Kevin Ralston PhD Student University of Stirling
Mar 28, 2015
Kevin Ralston
PhD Student
University of Stirling
Fertility in Scotland
Duration (hazard) until First Birth
Total Fertility Rate, Scotland, 1951-2006 www.gro-scotland.gov.uk/files1/stats/annual-report2006
KR 26/08/09
Why fertility?
• Recently there has been academic and political concern about the fall to below replacement fertility levels across many counties.
e.g. the ESRC and Scottish Government funded a Demography Research Programme
http://www.scotland.gov.uk/Topics/Research/by-topic/public-services-and-gvt/Demography-Research-Prog
Why Scotland?... Wha’s like us?
• 2002 was a year of historically low fertility for Scotland, things were looking bad!
• Scotland was the only country in Europe to be undergoing a natural decline in population and was projected to fall below 5million inhabitants.
• Despite the this interest there has still been relatively little research in the field. Key centres for research have been St Andrews University, the CRFR (Centre for Research on Families and Relationships) and in conjunction with the General Registrars Office for Scotland.
Total Fertility Rate, Scotland, 1951-2006 www.gro-scotland.gov.uk/files1/stats/annual-report2006
The importance of geography?
• Studies combining academics and researchers from these institutions have found significant geographical effects related to childbearing.
• For instance in looking at the Scottish Social Attitudes Survey: Family Module, Boyle et al. found that childbearing expectations among childless women apparently varied geographically even when modelling individual level factors.
• Also the Registrar General’s Annual Report for 2002 argued that different demographic regimes exist within Scotland -see below- (GROS Report 2002)
City Cores Standardised Births per 1000 population, 2001-2005
Age Specific Fertility Rate per 1000 population, age 30-34, 2001-2005
Edinburgh 8.6 82
Aberdeen 9.0 84
Dundee 9.8 71
Glasgow 9.5 74
Scotland 10.4 87
Commuter Belts Standardised Births per 1000 population, 2001-2005
Age Specific Fertility Rate per 1000 population, age 30-34, 2001-2005
East Renfrewshire 11.9 132
Aberdeenshire 11.8 101
East Lothian 12.1 99
Midlothian 11.8 97
West Lothian 11.9 91
Angus 11.9 92
Perth and Kinross 11.5 101
Scotland 10.4 87
Rural Areas Standardised Births per 1000 population, 2001-2005
Age Specific Fertility Rate per 1000 population, age 30-34, 2001-2005
Highland 12.1 87
Moray 11.8 81
Argyll & Bute 11.2 88
Orkney 11.1 84
Dumfries & Galloway 11.7 84
Scottish Borders 11.7 99
Eilean Siar 11.4 101
Scotland 10.4 87
The importance of geography?
• Geographers and the GROS present evidence as to the spatial variation of childbearing.
• However elsewhere the primary importance of other factors is highlighted.
Class/ stratification (men) Ekert Jaffe et al. (2002)Educational Attainment (women) Simpson et al. (2006)
• Generally empirical research suggests that measures of stratification are the best indicators of childbearing outcomes for men. Whilst, measures of education/ attainment are the best for women (see references above).
Some Theory
• Irwin (2000) argues that we should utilise the concept reproductive regime when thinking about childbearing outcomes. This refers to the central childbearing relationship between a man and a women located within wider social relations and structures.
• This theoretically links mirco-level action to macro level social patterns. The reproductive regime is the individual level location of childbearing, where social action takes place that results in macro outcomes. This concept has been central to all my thinking.
Some more theory…
• The concept of the reproductive regime links well with wider sociological theory. Micro social theorists such as Erving Goffman (1959) and Berger and Luckmann (1966) argue that people gain socially from living accepted social norms.
• A thesis that I have been working with for several years: If the social benefits of having children are higher we will have more children. If the social costs of having children are higher we will have fewer children.
• Empirically measures of stratification/ attainment indicate different levels of childbearing and fertility. This would represent differing reproductive regimes being lived by people, as an outcome of relative advantage.
My Research…so far!
• Using the Scottish Longitudinal Study (SLS)I have been setting up an event history analysis of first birth.
• The SLS is a dataset based on Census data and provides a 5% sample of Scotland. It includes, for example, measures of stratification, attainment, geography and vital events data.The data is structured around the census’ of 1991 and 2001 with individuals being picked up between these time points in vital events data.
Event History Analysis?
• There is an established tradition of the use of Event History Analyses when looking at fertility (Steele et al. 2005, Miranda 2006, McDonald & Rosina 2001)
• Time is of substantive interest in fertility research. We are able to model processes of relationship formation and childbearing and the effects of time to and between events. It is then possible to include explanatory covariates and their effects.
• The SLS is appropriate for an event history analysis. I have an event (first birth) and I know time to the event and have a number of explanatory covariates.
• I have all events (i.e. first births to SLS members between 1991 and 2006). This is around 75,000 events. So about 5000 a year.
The importance of Geography?• Methodology: Cox Proportional hazard model to assess
the effects of geography, attainment and stratification on the hazard of first birth.
