Vernon Gayle Professor of Sociology School of Applied Social Science University of Stirling

1
Vernon GayleProfessor of Sociology School of
Applied Social ScienceUniversity of Stirling

• - A summary of your academic history and
  foundations
• - Where you would like to take your research in
  future
• - How you will build your research within the
  context of St Andrews
• vernon.gayle_at_stir.ac.uk
• www.staff.stir.ac.uk/vernon.gayle/
• www.dass.stir.ac.uk/staff/showstaff.php?id28

2
Summary of Academic History and Foundations

• Sociologist by training
• In an Applied Social Science Department since
  1994
• Working alongside non-survey based researchers
• Inter and multidisciplinary researcher
• Geographers, economists, statisticians, computer
  scientists, health
• Research based on detailed empirical analyses
• Large-scale social surveys (especially
  longitudinal data)
• Multivariate statistical analysis
• Emphasis on advanced techniques

3
Ongoing Collaborations with St Andrews

• Secondment to the Longitudinal Studies Centre
  Scotland (2003-6)
• Galvanise the already successful set of
  collaborative arrangements and rapidly make
  substantial progress
• Boyle, Graham, Feng, Feijten and Flowerdew
• Work in longitudinal data, migration, fertility
  and family life
• 5 ESRC funded awards (research and knowledge
  exchange) 6 publications 13 conference
  presentations 4 research reports 6 ESRC funded
  consultancies

4
Longitudinal Data

• Longitudinal data are not a panacea
• For many analyses cross-sectional data are
  suitable
• Most analyses can be improved when longitudinal
  data are incorporated
• I argue that some research questions require
  longitudinal data
• Flows into and out of poverty
• The effects of family migration on womans
  subsequent employment activities
• Evaluating policy interventions
• Investigating individual development

5
Moving to St Andrews

• Developing spatial elements in my work
• Intellectual ambition is to develop suitable
  collaborations with social geographers in order
  to provide more comprehensive analyses of both
  the temporal and the spatial elements of
  contemporary social life

6
Some Current Research Areas

• Sociological / Educational research in social
  stratification
• youth transitions, education, occupations
• Research in human geography
• family migration, ESRC Centre for Population
  Change
• Methodology
• better communicating results, quasi-variance,
  missing data methods

7
Some Other Current Research Areas

• Modelling ordinal panel data
• Gayle (1996) ESRC NCRM attitudinal data bi and
  tri variate outcome random effects models
  (correlated error structures)
• Data management
• ESRC NCeSS Node managing, enabling survey data
  constructing measures grid technology digital
  social research
• Knowledge transfer/capacity building
• ESRC RM Programme, RDI Phase 1 2, ESRC AQMeN
  training researchers building capacity
  statistical modelling longitudinal data
  analysis ONS Scottish Gov Local Authorities

8
Parental Occupations and Filial Attainment

• Extended analyses of the Youth Cohort Study of
  England and Wales
• Overall trend
• Increasing proportions getting 5GCSEs (A-C)
• Increasing mean number of A-C grade GCSEs
• Increasing mean GCSE points score
• Gender
• Female pupils outperforming male pupils
• Ethnicity
• Some groups doing better than white pupils (e.g.
  Indians)
• Other groups doing worse (e.g. blacks)
• Parental Occupation
• Observable gradient
• Lower levels of GCSE attainment from those pupils
  with less occupationally advantaged parents
• Sensitivity analysis of 9 popular occupational
  measures (Adj. R2 .15 through to .20)

9
Exploring parental influences at occupational
unit group (OUG) levelNational Statistics
Socio-economic Classification (NS-SEC)

• NS-SEC No. of SOC90 Occupations
• 1.1 Large Employers and higher managers 10
• 1.2 Higher professional occupations 38
• 2 Lower managerial and professional
  occupations 78
• 3 Intermediate occupations 42
• 5 Lower supervisory and technical
  occupations 41
• 6 Semi-routine occupations 88
• 7 Routine occupations 74
• Total 371
• Employees
• Possible interesting variations within NS-SEC
  categories?

10
(No Transcript)
11
YCS Data
Secondary Teachers Publicans
Count 1320 222
5 A-C 78 25
Mean No. A-C 7.00 2.80
Mean GCSE Score 49.05 29.64

McKnight and Elias (1998) 371 Database
Secondary Teachers Publicans
Male Earnings Band 450-500 250-300
Female Earnings Band 350-400 150-200
Male Graduates in Occupation 85 4
Female Graduates in Occupation 71 1
Regrettably the micro-data used to construct the
371 database is no longer available! Working to
reconstruct this information from summary 371
database files Working to construct similar
measures from the Labour Force Survey
12
Microclass Analysis

• There might be extra insights somewhere between
  big class categories and individual
  occupations?
• Exciting debate emerging
• Punch up between heavyweights
• For microclasses Grusky, Weeden and Jonsson
• Against Goldthorpe and Erikson
• Jonsson et al 2009 AJS Grusky and Weeden (2005,
  2006)
• Between 8 categories and 371 unorganised
  occupational unit groups, could there be 80-120
  microclasses defined by their professional
  cultures and practices?

13
Microclass Analysis

• Microclass regime The microclass approach
  shares with the big-class model the presumption
  that contemporary labor markets are balkanized
  into discrete categories, but such balkanization
  is assumed to take principally the form of
  institutionalized occupations (e.g., doctor,
  plumber, postal clerk) rather than
  institutionalized big classes (e.g., routine
  nonmanuals, proprietors)
• (Jonsson et al 2009 pp.982-983)

14
Microclass Reproduction

• Mechanisms of Intergenerational Reproduction
• (Jonsson et al 2009 Table 1 p.986)
• Human capital
• Occupation-specific skills (e.g. carpentry)
• Cultural capital
• Occupation-specific cultures and tastes
• (e.g. aspirations, medicine, help with UCAS
  application)
• Social networks
• Occupation-specific networks
• (e.g. doing the knowledge, job interviews,
  internships)
• Economic resources
• Fixed resources (e.g. farms, market stalls,
  business in general)

15
Microclass Analysis

• The initial appeal is the prospect of clearer
  resolution regarding
• Occupation-Specific Human Capital
• Occupation-Specific Cultural Capital
• Other Occupation-Specific Mechanisms
• First attempt (that we are aware of) to construct
  a British microclass scheme
• Example (from Gayle and Lambert 2011)
  http//www.staff.stir.ac.uk/vernon.gayle/documents
  /gayle_lambert_rc28_v1.pdf

16

• Examples of the Composition of Microclasses
• Health Professionals Health Semi-Professionals
• 220 Medical practitioners 222 Ophthalmic
  opticians
• 221 Pharmacists / pharmacologists 340 Nurses
• 223 Dental practitioners 341 Midwives
• 224 Veterinarians 342 Medical radiographers
• 343 Physiotherapists
• Workers in religion 344 Chiropodists
• 292 Clergy 345 Dispensing opticians
• 347 Occupational and speech therapists
• Elementary and Secondary teachers 348
  Environmental health officers
• 233 Secondary school teachers 349 Other health
  associated professionals
• 234 Primary school teachers
• 235 Special education
• 239 Other teaching (e.g. dance)

17
(No Transcript)
18
(No Transcript)
19
Microclass Analysis
Least Squares Dummy Variable Models GCSE Score

No. Units Adjusted R2
NS-SEC (8 category) 8 .19
SOC 90 units 369 .22
ISCO 88 102 .21
Microclass units 81 .21
Controls CohortGenderEthnicity n55120
20
Microclass Analysis

• First attempt to construct a British microclass
  scheme
• Extra explanatory power (for GCSE attainment)
  questionable
• The initial appeal was the prospect of clearer
  resolution regarding
• Occupation-specific human and cultural capital
  and occupation specific mechanisms
• Family migration and microclasses / beyond big
  classes
• Mobility / immobility of microclasses
• Trailing spouses
• License to practice
• Geographical distribution of microclasses
• Unemployment at microclass level

21
Human Geography (Family Migration)
Boyle, P., Kulu, H., Cooke, T., Gayle, V. and
Mulder, C. (2008) The Effects of Moving on
Union Dissolution, Demography, 45(1), pp.
209-22.
Boyle, P., Feng, Z. and Gayle, V. (2009) A New
Look at Family Migration and Womens Employment
Status, Journal of Marriage and Family, 71, pp.
417-431.
Gayle, V., Boyle, P., Flowerdew, R. and Cullis,
A. (2008) Exploring the relationship between
family migration and social stratification
through the investigation of womens labour
market experiences in contemporary Britain,
International Journal of Sociology and Social
Policy (Special Issue), 28 (7/8), pp. 293-30.
22
Family Migration

