3
Report: The Spatial Analysis of Census Data Author(s): John Silk Source: Area, Vol. 14, No. 4 (1982), pp. 275-276 Published by: The Royal Geographical Society (with the Institute of British Geographers) Stable URL: http://www.jstor.org/stable/20001838 . Accessed: 16/06/2014 09:05 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . The Royal Geographical Society (with the Institute of British Geographers) is collaborating with JSTOR to digitize, preserve and extend access to Area. http://www.jstor.org This content downloaded from 194.29.185.145 on Mon, 16 Jun 2014 09:05:25 AM All use subject to JSTOR Terms and Conditions

Report: The Spatial Analysis of Census Data

Embed Size (px)

Citation preview

Report: The Spatial Analysis of Census DataAuthor(s): John SilkSource: Area, Vol. 14, No. 4 (1982), pp. 275-276Published by: The Royal Geographical Society (with the Institute of British Geographers)Stable URL: http://www.jstor.org/stable/20001838 .

Accessed: 16/06/2014 09:05

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The Royal Geographical Society (with the Institute of British Geographers) is collaborating with JSTOR todigitize, preserve and extend access to Area.

http://www.jstor.org

This content downloaded from 194.29.185.145 on Mon, 16 Jun 2014 09:05:25 AMAll use subject to JSTOR Terms and Conditions

The spatial analysis of census data

Report of a meeting organised by the Population and Quantitative Study Groups in collaboration with the RoyalStatisticalSociety, Birmingham Group, on 15-16 April 1982 atthe University ofBirmingham.

The workshop attracted an audience of about fifty- including geographers, statisticians, planners and other local authority representatives, together with at least one marketing consultant, to discuss the forms in which census data (particularly for 1981) were available and the allied prob lems of analysing data at a variety of spatial scales and for a multiplicity of areal units.

In the first session Chris Denham (OPCS) emphasised that considerable efforts had been made to respond to user demands before the 1981 census. Both he and a knowledgeable user (David

Rhind, Birkbeck) agreed that, compared with 1971, results were being made available far more speedily and in more sophisticated form. A 'change file' allowing comparisons at two spatial scales 1971-1981 was also being produced. It appeared that academics would be well-provided for as the SSRC had agreed to purchase a software package initially developed for local authority use and negotiations were being finalised to purchase all the 1981 census data. Keith Francis (Hampshire County Planning Office) commented that local authorities should be in a good position to produce basic information but expressed concern that more difficult tasks, such as projection or problem area definition, were fraught with problems (identified in technical papers given at the workshop) and that solutions were urgently required because planners would be under pressure to carry out analyses in the near future. In discussion, a number of people asked why it was still necessary to make data available by, often unsuitable, enumeration districts (EDs) and why individual data should be confidential in Britain when this was not the case in Italy. Few were very happy with the reply, particularly on the latter point that an undertaking of confi dentiality had been given although it was not a problem of parliamentary legislation.

Helen Mounsey (Edinburgh) described in detail a procedure for calculating populations on a constant areal base for the period 1901-79, based on the census, in order to produce a carto graphic film of population change. Likewise, Peter Norris (Durham) dealt with the problem of noncomparable units at successive censuses. He argued that the 1971 and 1981 censuses were far less comparable than OPCS claimed in terms of ED boundaries and used examples from

County Durham to support his view that census tracts intermediate in size between wards and EDs should be drawn up by OPCS and interested parties. In a later session, Graeme Thomsom (Bath) reviewed a number of automatic contouring procedures and proposed a technique based on a superior interpolation procedure allied to a method for piecing together the contours of a set of quadratic functions fitted to different areas of the map.

In a session devoted to in-depth consideration of the ecological fallacy, Kelvyn Jones (Portsmouth Polytechnic) argued that results obtained at differing spatial scales could well be providing us with information of genuine value rather than simply being a nuisance. He also argued that in certain cases changes attributed to modifiable units could be explained in terms of model mis-specification. Stan Openshaw (Newcastle) provided yet more empirical studies, some based on Italian census data which could be disaggregated to the level of the individual, and which demonstrated the magnitude of the ecological fallacy problem and the sensitivity of different analytical methods to it. In discussion, it was pointed out that geographers might benefit from standard statistical practice in which no one would dream of simply comparing patterns of zone or group averages with individual observations across all groups. After dinner, Roger

Mead (Reading) reviewed a variety of techniques for detecting and describing spatial patterns in contiguous quadrat data, pointing out that non-randomness could be detected at various scales in hierarchical analysis and that it was possible to deal with odd-shaped and odd-sized areas.

In the final session, Michael Coombes (Newcastle) outlined a functional classification of re gions in England and Wales developed to cope with the diverse elements in the urban system (including rural areas). The classification is to be meshed with census data so that the 1971-81 change file can be related to it. Phil Rees (Leeds) showed graphically how variations in defi

275

This content downloaded from 194.29.185.145 on Mon, 16 Jun 2014 09:05:25 AMAll use subject to JSTOR Terms and Conditions

276 The spatial analysis of census data

nitions, including those of regions, used in countries contributing to IIASA's Comparative Migration and Settlement project influenced the results obtained, and stressed the difficulties

in matching model and data concepts when estimating migration and population changes. Overall, the impression emerged that the census carthorse, although responding to demands

for change, found it difficult to fundamentally alter direction. There also appeared to be difficulty in marrying up appropriate analytical techniques with data providers and users-it seems, there fore, that further meetings involving all three groups would be very welcome.

John Silk University of Reading

Liverpool Papers in Human Geography The following have been added to this series:

7. Industrial estates in Brazil: problems and definition, by J. P. Dickenson. 8. Regional development and employment generation in Western Kenya, by W. T. S. Gould. 9. A comparative study of the spread of agricultural disturbances of 1816, 1822 and 1830,

by Andrew Charlesworth. 10. Population trends in London: 1841-1911, by Anthony Jones. 11. Family household structures in Savoy, 1561-1968, by Anne M. Jones and D. J. Siddle.

These working papers are now available at L1 50 each (including postage) and may be pur chased from the Map Curator, Department of Geography, The University, Liverpool L69 3BX.

Reading Geographical Papers The following have been added to this series: 79. The history of planning methodology: a framework for the assessment of Anglo-American

theory and practice, by M. Breheny and P. W. J. Batey. 80. Public provision and urban politics: papers from the IBG Annual Conference January 1982,

edited by Andrew Kirby and Stephen Pinch. 81. Location in a system of global extent: a social model of settlement, by James E. Vance

Jr. 82. The effect of geographical scale on manufacturing location decisions, by Peter Mcloughlin.

These papers are now available at ? 1 50 each from The Secretary, Department of Geography, University of Reading, Whiteknights, Reading RG6 2AB.

Retirement of T. H. Elkins Tom Elkins has been granted early retirement as Professor of Geography in the University of Sussex. He will, however, continue to have an honorary appointment in the University at Profes sorial level, and will continue to be involved in teaching and research in Geography, especially in the School of European Studies. He has been succeeded as Chairman of Geography by leuan

Griffiths, Reader in Geography.

This content downloaded from 194.29.185.145 on Mon, 16 Jun 2014 09:05:25 AMAll use subject to JSTOR Terms and Conditions