Information Management Consultants
For many museum documentation projects we need to estimate in advance the amount of staff time that will be needed, both for inputting existing data into a computerised collections management system and for creating new catalogue records for previously undocumented objects. These estimates may be used to justify the cost of a project, to apply for funds, to allocate staff resources and to decide on priorities. There seems, though, to be a dearth of reliable data on which to base such estimates.
In January 2001 I therefore did a short survey asking for data from projects that have already been completed which could be used to compile guidelines on which to base future estimates. Clearly there are many variables and any collection presents its own problems, but I think that even ranges of time can be valuable. The survey took the form of messages to the email lists of the [UK] Museum Computer Group, the Museum Computer Network (MCN) and to the general list MUSEUM-L
To make figures comparable, I asked respondents to give them in the following categories:.
When people quoted a number of objects per hour (or a number of hours per object) I have assumed an effective working day of 6 hours in converting the figures to objects per day. Figures given in square brackets have been interpolated from maximum and minimum figures supplied.
I asked respondents to say (1) what type of objects were involved, (2) whether authority files for names and subjects were pre-existing or were created during the project, (3) whether the staff concerned were curatorial (able to assess and modify the correctness of the data) or clerical (primarily doing straight transcription of existing records). Any other background information that helps to explain the rates achieved was also requested. Various comments were received on these matters, but not in a sufficiently structured way to allow systematic analysis. The division between "Inventory level" and "Full cataloguing" is not clear-cut, and it was often unclear in which category some figures should be included. All the data should therefore be taken as broadly indicative of rates achieved, and any realistic projections should be based on trials with the actual data and procedures applicable to a specific project.
As some museums did not wish to be identified, I have not given any names except for four entries taken from a published source, but the figures come from 21 different national, local authority, university and independent museums, mostly in the UK but with some in the USA and Australia. There does not seem to be any obvious correlation between type of museum and cataloguing rates, but the data is insufficient in quantity, and insufficiently controlled for other variables, to draw any conclusions of this type.
I am most grateful to all the museums that responded, and would welcome any further data for addition to these tables.
|INPUT OF EXISTING RECORDS||Objects per person-day|
|Inventory level||Full cataloguing|
|[Museum A] Four fields only: accession number, simple name, location and a general notes/description field.||700||1,000||1,200|
|[Museum C] Entering data from published catalogue for furniture, metalwork, ceramics and jewellery. Includes adding to name authority file for donors and artists.||13|
|[Museum E] Paintings, drawings and prints.||30|
|[Museum E] Applied arts.||15||27||60|
|[Museum F] Art, numismatics, scientific instruments etc., but especially geology.||200||||300||10||25||40|
|[Museum G] Maritime items, USA. About 12 to 16 fields plus creating very brief source & maker entries||88||127||166|
|[National Museums of Scotland - Morrison, 1996, cited in Stewart and Lonie, 1996.] Average rate over a six-year project, 1988-1994. Records varied from very brief to very full, reflected in the maximum and minimum rates shown.||1||66||1,500|
|[Museum K] Waterways objects - fullness not specified.||50|
|[Museum L] Archaeology.||40||50||70|
|[Museum L] Social history.||71|
|[Museum N] Natural history.||168|
|[Museum N] Mixed collection excluding natural history.||14||20||27|
|[Scotland Street School Museum] A museum of school education. Stewart and Lonie, 1996||108||162|
|[Strathyer - cited in Stewart and Lonie, 1996]||180|
|[Figure in Willoughby promotional material, 1991 - cited in Stewart and Lonie, 1996]||45|
|[Museum V] Brief catalogue records, input in 1986-87 by someone who knew the collections well.||250|
|NEW CATALOGUING||Objects per person-day|
|Inventory level||Full cataloguing|
|[Museum B] Waterways objects - Brief descriptions existed for most objects; includes recording more detailed information, measuring, marking and labelling.||9|
|[Museum C] Archival material, newspaper cuttings, letters, articles.||14|
|[Museum D] City history museum, USA. Includes describing, measuring, marking, photographing and entering the record on computer.||2|
|[Museum F] Art, numismatics, scientific instruments etc., but especially geology.||30||15|
|[Museum J] Mixed collection, including transport, furniture, jewellery. Filling out A4 data sheets, not computer entry.||1||6||72|
|[Museum H] Small objects relating to transport. Includes accessioning, cataloguing, taking digital photos of all objects, downloading them onto server and attaching images to the catalogue to create thumbnails for all new records.||10|
|[Museum K] Waterways objects - fullness not specified. Includes marking, labelling and photographing.||20|
|[Museum L] Social history.||12||21||50|
|[Museum O] Military items. Includes checking and adding to name and subject authority files. Test run of 55 items only.||30|
|[Museum O] Military items. Entry from existing accession records; not full cataloguing, but includes checking and adding to name and subject authority files. Data from actual project: 5 staff, 7000 items.||30||37||48|
|[Museum Q] Industrial and working life collections: retrospective documentation||10||20|
|[Museum S] Art works: adding subject index terms to existing inventory level records||125||150||200|
|[Museum T] Mounted photographs. Catalogue records have 11 fields; artist and place name authority files created as work progresses. Figures estimated from combined data supplied.||150||||180||75||||90|
Factors mentioned that affected rates included the following:
For the processing of archives, including physical organisation and listing, times vary widely depending on the initial condition of the material, but as a rule of thumb an average of 1 cubic foot per day for straightforward modern material seems to be generally accepted. , Older and disorganised material, and photographs, can take much longer. For a discussion of this, and other references, see the archives of the ARCHIVES discussion list for December 2001, week 2, under the heading "Rough processing statistics". There was a further discussion on the same list in February 2005, week 4 under the heading "Cost to process a collection", and again in February 2007 under the heading Estimate of time spent processing per linear foot?. The New York State Archives has published some guidelines on processing rates, including a table based on Karen T. and Thomas E. Lynch's "Rates of Processing Manuscripts and Archives," (The Midwestern Archivist, 7:1, 1982).