• ESRC Centre for Population Change
• Collaboration with Elspeth Graham and Marina
  Shapira (GROS)
• Greatly extends our previous BHPS based research
• Huge data preparation exercise
• Data in an advanced state of readiness
• Combining detailed migration information with
  fertility, partnership, employment and
  occupational data
• Paper accepted Understanding Society / BHPS
  Conference July 2011

23
Family Migration

• Colleagues MRC/CSO Social and Public Health
  Sciences Unit, Glasgow
• The relationship between childhood residential
  mobility and health in the UK is not well
  established
• Research elsewhere suggests that frequent
  childhood moves may be associated with poorer
  health outcomes and behaviours
• Comparison of people in the West of Scotland who
  were residentially stable in childhood with those
  who had moved in terms of a range of health
  measure (West of Scotland Twenty-07 Study)

24
Family Migration

• Submitted to Journal of Epidemiology and
  Community Health
• In a nutshell
• Risk of poor health was elevated in adolescence
  and adulthood with increased residential mobility
  in childhood, after adjusting for
  socio-demographic characteristics and school
  moves
• Childhood mobility associated with
• overall subjective health
• psychological distress
• health behaviours
• but not physical health (medical data)

25
Methodology

• Missing data (item non-response) enduring survey
  problem
• Practical issue - Youth Cohort Study young
  people being asked about their parents
  occupations
• In the 1990s cohorts approximately 12 of pupils
  with missing parental occupation data
• Nobel et al (2008) testing pupils with the YCS
  question and checking with parental interview
  data
• 60 of young people correctly reported parents
  occupations at 4 digit Occupational Unit Group
  (e.g. 2111 Chemist)
• Disappointingly only 74 managed it at the 1
  digit level either they know exactly or they
  dont know at all
• Nobel et al (2008) report no significant social
  class pattern (using NS-SEC)!

26
Missing Data Multiple Imputation

• Can we get further using some of the recent
  insights from the missing data and multiple
  imputation literature?
• Carpenter Bartlett (LSHTM), Goldstein (Bristol)
• Multiple imputed datasets (creation and analysis)
• Creating imputations by chained equations (ice)
  in Stata (n64K not 55K)
• Results are promising
• Important first step, our focus was missingness
  on parental social class, but original models
  were underestimating ethnicity effects
• Richer (congenial) models for imputation
• Breakthrough is fitting survey weighted models
  for imputation
• Compared results with other estimation techniques
  (e.g. Realcom)
• We are looking into a generalisation to
  multilevel framework
• Application to spatially clustered data!

27
Understanding youth transitions in the context of
contemporary home and family life
Possible UKHLS (BHPS) data sources
28
How I will build my research in the St Andrews
context

• Developing spatial elements within my work
• Intellectual ambition for more comprehensive
  analyses of both the temporal and the spatial
  elements of contemporary social life
• What do I bring?
• Enthusiasm, commitment, energy
• Methodological skills
• Sociological insights
• Inter and multidisciplinary researcher expertise

29
Research and Teaching

• Research led teaching
• Research intensive university
• UG teaching is extremely important (growing
  postgraduates)
• Methodological teaching
• Substantive teaching researching with large-scale
  datasets
• Research supervision
• Locate within Population, Health and Welfare
  Group
• Continuing to work within the ESRC Centre for
  Population Change
• Migration work

30
Developing the Longitudinal Studies Centre
Scotland

• Maintaining continuity and recognising
  opportunities
• Flagship product Scottish Longitudinal Study
• Synergies with e-Social Science, data linkage,
  ADLS, secure data etc.
• Always room for methodological work (missing
  data)
• Generating research income
• ESRC application YCS / BHPS Youth latent
  variables
• Top secret GTC/Scot Gov Teachers Panel Study
  (occupation and geography)
• Youth transitions contemporary home and family
  life (ESRC application)
• Growing unregulated markets
• Enhanced knowledge exchange and capacity building
• Training in survey data analysis longitudinal
  methods data management
• Current ESRC Researcher Development Initiative
  call

31
Developing the Longitudinal Studies Centre
Scotland

• An increasingly devolved political climate?
• Scottish data analytical expertise
• Scottish data housing
• The Scottish Essex?
• Institute of Social and Economic Research,
  University of Essex
• UK Longitudinal Studies Centre
• MISOC Research Centre on Micro-social Change
• UK Data Archive