INTRODUCTION BIBLIOGRAPHIC REVIEW

NATIONAL VERSIONS

SCALE CONSTRUCTION

CAMSIS: Social Interaction and Stratification Scale

The CAMSIS project is an internationally comparative assessment of the structures of social interaction and stratification across a number of countries. At its core lies the construction - and dissemination - of occupational scales for each constituent country. The scale values represent an occupational unit's relative position within the national order of social interaction and stratification.

 


Introduction 

 

Links and further information :

  • Download CAMSIS! Quick access page for downloading versions of CAMSIS scales for different countries and different time periods
  • National versions: Further details of versions currently being developed including downloadable data and metadata when available
  • Use of the CAMSIS scores: Guidelines on accessing and using the downloadable CAMSIS scales
  • Bibliographic review: References to literature on the background to the CAMSIS approach and its application
  • Details of construction: A practical account of the principles of constructing CAMSIS scales, including instructions on using the methods in selected statistical packages
  • Work in progress: Selected conference papers and information on other ongoing work
  • Citing CAMSIS scales

 

  • Occupational unit information: An assortment of notes and downloadable files concerned with handling occupational unit information across those countries with CAMSIS scales.
  • GEODE: A related project concerned with distributing occupational information via 'eScience' technologies ('Grid Enabled Occupational Data Environment'). Since 2008, this project is funded by the ESRC as part of the DAMES ('Data Management through e-Social Science') research Node.
  • Is Britain Pulling Apart? Analysis of Generational Change in Social Distance - An ESRC-funded project undertaken 2013-14 by Paul Lambert, Vernon Gayle, Mark Tranmer and Dave Griffiths, which focusses upon measuring patterns and trends in social distance between groups
  • 'Social Networks and Occupational Structure' - An ESRC-funded project undertaken 2010-2012 by Paul Lambert and Dave Griffiths which focusses upon the relationships between social interaction distance analysis (i.e. CAMSIS) approaches to studying occupations, and social network analysis.
  • The 'Social Stratification Research Seminar' - This long-running annual seminar has often been a focal point for researchers involved in making CAMSIS scales, and regularly features papers on, or using, CAMSIS.

Contact details / Register your interest

 


Introduction

Deutsch Español Français Italiano

 

The theoretical basis of CAMSIS (Cambridge Social Interaction and Stratification) scales is the idea that differential association is an essential feature of social stratification arrangements. It is a familiar argument in stratification theory that persons sharing a similar social position, in terms of social class or status group membership, are more likely to interact socially on the basis of equality with members of the same group than with members of other groups. So, acquaintances, friends and marriage partners will all tend to be chosen much more frequently from within the same group than from without. The usual approach, though, is to define a structure composed of a set of classes or status groups and then to investigate social interaction between them.

The CAMSIS approach reverses this, using patterns of interaction to determine the nature of the structure. The crucial point is that differential association can be seen as a way of defining proximity within a social space and that this social space can be reconstructed from the distances between groups. Social interaction will occur most frequently between persons who are socially close to one another and relatively infrequently between those that are socially distant. For this reason CAMSIS scales are often labelled as 'Social Interaction Distance' (SID) scales. The CAMSIS project website offers one methodological approach to deriving SID scales on occupational data. It is important to note that some other research projects outwith the CAMSIS have also used similar approaches to derive related SID scales (see especially the 'Social Status, Lifestyle and Consumption' project).

The further, theoretical, significance of the CAMSIS approach arises from its very nature and from its analytical utility. It challenges conventional ideas about the existence and nature of social classes, and of a distinction between social class and social status. Its predictive success in the analysis of what have commonly been thought of as areas of 'class' analysis, voting behaviour in particular, on the one hand, and its origins in what has conventionally been seen as an aspect of status, social interaction, on the other, reinforces the theoretical argument that the scale measures social and material advantage and that these are indivisible concepts.

Like virtually all other stratification measures, CAMSIS uses occupational groups as its basic units. Employment, directly or indirectly, still provides the major mechanism by which material rewards are distributed in modern society; it is also an important source of other forms of social and psychological rewards or costs. Occupation is still the single most significant and convenient indicator of someone's location in the overall structure of advantage and disadvantage, as well as a major source of social identity. It is important to understand that 'occupational group' is here being broadly defined so as to include, in addition to occupation as normally understood, differences in status in employment (self-employed or supervisory, for example). However, it is possible (and usual) for the scale to be gender- specific and for other bases for groupings, such as ethnicity and education or qualifications (for those not in paid employment), to be incorporated. It is a general principle underlying the construction of the scales that as much detail as possible should be retained. If two groups turn out to be indistinguishable, then they can be combined, but if they are combined from the beginning it is impossible to determine whether or not they are genuinely similar to or different from one another.

For practical reasons - the easy availability in very many countries of large-scale representative data (usually from the census) - the preferred basis on which to construct a CAMSIS scale is information on the occupations of husbands and wives (or cohabiting partners). The evidence suggests that the same structure underlies various forms of association, including marriage, friendship and social reproduction across generations.

When people are aggregated into occupational groups, those groups that are socially close will have similar distributions of associates - marriage partners and friends - while groups that are more distant will have less similar distributions. From a cross-tabulation of the occupations of marriage partners or friends it is possible to determine, for every pair of occupations, the distance between them. The more similar the distributions, the closer they are; the more dissimilar, the more distant. The empirical issue is then one of determining the nature of the social space that is defined by the whole set of pair-wise distances. It might be that the whole set of distances is consistent with a simple one-dimensional ordering; or it might require several dimensions in order for some sets of distances to be made consistent with one another. Similarly, occupational groups might be clustered together, with large gaps between the clusters; or they might be distributed fairly evenly and continuously in the space.