Archivists normally record groups of material, at various levels of aggregation, rather than individual items, so although a box of volume 1 cubic foot may contain around 2000 sheets of paper, each of these would not normally be separately catalogued. For the recording of individual archival items, I proposed the following guideline, which was discussed on the ARCHIVES mailing list; I have adjusted the times suggested after receiving feedback from archivists that my original estimates were too low: Clearly these figures are just suggestions; if we had some experimental data, or data from real projects, we could adjust this formula to fiit the data and be able to see the range of variation between projects, which would help us to estimate the uncertainty in projections for future projects. If anyone has any such figures I would be most interested to hear from them.
Estimate the overall time to be allowed per item as (5 + C + D + S) minutes, where:
If we give each of these three factors a rating as follows
- C = Complexity of material : from simple letters with clear handwriting and routine subjects, to complex, illegible, philosophical or technical manuscripts.
- D = Depth of cataloguing or listing: information recorded about each item might range from the name of the writer and the date to a digest of the content with many name and subject index terms.
- S = System used : free text descriptions or structured records. Availability of authority files for name, place and subject access points. Ease of use.
this gives a range from 20 minutes per item for the simplest material to 95 minutes per item for complex material, thoroughly catalogued in a difficult system.
Easy 5 minutes Moderate 15 minutes Difficult 30 minutes
To obtain a file of messages from the AUTOCAT email discussion list on the costs of cataloguing library materials, send a message with the content GET CATALOG COST to LISTSERV@listserv.buffalo.edu. These messages date from 1993 to 2003 and are mainly in the context of libraries in the United States, often obtaining catalogue records from a central utility, but the methods of calculation may be of interest as well as the results.
Beneficial shocks : the place of processing-cost analysis in archival administration. - American Archivist, vol. 58, no. 1, Winter 1995.
Cave, Alexandra, 2003-01-13
Logjam project. - Announcement of a project to survey uncatalogued archives by the [UK] North West Regional Archives Council and its partners. Data on previous projects is being collected to devise a formula for estimating the resources required.
An experimental approach to specimen databasing. - Plant Cuttings : the Department of Botany on-line newsletter - No.9, January 2003. - London : Natural History Museum.Announcement of an experiment to measure rates of data entry for natural history specimens, with comparison of different methods. Fran Kern has now left the Natural History Museum but the results of the experiments will be made available by the Botany Department�?s Collections�? Leader when they are complete.
Lynch, Karen T. and Lynch, Thoms E
Rates of processing manuscripts and archives. - Midwestern Archivist, vol. 7, no.1, 1982.
O' gudes an' gear an' a' my graith / Ian O. Morrison. - Scottish Museum News, vol.12, no.1, Spring 1996. - [Cited in Stewart and Lonie, 1996]
Stark, Bruce P.
A guide for processing manuscript collections. - [Connecticut : B.P. Stark], 2001. - 94, 6 p. ; 28 cm. - $12.
Stewart and Lonie, 1996
Scotland Street School Museum inventory project / Dorothy Stewart and Nuala Lonie. - Scottish Museum News, vol.12, no.2, Autumn 1996, p17-19.
Museum handbook / National Park Service. - Washington, D.C. : U.S. Dept. of Interior, 1990 - . Looseleaf printed and online (pdf) versions.Part 2: Museum records, Appendix B: Cataloging costs (170K pdf file) gives guidance on ways of estimating cataloguing costs and has tables of actual costs found in cataloguing projects for different types of materials in several states of the USA for the years 1997-1999.
This document is at
Revised 2010-02-16 17:10
Comments and feedback on content or presentation are welcome and should be sent to Leonard Will
Copyright © Willpower Information, 2001-2010.
Return to Willpower Information home page