Longitudinal studies and the life course abstracts

Strand organisers: Dr. Myles Gould, University of Leeds & Dr. James Robards, University of Southampton 

Longitudinal studies: Analytical approaches to longitudinal studies - Tuesday 8 September, 1.45pm 

The state of the art of longitudinal data analysis in population studies?
Myles Gould 1, James Robards 2  1University of Leeds, 2University of Southampton 

There has been an explosion of longitudinal data for population and health studies over the past 40 years. At present this trend continues with the advent of administrative and other ‘big data’ where response variables are repeatedly and dynamically measured. Whilst there is often appreciation of the complexity of data structures, researchers sometimes use techniques and approaches with which they are most familiar and experienced in using. This paper charts the application of longitudinal data analysis methods and techniques in demography, population studies and related disciplines (epidemiology, health studies and applied statistics) over time. The paper goes on to highlight ‘state of art’ and ‘lesser used’ methods which may assist demographers in answering pertinent research questions using new data becoming available. In so doing, this will provide context for rest of the conference session which promotes innovative approaches and novel quantitative techniques for the analysis of longitudinal / panel and cohort data and will hopefully provoke discussion and sharing of experience, whilst also encouraging people to try new approaches. We will illustrate the presentation drawing on our own work using longitudinal data including event history analysis and census data, multilevel modelling frameworks as example approaches; whilst also presenting a typology of different datasets and analysis tools that have been used to tackle different research questions/problems. 

E-mail: m.i.gould@leeds.ac.uk 

How does social care need amongst older adults change over time in the UK? Amos Channon, Joe Viana, Sally Brailsford, Stuart Rossiter, University of Southampton 

The provision of state funded social care for older adults in the UK is driven by the concept of need. Individuals who have care requirements that surpass a threshold are eligible to receive care. A wide range of methods are used to ascertain need, but the ability to conduct simple activities of daily living (ADL), such as bathing and dressing, is often used. However little is known how the abilities to conduct these activities change over time, alongside the factors that are associated with changes in care need. This is of clear interest to those providing care services to older adults. Using all five available waves of the English Longitudinal Survey (between 2002 and 2012) the care need for individuals over the age of 50 was calculated for each time period. ADLs were used to calculate need, which was divided into none, low, intermediate and high. Multistate life tables were used to estimate the probabilities of transitions between care need states between survey waves, as well as between care need state and death. These life tables were calculated for the whole survey, before calculating them separately by gender, age groups and smoking status and combinations of these factors. Results indicate that there is a substantial variation of care need between waves, with many individuals recovering between waves. Recovery is less likely amongst males, the old and those who smoke. This indicates that assessments of need should be conducted on a regular basis in order to ensure the appropriate level of care is provided when needed. 

E-mail: a.r.channon@soton.ac.uk 

Variation in the intersection between partnership and fertility: A comparison across three cohorts in 15 countries
Mark Lyons-Amos, UCL Institute of Education 

The intersection between fertility and partnership is increasingly complicated within Europe and the United States. The variety of union types has increased from a homogenous pattern of direct marriage to include cohabitation before marriage, short term cohabitation and lifelong cohabiting relationships. The intersection between union and childbearing behaviour has also shifted: marriage is no longer a prerequisite for childbearing, which increasingly occurs within cohabitation. Children are more likely to see their parents separate; although childbearing unions of all types are generally more stable than childless ones. The extent to which these trends have emerged has depended on country context. As such, the evolution of the intersection between fertility and partnership varies across national borders both in terms of the type of patterns exhibited and the rapidity of change. This paper uses multichannel sequence analysis to evaluate the relationship between partnership and fertility across the US and Europe. Multichannel sequence analysis is a longitudinal technique which allows multiple processes to be analysed simultaneously and clustered into similar groups. The main aims of the paper are to establish how the association between partnership patterns and the timing of births varies between countries and across birth cohorts, by examining the prevalence of fertility-partnership groupings across three cohorts (1945-54, 1955-64, 1965-74). I analyse retrospective union and fertility histories from the Harmonized Histories database for women age 15-45. Results suggest that the relationship between partnership and birth timing have shifted. However, the nature of this shift varies by country reflecting heterogeneous evolution of partnership-fertility interaction. 

E-mail: M.Lyons-Amos@ioe.ac.uk 

Longitudinal studies & the life course: Wednesday 9 September 9.00am 

Mismatches and matches in address information from the Census and the BSO: A longitudinal perspective 2001-2011
Brian Foley, Ian Shuttleworth, School of Geography, Archaeology and Palaeoecology, Queen’s University Belfast 

Accurate address information from health service providers is fundamental for the effective delivery of health care and population monitoring and screening. While it is currently used in the production of key statistics such as internal migration estimates, it will become even more important over time with the 2021 Census of UK constituent countries integrating administrative data to enhance the quality of statistical outputs. Therefore, it is beneficial to improve understanding of the accuracy of address information held by health service providers and factors that influence this. This paper builds upon previous research on the social geography of address mismatch between census and health service records in Northern Ireland. It is based on the Northern Ireland Longitudinal Study; this is a large data linkage study including about 28 per cent of the Northern Ireland population, which is matched between the census (2001, 2011) and healthcard register maintained by the Health and Social Care Business Service Organisation (BSO). This research compares address information from the Spring 2011 BSO download (Unique Property Reference Number, Super Output Area) with comparable geographic information from the 2011 Census. Multivariate and multilevel analyses are used to assess the individual and ecological determinants of match/mismatch between geographical information in both data sources to determine if the characteristics of the associated people and places are the same as the position observed in 2001. It is important to understand if the same people are being inaccurately geographically referenced in both Census years or if the situation is more variable. 