Although, strictly speaking, a CAMSIS scale is ordinal (or ordered interval), the large number of categories means that it can be treated as effectively continuous. In areas such as social mobility, voting behaviour, educational attainment or health a CAMSIS scale has been shown to be no less, and often more, efficient in predictive terms than class schemas. Because it is continuous, it generally reflects a social reality of finer gradations more clearly than do the latter.

The CAMSIS approach to social stratification is, then, novel and distinctive. It provides a measure of social stratification arrangements that is firmly grounded in actual social behaviour; it has a more adequate theoretical underpinning; and it has been shown to be very successful, in terms of statistical prediction, in a variety of significant areas of analysis.

Practical issues

The CAMSIS project involves the construction and dissemination of versions of CAMSIS scales for a range of countries and time periods. Previous research using this approach was the basis of the 'Cambridge Scale' for Britain, first published in full in 1980 and subsequently updated in 1990 (see the bibliography). Between 1999 and 2003 the UK Economic and Social Research Council funded the CAMSIS project to expand the range of countries and time periods on which CAMSIS scales applied, leading to the dissemination of around 20 national scales and the creation of these webpages (which were hosted at Cardiff University from 2000-2005). Since 2003 researchers from the CAMSIS project have continued to work on creating new scales as part of a number of other projects and collaborations (including HIS-CAM, funded by the Dutch NWP, and SONOCS, GEODE and DAMES, all funded by the UK ESRC).

Given their empirical basis, new CAMSIS scales (for new countries or time periods, or even updates of existing scales) can always be created. Within the CAMSIS project new scales continue to be occassionally added to these webpages. Anyone interested in computing a new scale using these approaches is most welcome to make contact with us through these webpages (paul.lambert@stirling.ac.uk). An abbreviated listing of existing CAMSIS scales is available in this MS Excel file though it is not routinely updated.

The CONSTRUCTION section of these webpages covers, in considerable detail, the technical and substantive issues involved in scale constructions, whilst the VERSIONS section of these pages can be used to find out which countries are being considered in the project, how far the work has progressed, as well as providing links to downloadable files with the completed CAMSIS scale values when available. In addition an account of the key issues in utilising the CAMSIS scale scores can be found on a page on accessing and using the CAMSIS scale scores.

The CAMSIS webpages which you are reading at present were originally written by Ken Prandy in 1999, then were substatially expanded by Paul Lambert during the 1999-2003 project mentioned above (then available at at URL at Cardiff University). Ken was responsible for the original drafts of many of the webpage sections, whilst Paul added many of the sections featuring technical details and practical instructions (Paul also admits responsibility for most of the spelling mistakes on these pages!). Since 2005, the CAMSIS project pages have been hosted at Stirling University, and are maintained primarily by Paul Lambert.

 

 
Contact details / Register your interest

We should like to maintain a register of researchers who have expressed an interest in using CAMSIS measures, so that we can keep them informed of future developments. If you would like to be included on the register, please send an email message, with the subject heading 'CAMSIS' (with an additional message if you wish), to: paul.lambert@stirling.ac.uk

The authors of these pages are Ken Prandy and Paul Lambert. The pages were jointly authored when both worked on the CAMSIS project at Cardiff University over the period 1999-2003. Thereafter the pages have been hosted at Stirling University and are maintained and occasionally updated by Paul Lambert.

Many others have contributed towards materials which are linked to the CAMSIS approach, the related Cambridge Scale (first publication in 1973), and the individual CAMSIS versions which are described at this site. As much as possible, accurate recent contact details for contributors to the available scales have been listed from the 'versions' page at this site. Though they are not the only contributors, we would highlight our colleagues Max Bergman, Erik Bihagen, Bob Blackburn, Wendy Bottero and Richard Zijdeman who have all made considerable contributions towards materials at this site in the relatively recent past. The name CAMSIS itself was coined, as best we remember, by Max Bergman.

Comments concerning these webpages, the CAMSIS project, and technical issues in the operationalisation of CAMSIS measures, are also welcome and we will make every effort to reply - emails can be sent to Paul Lambert or Ken Prandy .

 

 
Citing CAMSIS scales

Whenever you use a CAMSIS scale accessed from this site it good practice that you should cite it appropriately.

A few of the scale versions have specific citation requests associated with them (if so these are listed in a note where the versions can be downloaded from). Also, some of the CAMSIS versions at this site have been published in specific research articles or other publications, in which case you should cite the relevant publication.

Otherwise, we recommend that you should use at least one of the three following citation forms:

Lambert, P.S. and Prandy, K. ({year that page was last modified}) CAMSIS project webpages: Cambridge Social Interaction and Stratification Scales, Retreived {insert date of access} from http://www.camsis.stir.ac.uk/ .

and/or

Lambert, P.S. (2012) CAMSIS for Britain, SOC2010 (electronic file, version 1.0, date of release: 8 September 2012), Retrieved {insert date of access} from http://www.camsis.stir.ac.uk/Data/Britain2010.html .

(As noted above, in some instances the specific scale has been described in a research article or other publication, in which case we would encourage you to cite that publication as a means of citing the page).

and/or

Prandy, K., & Jones, F. L. (2001). An international comparative analysis of marriage patterns and social stratification. International Journal of Sociology and Social Policy, 21, 165-183.

Thank you in advance for taking the time to cite CAMSIS scales appropriately in your research!


Last modified 9 Sep 2012
This document is maintained by Paul Lambert (paul.lambert@stirling.ac.uk)

 

August 2005 url change: These pages moved in August 2005 to a url at Stirling University: www.camsis.stir.ac.uk (they were previously at Cardiff University at: www.cf.ac.uk/socsi/CAMSIS/ ).