Understanding Trends in Occupational Sex Segregation By Daniel Guinea-Martin Advanced Centre for Scientific Research, Spain (formerly at the Office for National Statistics, UK) Presented at the Research Methods Festival, Oxford University 1 st July 2008
27
Embed
Understanding Trends in Occupational Sex Segregation By Daniel Guinea-Martin Advanced Centre for Scientific Research, Spain (formerly at the Office for.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Understanding Trends in Occupational Sex Segregation
ByDaniel Guinea-Martin
Advanced Centre for Scientific Research, Spain(formerly at the Office for National Statistics, UK)
Presented at the Research Methods Festival, Oxford University
• Substantive scope: To study occupational sex segregation / occupational mobility in the 1990s
• How? Complementing Census-based indices of segregation (Blackwell and Guinea-Martin 2005) with longitudinal research of one cohort born in the late 1950s:
• NCDS
• A similar cohort from the ONS Longitudinal Study
ONS LS 0.81 0.80 0.77 - 0.70 0.72 (no person imp.)
Source: Blackwell, L. and D. Guinea-Martin (2005) ‘Occupational segregation by sex and ethnicity
in England and Wales, 1991 to 2001’, in Labour Market Trends, Vol. 113, No. 12, pp. 501-516
Occupational sex segregation in the NCDS & LS50s
1991 2000/1
Gini ID Gini ID Change
Gini Change
ID
NCDS 0.78
(0.76-0.79) 0.61
(0.59-0.63) 0.77
(0.76-0.79) 0.60
(0.58-0.62) -0.01 -0.01
LS50s 0.76
(0.75-0.77) 0.60
(0.58-0.61) 0.75
(0.74-0.76) 0.58
(0.57-0.59) -0.01 -0.02
Notes: Indices based on the 77 minor groups of SOC90, but coefficients standardized to 200 occupations (using the formulae presented in Blackburn and Jarman, 2005) Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
What are these indices of segregation? Example: The Segregation Curve for NCDS 2000
0
10
20
30
40
50
60
70
80
90
100
0 10 20 30 40 50 60 70 80 90 100
Proportion of women
Pro
po
rtio
n o
f m
en
Integration Segregation
AB
Line of Perfect Integration
Distance=D
Gini=A/(A+B)
Male/female ratio
How to calculate the indices
Index of dissimilarity
Gini coefficient
||21
1
Mi
K
i
Fi ppD
||1
1 1
Mi
Fj
Mj
K
i
K
ij
Fi ppppG
Decomposing segregation: Theil’s index
• Theil’s entropy index of segregation (H) is a multi-group measure of segregation
• Allows the incorporation of various dimensions, for example employment status (full time vs. part time) or age groups
• Currently working on these dimensions to disaggregate the changes in occupational sex segregation during the 1990s
Theil’s entropy index of segregation
‘Entropy’ or diversity
Theil’s index
Decomposition of
Theil’s H
E
EET
t
H
k
ii
i
1
)(
cbcba
cbbcbca
cba
bcacba H
E
EQH
E
EH \
\\
\\
\\
\\\
n
r rr QQE
12
1log
Example: Decomposition of segregation among ISCO-88 occs in UK
Multigroup HComponent 0.298% share 100%
Male/Female HComponent 0.256% share 86%
Women full-time/part-time HComponent 0.044% share 14%
Index of DissimilarityMale/Female ID 0.58
Source: Table 3 in Elliott (2005)
Occupational sex segregation in the NCDS & LS50s
1991 2000/1
Gini ID Gini ID Change
Gini Change
ID
NCDS 0.78
(0.76-0.79) 0.61
(0.59-0.63) 0.77
(0.76-0.79) 0.60
(0.58-0.62) -0.01 -0.01
LS50s 0.76
(0.75-0.77) 0.60
(0.58-0.61) 0.75
(0.74-0.76) 0.58
(0.57-0.59) -0.01 -0.02
Notes: Indices based on the 77 minor groups of SOC90, but coefficients standardized to 200 occupations (using the formulae presented in Blackburn and Jarman, 2005) Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Occupational sex segregation for various quasi-cohorts in the 1990s
Gini coefficientsa Cohort born in 1991b 2001c Diff. 1927-1936 0.82 n/ad n/a 1937-1946 0.82 0.75 -0.07 1947-1956 0.81 0.74 -0.07 1957-1966 0.75 0.73 -0.02 1967-1976 0.72 0.66 -0.06 1977-1986 n/ae 0.60 n/a Overall index 0.78 0.69 -0.09 Notes: People of working age and in employment only. a Coefficients standardised to 200 occupations (using the formulae presented in Blackburn and Jarman 2005) b SOC90 – used 77 minor groups
c SOC2000 – used 81 minor groups
d Part of this quasi-cohort was above working age in 1991
e Part of this quasi-cohort was below working age in 1991 Sources: Author’s analysis. 100 per cent 1991 and 2001 Censuses of England and Wales.
