Measuring class - Cambridge Social Interaction and Stratification

Measuring class - Cambridge Social Interaction and Stratification
Measures of social stratification and their
consequences: Occupational and nonoccupational measures in the study of social
stratification and mobility
Paul Lambert1, Dave Griffiths1, Richard Zijdeman2
1. University of Stirling, contact paul.lambert@stirling.ac.uk
2. University of Utrecht / International Institute of Social History
Paper presented to the Royal Statistical
Society, Social Statistics ordinary meeting
29 May 2013
1
1) Introduction: Measuring stratification position
2) Different occupation-based measures of
stratification position, and the evidence they give
on social mobility
3) Measuring stratification position with measures
that are not based on occupations
4) Comparing the empirical properties of occupationbased and non-occupation-based measures of
stratification position and of inter-generational
social mobility
2
(1) Introduction: Measuring stratification position
Traditional sociological perspective: stratification position
= enduring relative location of people and their families in
a system of socially organised, consequential, economic
inequalities
Illustrative occupational positions in the UK
SOC90
CAMSIS
2
NS-SEC
Large employers/Hi. managerial
900
3
4
Higher professional
5
6
700
(e.g. Bottero 2005; synonymous with
measures of ‘social class’ and/or
socio-economic status)
Lower manag. & professional
7
500
9
Intermediate occupations
10
11
Sml employers/own-account wrks
12
300
13
14
Lower supervisory & technical
15
16
100
Many things might indicate
stratification positions, but
occupations make the most plausible
single option
 Occupational structure the
‘backbone’ of distribution of
inequality (Parkin 1972)
Vingtiles
8
Semi-routine occupations
17
18
19
Male
Female
Routine occupations
Male
Female
M
Source: BHPS adults in Britain, 1991, current job of those in work (16yrs+).
SOC90 plot shows range of occupational unit groups (with 'jittering'). NS-SEC & CAMSIS plots show proportion of cases.
 Data on occupations is reasonably easy to record
 …..& is reasonably stable over time & over the life-course
From Lambert &
Bihagen (2012)
3
F
(1) Introduction: Measuring stratification position
Traditional perspective is increasingly challenged
• Multidimensional character of social circumstances
– Different dimensions can be measured and seem to matter
– Occupation(s); income/wealth; lifestyle; material assets
– {Alleged} decline in centrality of occupations to individuals
• Longitudinal information on circumstances
– Longitudinal change can be measured and ought to matter
• Certain social inequalities of heightened interest may
not be well captured by occupation-based schemes
 Underclass/poverty; home-ownership; elites
and power
 Growing interest in social inequality in
disciplines outside sociology, with nonoccupational focus (e.g. social geography;
economics; public health)
4
(1) Introduction: Measuring stratification position
Motivation for this paper
“..we have found no clear affiliation between specific occupations and our
latent classes. Perhaps, rather than seeking to locate class fundamentally in
occupational ‘blocks’, the time is now ripe for a different, multi-dimensional
perspective, in which occupational membership is spread (though unevenly)
between different classes” (Savage et al. 2013: 245)
? Even though it’s possible to think of counter-examples, on the
whole social stratification is probably still best studied in
terms of occupations!
– Enduring debates on occupation-based measures
– Comparisons with non-occupation-based measures and
their data sources and spurious correlates
– Particularly interesting to study a traditional sociological
topic – social mobility - from traditional and nontraditional perspectives
5
(2) Different occupation-based measures of stratification
position, and the evidence they give on social mobility
• There are many possible measures!
