CAMSIS - 13 May 2008 1

Understanding social stratification through social interactions between occupations: The CAMSIS approach

Paul Lambert, University of Stirling, UK Wendy Bottero, University of Manchester, UK

Presentation to the conference ‘Occupational Stratification: Social change and methodological issues’, University of

Eastern Piedmont, 13-14 May 2008

CAMSIS - 13 May 2008 2

The CAMSIS approach

CAMSIS: ‘social interaction distance’ scales

1.1) Introduction to approach

1.2) Methodology and its empirical features

1.3) Interpretations

Evidence about occupational structures

2.1) Universal and specific alternatives

2.2) Evidence on change and stability

CAMSIS - 13 May 2008 3

1.1) CAMSIS in brief

Cambridge Social Interaction and Stratification scales

• [Stewart et al 1973, 1980; Prandy 1990; Prandy & Lambert 2003]• www.camsis.stir.ac.uk

– One dimensional summary of a structure of social distance between occupations that is interpreted as a measure of social stratification

– Calculated according to empirical patterns of social interaction between the incumbents of occupations, using data on friendship, marriage, or father-son intergenerational mobility

– Family of scales for different countries, time periods, men and women

CAMSIS - 13 May 2008 4

Social stratification & social interactions

• Central concern is the reproduction of social inequalities • Social relations are key agents in reproducing inequalities

– E.g. social interactions, homophily, social networks

Classically, intergenerational social relations show patterns which tell us about the structure of social inequalities (e.g. Weber)

Generally, other patterns of social relations (social interactions) also reveal the same patterns of structured social inequalities

Social stratification [Bottero 2005, p3] = ‘the patterning of inequality and its enduring

consequences on the lives of those who experience it’

CAMSIS - 13 May 2008 5

Social stratification & occupations

• Occupational titles as convenient ‘tags’ to locate people within a social structure– Long term indicators of lifetime social circumstances– Easily measured at high level of detail

• Complications– Occupational restructuring - tags have potential to change over

time and between countries in their relative positions– Gender - male and female occupational distributions

• For SID scales, objective features of occupations are (potentially) irrelevant– Similar to prestige rankings – Departure from conventional class categorisations

CAMSIS - 13 May 2008 6

1.2) CAMSIS methodology

www.camsis.stir.ac.uk

Website sets out principles for deriving scales using correspondence analysis or RC-II association models

Also site for distributing databases with CAMSIS scales (c30 countries, scales for periods 1800-2001)

• Derivation: – Requires a dataset of pairs of occupations linked by a social

interaction – Instructions on methods using SPSS, lEM {and Stata}

CAMSIS - 13 May 2008 7

Tabular analysis (correspondence analysis; RC-II association models)

Husband’s Job Units Occ. Units ↓ → 1 2 .. 407

Derived dimension scores ↓ → 75.0 70.0 .. 10.0 Wife’s 1 72.0 30 15 .. 0

Job 2 72.5 13 170 .. 1

Units .. .. .. .. .. ..

407 11.0 0 2 .. 80

A large cross-tabulation of pairs of occupations is modelled; dimension scores help predict frequency of occurrences in cells; scaled dimension scores arethen presented as CAMSIS scale scores.

CAMSIS - 13 May 2008 8

CAMSIS - 13 May 2008 9

From: Bozon and Heran (1989), ‘Finding a spouse: A survey of how French couples meet’, Population, 44(1):91-121.

10

Ongoing methodological issues• Dimensionality

– We find there is always one discernible ‘stratification’ dimension (continuous in character – challenges class accounts)

– Other dimensions to the social interaction space between occupations include farming; gender segregation; regions

• Current practice of controlling for ‘Pseudo-diagonality’– manually identify and exclude institutionally connected occupations

• Sparsity / re-coding occupations • Gender

– For male-female social interactions, row and column scores (m/f) are discernibly different; a male and female scale is published

– For cross-gender analysis, use male for both; or m for m and f for f...?