Occupational movers and stayers
NCDS LS50s Men Women All Men Women All Mover 62.9 69 65.5 60.8 67.5 63.5 Stayer 37.1 31 34.5 39.2 32.5 36.5 Total 100 100 100 100 100 100 Notes: People of working age and in employment only. Sources: Author’s analysis. ONS Longitudinal Study and NCDS
Transitions across sex-typed occupations (occupational movers only)
Male transitions across sex-typed occupations (cell percentages)
NCDS LS50s 2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Total Male Integrated Female Total Male 51.9 13.3 3.3 68.5 55.5 12.0 4.1 71.6 Integrated 13.6 8.1 2.3 23.9 11.6 6.6 2.3 20.5 Female 3.5 2.9 1.2 7.6 3.8 2.8 1.4 7.9 Total 68.9 24.3 6.8 100 70.8 21.4 7.8 100 Notes: Data based on occupational movers in employment both years only. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998). Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Female transition across sex-typed occupations (cell percentages)
NCDS LS50s 2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Total Male Integrated Female Total Male 3.3 4.2 4.4 11.8 4.8 4.3 4.7 13.9 Integrated 3.7 11.2 13.6 28.5 4.9 9.0 13.3 27.1 Female 5.5 15.8 38.4 59.7 7.4 14.9 36.7 59.0 Total 12.6 31.1 56.3 100 17.1 28.2 54.7 100 Notes: Data based on occupational movers in employment both years only. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998).
Transitions across sex-typed occupations (occupational movers only)
Male transitions across sex-typed occupations (cell percentages)
NCDS LS50s 2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Total Male Integrated Female Total Male 51.9 13.3 3.3 68.5 55.5 12.0 4.1 71.6 Integrated 13.6 8.1 2.3 23.9 11.6 6.6 2.3 20.5 Female 3.5 2.9 1.2 7.6 3.8 2.8 1.4 7.9 Total 68.9 24.3 6.8 100 70.8 21.4 7.8 100 Notes: Data based on occupational movers in employment both years only. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998). Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Female transition across sex-typed occupations (cell percentages)
NCDS LS50s 2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Total Male Integrated Female Total Male 3.3 4.2 4.4 11.8 4.8 4.3 4.7 13.9 Integrated 3.7 11.2 13.6 28.5 4.9 9.0 13.3 27.1 Female 5.5 15.8 38.4 59.7 7.4 14.9 36.7 59.0 Total 12.6 31.1 56.3 100 17.1 28.2 54.7 100 Notes: Data based on occupational movers in employment both years only. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998).
Transitions across sex-typed occupations (occupational movers only)
Male transitions across sex-typed occupations (cell percentages)
NCDS LS50s 2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Total Male Integrated Female Total Male 51.9 13.3 3.3 68.5 55.5 12.0 4.1 71.6 Integrated 13.6 8.1 2.3 23.9 11.6 6.6 2.3 20.5 Female 3.5 2.9 1.2 7.6 3.8 2.8 1.4 7.9 Total 68.9 24.3 6.8 100 70.8 21.4 7.8 100 Notes: Data based on occupational movers in employment both years only. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998). Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Female transition across sex-typed occupations (cell percentages)
NCDS LS50s 2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Total Male Integrated Female Total Male 3.3 4.2 4.4 11.8 4.8 4.3 4.7 13.9 Integrated 3.7 11.2 13.6 28.5 4.9 9.0 13.3 27.1 Female 5.5 15.8 38.4 59.7 7.4 14.9 36.7 59.0 Total 12.6 31.1 56.3 100 17.1 28.2 54.7 100 Notes: Data based on occupational movers in employment both years only. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998).
Transitions across sex-typed occupations and non-work
Notes: All members of the LS50s longitudinal sample (1991-2001) included. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998). Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Female transitions across sex-typed occupations and non-work (cell percentages)
NCDS LS50s
2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Nonwork Total Marginal diff. Male Integrated Female Nonwork Total Marginal
Notes: All members of the LS50s longitudinal sample (1991-2001) included. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998). Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Male transitions across sex-typed occupations and non-work (cell percentages)
NCDS LS50s
2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Nonwork Total Marginal diff. Male Integrated Female Nonwork Total Marginal
Transitions across sex-typed occupations and non-work
Notes: All members of the LS50s longitudinal sample (1991-2001) included. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998). Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Female transitions across sex-typed occupations and non-work (cell percentages)
NCDS LS50s
2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Nonwork Total Marginal diff. Male Integrated Female Nonwork Total Marginal
Notes: All members of the LS50s longitudinal sample (1991-2001) included. Occupations classified into sex-types on the basis of one lookup table with percentage female derived from one 10 per cent sample of the 1991 Census (Hakim 1998). Sources: Authors’ analysis. ONS Longitudinal Study and NCDS
Male transitions across sex-typed occupations and non-work (cell percentages)
NCDS LS50s
2001 sex types 2001 sex types
1991 Sex types Male Integrated Female Nonwork Total Marginal diff. Male Integrated Female Nonwork Total Marginal