• They are substantially correlated with each other
• Their qualities are influenced by their functional form
From Lambert &
Bihagen (2012)
Predictors of ‘poor health’ in Sweden
6
(2) Different occupation-based measures of stratification
position, and the evidence they give on social mobility
0
.5
1
1.5
• Could be fruitful to interpret differences between occupation-based measures
• Leads to new theoretical
interpretations (e.g. Chan
Difference in residuals: CG resid - EGP resid
& Goldthorpe 2007)
• Risk that empirical
differences between
measures might not
reflect what is
theoretically intended
• Lambert & Bihagen 2007
-1
0
1
cf. Bukodi et al. 2011
Total variance of residuals
Region (0.2%)
• Netuveli & Bartley 2012
Gender+Age (5%)
Educ (indv) (7.5%)
cf. Blane et al. 2007
Marital status (0.03%)
Employment status (70%)
Asset Sp./Autonomy (occ) (0.01%)
Manual/non-manual (2%)
From Bihagen &
Lambert (2012)
Edyrs (occ) (7.5%)
Gender profile of job (7.6%
Source: Analysis of difference in residuals predicting income using CG or EGP.
Data from 5266 adults in work in Britain, 1991 BHPS. Value modelled is CG resid - EGP resid.
Percents & lines refer to r2 increment from adding relevant variables to model predicting residuals
(sum of these increments is approximately equal to overall model r2 = 0.43).
7
(2) Different occupation-based measures of stratification
position, and the evidence they give on social mobility
Ongoing debates over occupation-based measures
– Decide by fiat? Encourage sensitivity analysis?
– Age and gender correlations are a substantial element of
empirical differences between measures
– Much, but not all, of stratification difference is hierarchical, so
using a gradational measure is often compellingly parsimonious
• Generic issues
– What to do with the non-working?
– How to account for the household?
– Whether to use detailed or broad occupational data, and many
or few model parameters?
8
(2) Different occupation-based measures of stratification
position, and the evidence they give on social mobility
• Even though it’s possible to think of counter-examples, on the whole
social stratification is probably still best studied in terms of
occupations!
• Occupational data can
be readily measured,
stored, processed, and
linked to nearly everybody
• Occupations have
strong correlations to
other outcomes and are
important to people
• Relative occupational
positions are fairly stable
through time
From Lambert and
Gayle 2009
Importance of 'Having a fulfilling job'
I'm going to read out a list of things that people value. For each one I'd like you to tell me
on a scale of 1 to 10 how important each one is to you.
Not important
2
3
4
5
6
7
8
9
Very important
male
female
Source: BHPS wave M (2003), Scottish respondents, valid N=2733, variables 'mlfimp*'.
Other options (mean): Health (9.5); money (6.5); children (7.7); job (7.9);
independence (8.7); own own home (7.7); good partnership (8.9); good friends (9.3) 9
(2) Different occupation-based measures of stratification
position, and the evidence they give on social mobility
(Parent-child correlation by birth cohort and survey)
Most occ.
measures
suggest
long-term
rise in social
mobility
(Updated results
from dataset
described in
Lambert et al.
2007). Men, 20100, N=106k (all);
80k (yob > 1900).
10
Social mobility in Britain by year of birth (splines)
.1
.2
.3
.4
.5
.6
Ages 20 to 100. Men only.
1900
1920
CAMSIS
CAMSIS
1940
Birth cohort
RGSC
RGSC
1960
1980
EGP
EGP
Data from the 'Slow degrees' pooled survey dataset - see Lambert et al. (2007). N = 117199.
Points are correlation statistics for father-son association, 5 year surveys / 10 year birth cohorts.
11
Social mobility trends in Britain by year of birth
.6
.4
.2
0
-.2
-.2
0
.2
.4
.6
Ages 20 to 100. Women only.
1800
1850
1900
Birth cohort
1950
CAMSIS
CAMSIS
2000
1900
RGSC
RGSC
1920
1940
1960
Birth cohort
1980
EGP
EGP
Data from the 'Slow degrees' pooled survey dataset - see Lambert et al. (2007). N = 95452.
Points are correlation statistics for father-son association, 5 year surveys / 10 year birth cohorts.
12
.6
Ages 20 to 40. Women only.
-.2
0
.2
.4
Trends, and values ,
are somewhat
sensitive to the age
of the child
1900
1920
CAMSIS
CAMSIS
1940
Birth cohort
RGSC
RGSC
1960
1980
EGP
EGP
Data from the 'Slow degrees' pooled survey dataset - see Lambert et al. (2007). N = 31329.
Points are correlation statistics for father-son association, 5 year surveys / 10 year birth cohorts.
Social mobility in Britain by year of birth (splines)
0
.1
.2
.3
.4
.5
Ages 40 to 80. Women only.
1900
1920
CAMSIS
CAMSIS
1940
Birth cohort
RGSC
RGSC
1960
1980
EGP
EGP
Data from the 'Slow degrees' pooled survey dataset - see Lambert et al. (2007). N = 23735.
13
Points are correlation statistics for father-son association, 5 year surveys / 10 year birth cohorts.
Social mobility in Britain by year of birth (splines)
0
.2
.4
.6
.8
Ages 25 to 80. Men only.
1900
1920
CAMSIS
CAMSIS
1940
Birth cohort
RGSC
RGSC
1960
EGP
EGP
1980
Income
Data from the 'Slow degrees' pooled survey dataset - see Lambert et al. (2007). N = 72509.
Points are correlation statistics for father-child association, 5 year surveys / 10 year birth cohorts.
(Income figures from Blanden et al. 2004.)
 Different types of measure of stratification have hitherto led to different
(influential) interpretations
14
(3) Measuring stratification position with
measures that are not based on occupations
Intuitive, theoretical and empirical reasons to expect
non-occupation based measures to be revealing
1) Indices of income, wealth, etc.
–
Esp. Corak 2004; Dorling 2013
2) Longitudinal life-course summary indexes
3) Non-traditional occupation-based or incomerelated measures
4) Multidimensional summaries
15
Indices of income, wealth, etc
Access
• Easy to measure current income for individuals and households
• Some surveys have data on wealth, assets, investments
• Hard to reliably measure family income, life-course income, or
social origins income, particularly for suitably mature adults
Qualities
• Convenient metric outcome (& readily converted to key contrasts)
• High correlations with age, life-course stage, family structure
Findings on
intergenerational
mobility
• Rising intergenerational income correlations in recent decades
• UK findings based on income of young adults from 1958 and 1970
cohorts at fixed points, with varying parental ages and high attrition
First 99 percentiles of income measures for BHPS 2005
Household savings, BHPS
None
Up to £1000
£1-10k
0
£10-100k
0
2000
4000
Personal income
6000
Household income
8000
10000
> £100k
16
Longitudinal life-course measures
• Traditional limitation of stratification studies is their focus on
outcome at a single point in time
• Longitudinal data could address this in various ways
• One idea has been to try to construct categories indicative of
typical life-course trajectory
• Occupation-based measures may argue they have already done
this (e.g. Stewart et al. 1980; Goldthorpe and McKnight 2006)!