• How does occupational status affect hazard of first birth in Scotland?How does (educational) attainment affect hazard of first birth in Scotland?
Hypotheses:• Regional differences in occupational status and
attainment accounts for variation in hazard of first birth.• People of the same occupational status and attainment
have different hazard of first birth by region.
Models• The risk set here is women aged 15-25 in 1991
• Number of observations = 33257
• Number of births = 21692
(Number of first births to sample between 1991 and 2006)
* p= 0.05 **p=0.01 ***p=0.001
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Single 91
Married 91 .501*** .448*** .549*** .599*** .478*** .445*** .418***
Remarried 91 .640***
.518*** .602*** .719*** .598*** .511*** .596***
Divorced 91 .373*** .267*** .396*** .458*** .346*** .260*** .254***
Widowed 91 -.786 -.915 -.850 -.896 -.775 -.900 -.959
Aged 15-19 .335*** .0843*** .157*** .201*** .310*** .0856*** .132***
Aged 20-25
Professional 91
Managerial Technical 1991 .489***
.457*** .339***
Skilled non-manual 1991 .452***
.414*** .219***
Skilled Manual 91 .717*** .669*** .486***
Partly Skilled 91 .531*** .485*** .302***
Unskilled 91 .444*** .403*** .233***
Degree or Higher 91
Further Qualification 91 .728*** .715*** .512***
No Higher or Further Qualifications 91 .870*** .856*** .711***
Not Stated 91 .774*** .761*** .693***
Not Completed U-18, in 1991 .888*** .873*** .859***
City 91
Town 91 .282*** .168*** .054*** .046***
Small Accessible 91 .254*** .150*** .025 .025
Small Remote 91 .323*** .201*** .089* .081
Accessible Rural 91 .193*** .090*** -.032 -.029
Remote Rural 91 .176*** .083** -.050 -.047
None
Catholic .219*** .298***
Church of Scotland .213*** .210***
Non-Christian .077 .545***
Other-Christian -.190*** .219***
Another-Religion -.412* -.249
Findings & the Future
• I need to check this analysis further» Multiple Births» More descriptive analysis
• The Importance of Geography?» Try other measures
• The nature of reproductive regimes in Scotland?
» Time varying effects. Education. Marriage» More complex models
Berger, P. and Luckmann, T. (1966) The Social Construction of Reality: A Treatise in the Sociology of Knowledge, Penguin: Harmondswoth.
Boyle, P. Graham, E. & Feng, Z. (2007) ‘Fertility Variations in Scotland: geographical influences’ CRFR Research Briefing, Edinburgh, Centre for Research on Families and Relationships.
Ekert-Jaffe, O. Joshi, H., Lynch, K. Mougin, R. Rendell, M., & Shapiro, D. (English Edition, 2002-) ‘Fertility, Timing of Births and Socio-economic Status in France and Britain: Social Policies and Occupational Polarization’ Population. 57, No3. pp. 475-507.
Goffman, E. (1959) The Presentation of Self in Everyday Life, Penguin: Harmondsworth.
Irwin, S. (2000) ‘Reproductive Regimes: Changing Relations of Inter-dependence and Fertility Change’ in Sociological Research Online, 5(1):<http://www.socresonline.org.uk/5/1/irwin.html>.
McDonald, J. W. and Rosina, A. (2001) ‘Mixture modelling of recurrent event times with long-term survivors: Analysis of Hutterite birth intervals, Statistical Methods and Applications, (2001) 10: pp. 257-272
Miranda, A. (2006) ‘Are young cohorts of women delaying first birth in Mexico?’, Journal of Population Economics, 19: 55-70
Simpson, R. Morton, S. and Wasoff, F. (2006) ‘Childbearing on Hold: delayed childbearing and childlessness in Britain’
Steele, F. Constantinos, K. Goldstein, H. Joshi, H. (2005) ‘The Relationship between Childbearing and Transitions from Marriage and Cohabitation in Britain, Demography, Vol. 42. No.4 (Nov., 2005) pp.647-673
CRFR Research Briefing. Edinburgh, Centre for Research on Families and Relationships. GRO Scotland (2002) General Registrars Office Scotland, Annual Report, 2002-2006, <http://www.gro-scotland.gov.uk/statistics/library/annrep/02annual-report>; accessed 10/12/06.
http://www.scotland.gov.uk/Topics/Research/by-topic/public-services-and-gvt/Demography-Research-Prog
References
• Thanks to students and staff at the University of Stirling
• Also, the ESRC, the organisation funding my PhD and research
• The help provided by staff of the Longitudinal Studies Centre - Scotland (LSCS) is acknowledged. The LSCS is supported by the ESRC/JISC, the Scottish Funding Council, the Chief Scientist's Office and the Scottish Government. The authors are responsible for the interpretation of the data. Census output is Crown copyright and is reproduced with the permission of
the Controller of HMSO and the Queen's Printer for Scotland
Acknowledgements: