$         Cat Scr: d
Record ID :CRLD93-B33509                Transac:18200793-008876-1

020    $$    @a093699137
040    $$    @aCU@cCU
100    10    @aBlixrud, Julia C.,@c1954-
245    12    @a A manual of AACR2 examples tagged and coded using the MARC format /@cby Julia C. Blixrud and Edward Swason.
260    0$    @alake Crystal, Minn. :@bSoldier Creek Press,@c1982.
300    $$    @aiii, 166 p. ;@c28 cm.

500    $$    @a”An adjunct to the series of manuals illustrating cataloging using the Anglo-American cataloging rules, second edition, prepared by the  Minnesota AACR2 Trainers.”
650    $0    @aCataloguing.
700    10    @aSwason, Edward,@d1941-
710    20    @aMinnesota AACR2 Trainers.

References :

1. Crawford, Walt. MARC for library use: understanding integrated USMARC, 2nd ed. Boston, G.K. Hall & Co., 1989.
2. Goodacre, Christine. “Record structure”. Lecture. Library and Information Studies, University of Tasmania, 25 May 1993.
3. Lancaster, F.W. and E.G. Fayen. Information retrieval on-line. A Wiley-Backer & Hayes series book. Los Angeles, California: Melville Publishing, 1973.
4. Meadow, Charles T. The analysis of information system, 2nd ed. A Wiley-Backer & Hayes series book. Los Angeles, California: Melville Publishing, 1973.
5. Pao, M. Concepts of information retrieval. Littleton, Colorado: Libraries Unlimited, 1989.

Leave a Reply

Your email address will not be published. Required fields are marked *


Skip to toolbar