• Confidence intervals on scale scores? • Interactive or automated derivation?

– E.g. HIS-CAM – 10 national permutations*4 gender patterns*5 levels of occupational detail*5 time periods=1000 different scales

CAMSIS - 13 May 2008 11

1.3) Interpretation

SID scales identify a stratification space defined by social relations– Weber – reproduction in life chances and life styles – Bourdieu – reproduction through social space [e.g.. Bottero 2005]– Chan and Goldthorpe [2004; 2007] – social interactions reflect

deference and authority and therefore ‘status’

Understanding concepts and measures – SID scales using occupations correlate c0.8 or more with ISEI, SIOPS,

and other scales and schemes– Interpretation that they measure a generalised structure of social

stratification advantage [Stewart et al. 1980; Prandy 1990; Rytina 1992] All occupation-based social classifications measure this same structure –

but SID scales, which emphasise both the economic and cultural aspects of social stratification, are better for recognising this

CAMSIS - 13 May 2008 12

2.1) Universal and specific alternatives

• ‘Universal’– one scaling of occupations is adequate across time /

between countries / for men and women (‘Treiman constant’)

• ‘Specific’– useful to have different scalings between countries,

time periods, etc

– SID approaches could be either universal or specific – CAMSIS scales have, a priori, been ‘specific’

• Website: c300 different scales from c30 countries

CAMSIS - 13 May 2008 13

Universality or specificity? [Lambert et al. 2008]

• Easy to demonstrate some specificity• Certain occupations change positions in SID scales in a manner that is

substantively plausible (e.g. farming) • Some scale derivations use nested loglinear models, and likelihoods and

BICs favour specificity • Social scientists / social historians are ordinarily interested in differences /

changes in occupations’ relative positions

• Whether there is enough specificity is unclear • Most scales correlate well with most others In most uses of occupation-based measures, universality is fine• With SID data, some ‘specificity’ could be measurement error• Practical obstacles to specificity (but see www.dames.org.uk /

www.geode.stir.ac.uk)

CAMSIS - 13 May 2008 14

2.2) Evidence on change and stabilitySociologically, change in the relative stratification advantage

associated with any particular occupational position is hypothesised

• Examine components of occupational activities and conditions– e.g. Euroccupations; Guveli 2006; Oesch 2006 – Describes occupations and occupational structure– Doesn’t necessarily tell us about how occupations are valued within the social

arrangements of the stratification structure

• Concentrate on social positioning of occupations– Prestige rankings – Average rankings by combinations of social outcomes (e.g. ISEI) – Social Interaction Distance scales

• Evaluating change using SID scales – Stability is the main pattern in all examples– Problem of non-comparable occupational unit groups – Problems of potential measurement error, and lots of occupational positions

CAMSIS - 13 May 2008 15Universal scale (disadvantage to advantage)

His

tori

cally

sp

ecif

ic s

cale

s

98420 Railway Brakeman (Freight Train)

56070 Presser (Hand)

39350 Insurance Clerk32140 Typist

32120 Stenographic Secretary

21300 Sales Managers

7210 Auxiliary Nurse

98420 Railway Brakeman (Freight Train)

56070 Presser (Hand)

39350 Insurance Clerk

32140 Typist32120 Stenographic Secretary

21300 Sales Managers

7210 Auxiliary Nurse

Later period

Earlier period

Figure 4: Universal to Historical-specific scale scores, HISCO unit groupsHIS-CAM V0.1:

16

National changes in SID scales – ISCO-88 major groupsGer. 1995 Hung. 1996 Swi. 90 Rus. 92 Slv. 94

m f m f m m m

1. Managers, legislators, senior officials

59.5 62.2 63.0 62.1 56.5 72.2 80.8

2. Professionals 79.6 74.0 75.1 71.0 72.9 78.1 89.9

3. Assoc. Professionals and technicians

55.6 55.3 59.5 55.1 56.7 59.6 61.3

4. Clerks 50.4 54.5 52.0 51.4 43.4 59.3 47.4

5. Service workers / sales 47.7 49.1 52.4 55.2 47.9 55.7 49.3

6. Skilled agricultural 42.5 28.4 34.4 33.9 43.3 55.0 40.1

7. Craft and trades 40.2 31.9 42.4 37.7 39.0 43.5 44.0

8. Plant and machine operators / assemblers

31.6 26.6 39.2 30.7 33.2 40.8 40.5

9. Elementary occupations

35.4 25.1 31.7 28.0 33.8 41.3 37.7

Scales from www.camsis.stir.ac.uk, versions originally derived using ISCO-88

17

Comparing SID scales – USA 1960 / 1990 / 2000, male CAMSIS scores for particular occupations

1960 1990 2000

Architect (13; 43; 130) 83.2 78.4 76.1

Sociologist (175; 168; 183) 84.1 73.2 75.0

Telephone operators (353; 348; 502) 51.2 49.3 44.3

Farmer (200; 473; 21) 46.3 46.0 50.1

Farm labourer (902; 479; 605) 31.1 27.3 26.3

Barber (814; 457; 450) 46.6 40.8 45.0

Boilermakers (403; 643; 621) 51.8 34.2 38.8

From www.camsis.stir.ac.uk, scales derived on SOC data from IPUMS. Figures in brackets show the SOC code used for each year

CAMSIS - 13 May 2008 18

Summary: occupational change and CAMSIS scales

• The main story is of similarity in occupational rankings over time

• It’s not easy to tell a coherent story about major changes in occupations’ meanings

• There are examples of occupations with plausible (non-measurement error) changes over time and between countries– Occupations with many female workers– Farming occupations– Occupations in declining / expanding sectors – Occupations in economies in rapid transition

CAMSIS - 13 May 2008 19

Summary - contributions of CAMSIS scales

• Summary measure of occupational positions• Differentiates finer occupational details

– Typically 300+ occupational units assigned different scores

• Emphasises a hierarchical structure of inequality• Measures relative advantages typically associated with

incumbents of an occupational position

• Explorative device for understanding occupations

• Measure multiple relative structures of stratification between countries, time periods, gender based groups..?

CAMSIS - 13 May 2008 20

References

• Bottero, W. (2005). Stratification: Social Division and Inequality. London: Routledge.• Bozon, M., & Heran, F. (1989). Finding a Spouse: A Survey of how French Couples Meet.

Population, 44(1), 91-121.• Chan, T. W., & Goldthorpe, J. H. (2004). Is There a Status Order in Contemporary British

Society. European Sociological Review, 20(5), 383-401.• Chan, T. W., & Goldthorpe, J. H. (2007). Class and Status: The Conceptual Distinction and its

Empirical Relevance. American Sociological Review, 72, 512-532.• Guveli, A. (2006). New Social Classes within the Service Class in the Netherlands and Britain:

Adjusting the EGP class schema for the technocrats and the social and cultural specialists. Nijmegen: Radbound University Nijmegen.

• Oesch, D. (2006). Redrawing the Class Map: Stratification and Institutions in Britain, German, Sweden and Switzerland. Basingstoke: Palgrave.

• Prandy, K. (1990). The Revised Cambridge Scale of Occupations. Sociology, 24(4), 629-655.• Prandy, K., & Lambert, P. S. (2003). Marriage, Social Distance and the Social Space: An

alternative derivation and validation of the Cambridge Scale. Sociology, 37(3), 397-411.• Rytina, S. (1992). Scaling the Intergenerational Continuity of Occupation : Is Occupational

Inheritance Ascriptive after all? American Journal of Sociology, 97(6), 1658-1688.• Stewart, A., Prandy, K., & Blackburn, R. M. (1973). Measuring the Class Structure. Nature.• Stewart, A., Prandy, K., & Blackburn, R. M. (1980). Social Stratification and Occupations.

London: MacMillan.