E-mail: b.foley@qub.ac.uk 

Relationships between time since entering the UK, languages spoken and proficiency in English
Oliver Duke-Williams, Department of Information Studies, UCL 

This paper reports the findings of an LS Beta test project, which used the ONS Longitudinal Study to explore some of the new variables in the 2011 Census. These included questions directed at those born outside the UK, such as the year and month of entry to the UK and the length of intended stay in the UK; and questions about main language used and language proficiency. Comparisons of the results of the question on year of entry to the UK can be made with other sources to test reliability and possible recall error: entry to the NHS, one-year migration questions in each separate Census, and a similar question asked in 1971. These are well correlated suggesting that the question response was robust. Analysis on the month of entry shows clustering in more recent years that may be related to student migrants. Whilst a relatively small set of languages provide the main language used by a large majority of respondents, there is a long tail of other languages also used. Combination of the main language question with the year of entry results shows clear patterns of different periods of entry by different language speakers. Indication of a 'main' language other than English does not necessarily mean that English proficiency is poor; one aspect that is of interest that is explored by this paper is the case of households where no adults use English as a main language, but the household contains at least one child that is an English speaker. 

E-mail: o.duke-williams@ucl.ac.uk 

Birth order and educational decision-making: Fields of study and expected outcomes
Kieron Barclay, Martin Hällsten, Mikko Myrskylä, London School of Economics 

This study investigates the relationship between birth order and preferences regarding educational field, measured by choices made about the field of study pursued at university. A large body of research has investigated the relationship between birth order and educational attainment. Research that has used fixed effects to perform a within-family comparison has consistently shown that, relative to first borns, later born siblings have a lower grade point average in school, a lower likelihood of making subsequent educational transitions, and lower completed educational attainment. However, it is not well understood whether first borns are more likely to enter fields of study that yield greater career opportunities or higher earnings potential than later borns. In this study we use complete individual-level Swedish population register data on individuals born between 1955 and 1977. We have data on the university courses that individuals are accepted to and enter, as well as university applications, giving us information on individual preferences about field of study. Our results show that first borns are more likely to enter courses for engineering and medical training, while later borns are more likely to study nursing, business degrees, and arts programs. Overall our results show that first borns are more likely to study subjects that have higher average earnings amongst university graduates. Our results complement the previous body of research on birth order and educational attainment and extend it by examining how greater attainment may be more or less valuable to future opportunities according to the field that is pursued. 

E-mail: k.j.barclay@lse.ac.uk 

Fertility, health and education of immigrants in England and Wales: The role of English language skills
Lualhati Santiago 1, Yu Aoki 2  1Public Policy Division, Social and Analysis Directorate, Office for National Statistics, 2Department of Economics, Business School, University of Aberdeen 

This paper aims to identify the causal effects of English language skills on fertility, health and education outcomes of immigrants in England and Wales. To identify the causal effects, we use the instrumental variable estimation strategy where age at arrival in the United Kingdom (UK) is exploited to construct an instrument for language skills. The idea of exploiting age at arrival is based on the phenomenon that a person who is exposed to a new language within the critical period of language acquisition (i.e., childhood) learns the language easily. This implies that immigrants who arrive in the UK at a young age will have on average better English language skills than those who arrive when they are older. Using a unique individual-level dataset that links census and life event records for the population living in England and Wales at the 2011 Census, we find that better English language skills significantly delay the age at which women have their first child, lower the likelihood of becoming a teenage mother, decrease the number of children a woman has, but do not affect child’s birthweight and self-reported health. The impact on educational achievement is also considerable: better English skills significantly raise the probability of obtaining post-compulsory qualifications and academic degrees and significantly lower the probability of having no qualifications or only compulsory level qualifications. 

E-mail: lualhati.santiago@onsgsi..gov.uk 

Health outcomes for informal carers in England and Wales – A longitudinal study using nationally representative census data for England and Wales
Athina Vlachantoni 1,2,3, James Robards 1, Maria Evandrou 1,2,3, Jane Falkingham 1,2   1EPSRC Care Life Cycle, Social Sciences, University of Southampton  2ESRC Centre for Population Change, Social Sciences, University of Southampton  3Centre for Research on Ageing, Social Sciences, University of Southampton 

Cross-sectional analyses of 2001 and 2011 Census data suggest an association between worse self-reported health and increasing hours of informal (unpaid) care. Research on health among informal carers has identified caring and health interactions as often being more positive than portrayed leading to arguments that studies require a more nuanced approach. The current study asks ‘how does poorer self-rated health at the 2011 Census relate to carer roles past (2001), present (2011) and transitions between 2001 and 2011?’ Using Office for National Statistics Longitudinal Study members at 2001 and 2011 we identify transitions in caring roles (2001-2011) and use baseline health status (2001) as a control. A binary logistic regression model is specified where the outcome measure is bad or very bad health at the 2011 Census. Results show (1) those providing informal care at 2011 exhibit lower odds of poor health than those who did not provide care in both 2001 and 2011 and (2) light informal carers (1-19 hours of care per week) at 2001 no longer caring at 2011 show no difference in odds of reporting poor health compared to non-carers (2001 and 2011). Results for heavy informal carers (20+ hours of care per week) show (1) heavy informal carers at 2001 not caring at 2011 show higher odds of reporting poor health compared to non-carers (2001 and 2011) and (2) heavy carers at 2001 and 2011 are less likely to report poor health at 2011 compared to non-carers (2001 and 2011). 

E-mail: james.robards@soton.ac.uk