Adults 30+ in the BHPS 2005: Career trajectories
Males
Females
Stable service
Decline service
Career intermediate
Career/rising manual
0
500
1,000
Approximation to typology of Tampubolon and Savage 2012 using BHPS work-life history records
1,500
2,000
17
Non-traditional occupational or income-related
Access
• ‘Microclass’ measures focus on detailed occupational differences
rather than the low-parameter summaries (e.g. 81 units on UK scheme)
• Measures of polarised inequality focus on the most or the least
advantaged in society, for instance using income, occupational or other
information as available
Qualities
• Microclasses:
• Strong empirical correlations but ambiguity in interpretation
• No standard, comparable scheme over time yet published
• Polarised inequality:
• Most sample surveys lack good coverage of extremes of inequality
(lack of cases at top or bottom + measurement challenges)
• No consensus on how to define extremes
Findings on
mobility
• UK analysis of microclass mobility through time yet to be undertaken
• Elites/most advantaged probably re-opening the gap from mainstream
on economic assets (e.g. Dorling 2013), but social origins of elements
of elites are increasingly heterogeneous (e.g. Griffiths et al. 2008)
• Unclear regarding trends in social origins of those in extremes of
poverty/deprivation
18
Multidimensional measures
Different aspects of individual lives matter, so it might be productive
to summarise social positions drawing upon several aspects
• Multidimensional deprivation measures often used
(e.g. Gordon 2006)
• Market research typologies use area/economic circumstances
• Recent efforts to make multidimensional social stratification
measures for social research purposes
• Hennig and Liao 2013: Combined economic circumstances
• Savage et al. 2013: Combining social, economic and cultural capital
• In the UK, multidimensional measures raise potentially ambiguous correlations
with age, life-course stage, family structure, gender and region
19
[http://www.bbc.co.uk/news/uk-22007058]
•
•
•
•
Intersection of social, economic & cultural capital
For theoretical/qualitative research
Online ‘class calculator’ and offline LCA derivations
Criticisms raised
–
–
–
–
–
‘Great British Class
Scheme’ (GBCS)
- Savage et al. (2013)
- Based on
https://ssl.bbc.co.uk/lab
uk/experiments/class/
BBC’s bumptiousness!
Correlation to age/family/gender
Probably not better than using the underlying measures
Probably not a neat 7 class solution; doesn’t address change over time
Ambiguity of interpretation
20
Operationalisation
• British Household Panel Survey (Univ. Essex 2010) with socioeconomic and behavioural questions on adults 1991-present
• With limitations, it’s possible to operationalise
representatives of all these measures of social position on the
BHPS respondents
• The same is not true for (most) BHPS respondents’
fathers/parents
– Compare with parental occupational positions
– Construct proxy measures which capture most likely position on other
schemes according to the parents occupations(!)
21
Empirical measures of social position – BHPS (wave 5 & 15)
Measure
#; % missing
CAMSIS*
Scale; 11%
Occupational scale score for most recent job
EGP-7*
7; 18%
``
RGSC-5*
5; 18%
``
Microclass*
81; 14%
Allocated via SOC90/ISCO88, most recent job
Personal Income
Scale; 6%
Household income
Scale; 5%
Household savings
Scale; 0%**
Adds together reported savings and investments
Top 1% money
2; 1%**
Top 1% by personal or hhld inc., or house value
Poverty (inc)
2; 5%
< half median hhld income
Life-history
categories
4; 43%
4 categories following Tampubolon & Savage 2012
(stable service; decline service; intermed.; manual)
Hennig-Liao
8; 8%**
See over
GBCS
7; 0%**
See over
* Available in original form for BHPS respondents’ parents
**Includes imputed values, where non-missing = valid data for any component item
22
…(Badly) reconstructing the multidimensional measures…
Hennig & Liao 2013: ‘Appropriate clustering for
mixed type variables with an application to
socio-economic stratification’
HL source (m)
HL-BHPS (m+f)
Savage et al. 2013: Seven latent classes identified
by distributions of social, economic and cultural
capital
GBCS source
GBCS-BHPS
Total monthly savings
Self reported savings
Household income
=
Total personal income
=
Household savings
Sum indv. savings (e~=o)
Years of education
Ranked educ. quals
House value
=
Number of checking
accounts
Has credit card
Social contact score
Mean CAMSIS of
friends/family
Number of savings
accounts
Savings income
category
Social contact number
# friends/family in work
Housing tenure
=
Highbrow cultural
capital
Proportion1/2 contacts
read broadsheets
Has life insurance
=
Emerging cultural
capital
Sports, eating out, pub,
visit friends, computer
Occupational class (6
+not working)
RGSC + not working (8)
23
Hennig-Liao (2013) classification for the BHPS
.8
– Seek to identify constellation of shared characteristics
related to income, housing, employment, & education
– HL publish a routine in R for mixed-type items
– Implemented below on BHPS, choosing 8-cluster solution
0
.2
.4
.6
8-cluster solution using Hennig-Liao items (BHPS adults>20, 2005)
ZIncome+2/4
Age/100
%Fem
• Higher and lower numbers of clusters are also plausible
• “If the model does not hold precisely, the truth may be best approximated according
to the BIC (or any consistent criterion) by a very high number of mixture
components if there are only enough observations, which is of little interpretative
24
value.” (Hennig and Liao 2013: 16)
Approximating GBCS for the BHPS
% cases; % female; mean age
BHPS, 1995
BHPS, 2005
Original
GBCS1
GBCS2
GBCS1
GBCS2
Elite
6; 50; 57
2; 48; 55
14; 43; 49
1; 41; 55
13; 51; 46
Est. middle
class
25; 54; 46
17; 43; 43
20; 46; 41
12; 45; 45
11; 44; 54
Tech. middle
class
6; 59; 52
4; 58; 55
14; 51; 41
6; 59; 57
17; 50; 45
New affluent
workers
15; 43; 44
27; 51; 52
10; 47; 38
21; 53; 44
14; 53; 50
Traditional
working class
14; 62; 66
25; 64; 61
15; 57; 49
28; 63; 64
21; 62; 56
Emergent
service wkrs
19; 55; 34
19; 48; 40
15; 63; 54
23; 47; 41
11; 48; 39
Precariat
15; 57; 50
7; 61; 59
11; 63; 65
9; 61; 62
13; 60; 66
GBCS1: Minimise sum of magnitude of residuals for 7 component items (modal imputation if missing)
GBCS2: Choose a 7 class solution from Hennig-Liao function for component measures
[Correlation GBCS1-GBCS2 = 0.417 (2005)]
25
Background comments on the GBCS
• BHPS clustering on similar items doesn’t correspond at all to GBCS
descriptions, but several possible limitations to my BHPS analysis
(unweighted; suboptimal measures; possible implementation errors)
• GBCS seeks to
distinguish social
capital from
economic capital,
but it’s social
capital score
seems to relate
more to age and
occupations
26
(4) Comparing the empirical properties of occupation-based and nonoccupation-based measures of stratification position and of intergenerational social mobility
Correlations between alternative measures of stratification
0
.2
CAMSIS
Income
HHinc
HSaving
EGP7
RGSC5
Microclass
Top1pc
Poverty
LifeType
HennigLiao
GBCS
.4
.6
.8
1
CAMSIS
EGP7
Poverty
0
.2
.4
.6
.8
1
Income
RGSC5
LifeType
0
.2
.4
.6
.8
1
HHinc
Micro
HennigLiao
0
.2
.4
.6
.8
1
Hsaving
Top1pc
GBCS
Graphs by vartype
Regression r2^0.5 when 1 variable is linear; Cramer's V when both categorical.
27
(4) Comparing the empirical properties of occupation-based and nonoccupation-based measures of stratification position and of intergenerational social mobility
#categ.s
Correlation to sex;
age2; educ
Correlation to smoking; age 1st
child; reads broadsheet newsp.
CAMSIS
-
5; 12; 50
16; 10; 23
EGP-7
7
21; 11; 26
9; 7; 11
RGSC-5
5
7; 11; 31
8; 11; 12
Microclass
81
24; 14; 23
8; 17; 10
Personal Income
-
26; 29; 36
2; 14; 14
Household income
-
8; 38; 32
4; 7; 12
Household saving
-
4; 13; 11
10; 7; 14
Top 1% money
2
8; 15; 24
8; 5; 17
Poverty (by hhinc.)
2
10; 42; 27
1; 16; 9
Life-history categories
4
6; 8; 34
12; 17; 18
Hennig-Liao
8
7; 36; 49
8; 12;11
GBCS
7
8; 32; 28
11; 13; 24
Figures for BHPS adults 2005, using x-sectional weight, N ~= 13000
28
(4) Comparing the empirical properties of occupation-based and nonoccupation-based measures of stratification position and of intergenerational social mobility
Correlation to parental
CAMSIS / EGP / RGSC
Correlation to parental measure
(*occupational proxy)
CAMSIS
30; 15; 15
30
EGP-7
26; 15; 14
15
RGSC-5
24; 12; 12
12
Microclass
33; 16; 16
24 {for scale scores}
16; 8; 8
11*
Household income
21; 11; 11
21*
Household saving
10; 5; 6
15*
Top 1% money
11; 6; 7
0*
Poverty (inc)
13; 6; 6
3*
Life-history categories
31; 15; 16
15*
Hennig-Liao
31; 16; 17
15*
GBCS
35; 18; 18
15*
Personal Income
Correlation values are usually obtained as square root of r2 or pseudo-r2 statistic from linear or mlogit
29
regression. Correlations are systematically larger for continuous variables.
0
.1
.2
.3
.4
Different social mobility trends by measures?
1900
1920
CAMSIS
EGP7
Poverty
1940
yob2
Income
RGSC5
LifeType
1960
HHinc
Microclass
HennigLiao
1980
HSaving
Top1pc
GBCS
Pooled data from BHPS 1995 and 2005 by 5-year birth cohort with weights. N ~= 19k cases, 14k individuals (age 20+).
Points analysed are correlations between measure and fathers CAMSIS score.
Lines 'calculates cross medians and then uses the cross medians as knots to fit a cubic spline' (Stata manual) 30
[correction: R2 values, not correlations as in footnote]
.4
Women, 30+ yrs
.1
.2
.3
Trends in parent-child r2 association?
0
.5
Men, 30+ years
1920
1940
yob2
1960
0
.1
.2
.3
.4
1900
1900
1920
CAMSIS
EGP7
Poverty
1940
yob2
Income
RGSC5
LifeType
1960
HHinc
Microclass
HennigLiao
1980
HSaving
Top1pc
GBCS
31
Some conclusions
• Occupation-based measures perform largely as well as any other
measures in correlating expected factors
• Correlations with age and gender are a complication to many
measures
– This isn’t necessarily a problem (though I’ve implied it is)
– An open question whether we want to use measures to indicate current
situation or general, lifetime circumstances
– But ambiguity with age: occupation-based measures are defined through
occupations, but other measures are more directly determined by age
• Hennig-Liao, GBCS, life history and household income measures do
all seem to pick up something of interest regarding stratification
reproduction/social mobility
– More work needed, but social mobility trends vary by stratification measures
– Cf. Marks (2013): questioning income-based measures
32
Conclusions, ctd
Advocates of non-traditional measures don’t generally seem to
prioritise implementability of their measure on relevant
secondary data!
• Longitudinal profiling works ok for specific age cohorts and
studies, but restricts population coverage
• Income data is rarely held across generations in studies
without attrition and life-course stage bias
• Constellation of measures approach is typically unique to the
relevant survey and hard to replicate
 …All points back to using occupations to me…!
33
References cited
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Bihagen, E., & Lambert, P.S. (2012). Class & Status: One or two forms of stratification? Paper presented at Equalsoc conference, Stockholm, 24-26 Sept.
Blanden, J., Goodman, A., Gregg, P., & Machin, S. (2004). Changes in generational mobility in Britain. In M. Corak (Ed.), Generational Income Mobility in North America
and Europe (pp. 147-189). Cambridge: Cambridge University Press.
Blane, D., Netuveli, G., & Bartley, M. (2007). Does Quality of Life at Older Ages Vary with Socio-Economic Position? Sociology, 41(4), 717-726.
Bottero, W. (2005). Stratification: Social Division and Inequality. London: Routledge.
Bukodi, E., Dex, S., & Goldthorpe, J. H. (2011). The conceptualisation and measurement of occupational hierarchies: a review, a proposal and some illustrative analyses
Quality and Quantity, 45(3), 623-639.
Chan, T. W., & Goldthorpe, J. H. (2007). Class and Status: The Conceptual Distinction and its Empirical Relevance. American Sociological Review, 72(4), 512-532.
Corak, M. (Ed.). (2004). Generational Income Mobility in North America and Europe. Cambridge: Cambridge University Press.
Dorling, D. (2013). How social mobility got stuck. New Statesman, 16 May 2013.
Goldthorpe, J. H., & McKnight, A. (2006). The Economic Basis of Social Class. In S. L. Morgan, D. B. Grusky & G. S. Fields (Eds.), Mobility and Inequality (pp. 109-136).
Stanford: Stanford University Press.
Gordon, D. (2006). The concept and measurement of poverty. In C. Pantazis, D. Gordon & R. Levitas (Eds.), Poverty and Social Exclusion in Britain: The Millenium Survey.
Bristol: The Policy Press.
Griffths, D., Miles, A., & Savage, M. (2008). The End of the English Cultural Elite. In M. Savage & K. Williams (Eds.), Remembering Elites Oxford: Blackwell (Sociological
Review Monographs).
Hennig, C., & Liao, T. F. (2013). How to find an appropriate clustering for mixed type variables with application to socio-economic stratification. Journal of the Royal
Statistical Society Series C, forthcoming 2013.
Lambert, P. S., & Bihagen, E. (2007). Concepts and Measures: Empirical evidence on the interpretation of ESeC and other occupation-based social classifications. Paper
presented at the ISA Research Committee 28 on Social Stratification and Mobility, Montreal (14-17 August).
Lambert, P. S., & Bihagen, E. (2012). Stratification research and occupation-based classifications. In P. S. Lambert, R. Connelly, R. M. Blackburn & V. Gayle (Eds.), Social
Stratification: Trends and Processes (pp. 13-28). Aldershot: Ashgate
Lambert, P. S., & Gayle, V. (2009). 'Escape from Poverty' and Occupations. Paper presented to BHPS Research Conference, 9-11 July 2009
Lambert, P. S., Prandy, K., & Bottero, W. (2007). By Slow Degrees: Two Centuries of Social Reproduction and Mobility in Britain. Sociological Research Online, 12(1).
Marks, G. N. (2011). Issues in the Conceptualisation and Measurement of Socioeconomic Background: Do Different Measures Generate Different Conclusions? Social
Indicators Research, 104(2), 225-251.
Marks, G. N. (2013). Reproduction of Economic Inequalities: Are the Figures for the United States and United Kingdom Too High? In G. E. Birkelund (Ed.), Class and
Stratification Analysis (Comparative Social Research, Volume 30) (pp. 341-363): Emerald Group.
Netuveli, G., & Bartley, M. (2012). Perception is Reality: Effect of Subjective Versus Objective Socio-economic Position on Quality of Life. Sociology, 46(6), 1208-1215.
Parkin, F. (1972). Class Inequality and Political Order: Social Stratification in Capitalist and Communist Societies. London: MacGibbon & Kee.
Savage, M., Devine, F., Cunningham, N., Taylor, M., Li, Y., Hjellbrekke, J., Le Roux, B., Friedman, S., & Miles, A. (2013). A new model of social class: Findings from the
BBC's Great British Class Survey Experiment. Sociology, 47(2), 219-250.
Stewart, A., Prandy, K., & Blackburn, R. M. (1980). Social Stratification and Occupations. London: MacMillan.
Tampubolon, G., & Savage, M. (2012). Intergenerational and Intragenerational Social Mobility in Britain. In P. S. Lambert, R. Connelly, R. M. Blackburn & V. Gayle (Eds.),
Social Stratification: Trends and Processes (pp. 115-131). Aldershot: Ashgate.
University of Essex, & Institute for Social and Economic Research. (2010). British Household Panel Survey: Waves 1-18, 1991-2009 [computer file], 7th Edition.
34
Colchester, Essex: UK Data Archive [distributor], July 2010, SN: 5151.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising