Poster abstracts


Posters alphabetically by first author 

Fear of female promiscuity: how partner’s perception impacts on family planning demand in Nigeria
Abiodun Idowu Adanikin 1, Sabu Padmadas 1, Nuala McGrath 1, 2, 1 Department of Social Statistics and Demography, University of Southampton, 2 Primary Care and Population Studies Academic Unit, Faculty of Medicine, University of Southampton 

Background: Evidence from the last three Demographic and Health Surveys (DHS) in Nigeria shows slow progress in increasing family planning use despite programmatic interventions. To unravel the cause, there has been a gradual shift in programme focus towards the role of men in family planning and the underlying social behaviours. A recurring theme from previous male-centred studies is the fear of promiscuity when women use contraception. However, little is known about how this perception impacts family planning demand. We hypothesise that men’s fear of female promiscuity deters demand for family planning.

Methods: Using the couple dataset from the 2013 Nigerian DHS, we examine the effect of men’s fear of promiscuity on the odds of family planning demand among fecund couples (n=7747), controlling for other socio-behavioural and demographic factors.

Results: About 40% of men held the perception that women who use contraception becomes promiscuous. The perception was most prominent among older men, those with no formal education, rural residents, those from the Northwest region and Muslims. Results from multivariate logistic analysis show that men’s fear of female promiscuity was associated with lower odds of family planning demand (OR: 0.86; 95% CI: 0.76-0.97), after adjusting for the partner’s and woman’s characteristics. The perception and demand for family planning varied considerably by region of residence.

Conclusion: Policy and program efforts targeted at changing men’s perception about family planning may yield success in improving contraceptive uptake in Nigeria.

Exploiting Routine Health Facility Data to Estimate Small Area Indicators for Sub-National Healthcare Decision making
Seth Kwaku Afagbedzi 1, Allan G. Hill 2, James A. Wright 2, 1 University of Ghana, School of Public Health, 2 University of Southampton

The health sectors of most countries in sub-Saharan Africa with high disease incidence collect huge amount of health facility data routinely and store them in databases. These dataset investments contain mortality, morbidity and other key information on common diseases that affect the population of the catchment areas of the health facilities. Unfortunately, these data are seldom used to provide evidence for local authority health activity planning and decision making. The principal objective of this paper/poster is to demonstrate the use of routine health facility data in estimating small area (catchment-level) indicators of disease incidence to inform local health policy decision-making. The study uses 2011 routine morbidity data stored in a national database (District Health Information Management System 2), collected from healthcare facilities in Greater Accra Region of Ghana. Case study diseases included malaria, diarrheal disease, anaemia, pneumonia and acute lower respiratory infection, drawn from the top ten causes of morbidity and mortality in the region. Public and private health facility locations have been geo-coded and spatially linked to enumeration area boundaries from the 2010 population census to assess consultation patterns and ultimately provide denominators for catchment-level incidence. Preliminary analysis of malaria morbidity reported via health facilities suggests similar age and sex incidence distributions, comparable to results from international and national surveys such as Demographic and Health Surveys and Ghana Living Standard Survey respectively.

The social and economic impact of international female labour migration on left-behind parents in East Java, Indonesia
Faishal Aminuddin1, , Saseendran Pallikadavath2, Amie Kamanda2,  Keppi Sukesi1, Henny Rosalinda1, Ngianga II Kandala2, Kieron Hatton3, 1Portsmouth-Brawijaya Centre for Global Health, Population and Policy, University of Brawijaya, 2 Portsmouth-Brawijaya Centre for Global Health, Population and Policy, 3University of Portsmouth, School of Health Sciences and Social Work 

Context: International labour migration from Indonesia is dominated by young females, who have traditionally been the carers of the parents, especially in rural areas. Previous studies documented the contribution of migration of daughters/daughters-in-law to the reduced social status, increased neglect, and poor mental health of the left-behind parents/elderly. This paper, for the first time, examined the socio-economic impact on the parents while their unmarried and married daughters/daughters-in-law were working aboard. The paper takes a holistic perspective while considering the impact within the wider social, economic and political context. Data and methods: Data from a household census (178 households) and in-depth interviews (71 left-behind members) conducted during two months fieldwork in two villages in East Java in 2015 was used in this paper. The qualitative data were analysed for pre-determined themes as well as those that emerged during analysis. Census data were analysed using descriptive statistics. Results and conclusions: Unmarried daughters helped parents by contributing to the family economy, where the parents were the direct beneficiaries. Remittances from the unmarried daughters were mainly saved for their marriages. While parents were concerned about the safety of their unmarried daughters, the concern with regard to married daughters were about children, and marital dissolution. Grandmothers provided home-based care and grandfathers provided support with schooling and health care. Grandparents experienced stress if there is inadequate communication and financial support from daughters/daughters-in-law for the children left-behind. International female migration had resulted in a reversal of care where parents were tasked with more responsibilities which traditionally were carried out by daughters. 

The association between alcohol consumption and the timing of union formation in a British cohort 
Giacomo Arrighini, University of Oxford 

This paper identifies the association between the recreational consumption of alcohol in early life and the timing of union formation using a large, representative cohort of the British population. Despite the vast amount of research available on how risky health behaviours are affected by entering a union, comparatively little has been done to identify the consequences of alcohol consumption on the selection out of singlehood, into cohabitation and marriage. The literature provides informative but conflicting glances on the topic: across different samples, cohorts, and countries drinking – captured in variety of ways – has been linked to early marriage, to late marriage, and to no impact on the timing of marriage. Different ways of framing cohabiting union further complicate the picture. Using a longstanding, prospective cohort of the British population, this study establishes the association between the consumption of alcohol and the timing of first exit from singlehood. The use of multi-state event-history models let us identify how alcohol consumption is associated to the transition from singlehood to first cohabitation, from singlehood to first marriage, and from first cohabitation to first marriage. Our results suggest that the associations between alcohol consumption and the three transitions considered (a) have a magnitude that justifies further attention, (b) are statistically significant, and (c) are based on gender. The analyses performed further suggest that the association cannot be explained away by – among the others – parental social class, intention to pursue a tertiary education, frequency of participation in social activities, reported ideal timing for marriage, and smoking.

Trends in Living Arrangements of Older Adults in Africa, 1976-2011
Paul K Ayernor, University of Oxford   

This study analyzes trends in living arrangement of persons aged 50 years and over in selected African countries between 1976 and 2011. After a review of the previous evidence, we present a conceptual framework that highlights how demographic transition, economic development, and urbanization combine to underpin the nuclearization of household structure in Africa. We use the Integrated Public Use Microdata Series (IPUMS) of two or three consecutive censuses for each country. The IPUMS census-based micro-sample data allows us to identify five types of living arrangements: (1) living alone (2) living with spouse only (3) couple with children (4) single-parent family (5) extended family and others. We find a steady decline in the proportion of older adults who co-reside with traditional extended family in several countries, and a significant increase in the percentage of older adults living alone in Ghana and South Africa. We then examine the variation of living arrangement by age, sex, education, place of residence, and disability status. The findings suggest variations based on the level of education and rate of urbanization of the country. When older adults have gained in human capital through education, they are more likely live independently rather than in an extended household living arrangements. We see this as an indication of the shift toward nuclearization of the family and the weakening of extended familial ties.

Changing partnership and fertility in Europe Insights from a microsimulation model for Italy, Norway and Great Britain
Eva Beaujouan 1, Maria Winkler-Dworak 1, Paola DiGiulio 1, Martin Spielauer 2, 1 Wittgenstein Centre for Demography and Global Human Capital, Vienna Institute of Demography/Austrian Academy of Sciences, 2 Independent Researcher 

Family dynamics are changing in Europe, but how cohort completed fertility is affected by partnership behaviours and how this has changed over time is rarely studied. We use microsimulation techniques to investigate the effect of the increasing prevalence of union dissolution on completed fertility levels in Italy, Norway and Britain, three countries with very different systems of value. We find that the net effect of union instability is to decrease fertility (by about 0.5 children for Italian and 0.2 to 0.4 children for British cohorts, explorations are ongoing for Norway) but the magnitude of the difference depends on the timing of union formation and separation. As expected, re-partnering produces more children in new partnerships if the separation occurs earlier. Nonetheless, it is only if separation takes place after the second birth and if all women re-partner that additional childbearing would almost compensate for births lost due to union disruption. 

Ethnic Differences in Returning Home: Explanations from a Life Course Perspective
Ann Berrington 1, Tom Kleinepier 2, Lenny Stoeldraijer 3, 1 University of Southampton, 2 Delft University of Technology, Faculty of Architecture and the Built Environment, 3 Statistics Netherlands   

Ethnic differences in leaving and returning home may reflect varying cultural norms regarding intergenerational coresidence, but also differences in transitions in linked domains, for example employment and partnership transitions. This study uses data from the Dutch population registers to compare returning home among second-generation Turks, Moroccans, Surinamese, and Antilleans with native Dutch who had left the parental home between age 16 and 28 in the period 1999-2011 (N=194,020). All second-generation groups were found to be more likely to return home than the native Dutch. A large part of these differences was attributable to the timing and occurrence of other key events in the life course, such as age at leaving home and partnership dissolution. The impact of partnership dissolution on returning home was found to be weaker for the Turkish and Moroccan second generation, lending support to notions of parental disapproval of cohabitation and divorce among these origin groups.

Austerity, familialism, and conceptualisations as to the age at which adulthood is reached 
Ann Berrington 1, Adriana Duta 2, 1 University of Southampton, 2 University of Edinburgh 

This paper explores the meaning of “adulthood”, and to examine whether there are increasing inequalities in trajectories to adulthood in the context of austerity, cuts to welfare benefits, and increasing reliance upon parental support for young adults in the UK. The paper begins by reviewing changes over the past decade in the way that “becoming an adult” is identified and conceptualised in education, social and housing policies. It becomes clear that there is considerable inconsistency across policy domains in the extent to which parents are presumed to be responsible for their adult children’s life chances. For example, in terms of supporting young adults through post-compulsory education, parental support is presumed, whereas in terms of out of work benefits, no reference is made to parental income or support is made in judgements of need. In terms of housing support, those who are and are not co-resident with their parents have always been treated differently, and recently, the age at which young adults are deemed as ready to leave the parental home has been increased. Using empirical evidence from large nationally representative data sources such as the Labour Force Survey which includes information on parental socio-economic background, the paper looks at timing of leaving full time education, gaining full time employment, moving out of the parental home, and home ownership according to parental socio-economic status. The paper concludes that the recent shift to familialism in the context of insecure labour markets and unaffordable housing has increased intra-generational inequalities for young adults.

Women’s employment and child nutritional status– The case of India
Stephanie Bispo Andrew Amos Channon, University of Southampton

Over the last few decades India started to experience great economic growth. However, levels of child undernutrition did not decrease, and are still among the highest in the world. Economic growth normally comes accompanied by increased employment opportunities, especially for women who have children. Previous studies suggest that female employment is linked to better child health and nutrition, but this association is not fully understood, especially within the context of economic growth, nutrition transition and considering cultural constraints. This study aims to examine the association between a range of aspects of female employment and child nutritional status. Data was obtained from the National Family Health Survey (NFHS) from 1998 and 2005. Children were classified as stunted or wasted according to the WHO Growth Curves. Multilevel logistic regression was performed using STATA 13. From 30% of mothers employed in both years, more than half were classified with some form of malnutrition, with significant difference from mothers who are not working for both years. When controlled by socio-demographic characteristics, being employed or not was no longer associated to child nutritional status. The only characteristics of the mother which remained in the model were maternal education and mother’s BMI. Differences according to regions were identified when considering the percentage of employed women and women empowerment in each place. This study suggests that working mothers can be an important target for policies, and improving maternity leave and gender inequality can have an impact in nutrition for future generations in India and other emerging countries.

Impact of devolution to Scotland on data for policy making
Jenny Boag, Falkirk Council   

Devolution has gone much further in Scotland than in the rest of the UK, at least down to a regional level, and that can be seen in the way in which the Scottish Government works in collecting and publishing statistics, in the way in which National Records for Scotland works and in the requirements and capacity of local government. I would want to discuss where Scotland is doing better than in the rest of the UK and has been able to be more responsive to users of data, for example, in household projections, small area population projections and in the Census. However, there have been issues, most notably the delay in publishing Census data, lack of capacity among statisticians both in the Scottish Government and in local authorities. There remains a question mark over whether we really do use evidence for policy making.

The implication of the modern decline in marriage for taxes, benefits and entitlements
Patrick Carroll, Jean Utshudiema, PAPRI   

A second demographic transition characterised by a decline in marriage has been noted in developed and developing countries. This has an impact on government revenues and in government provided benefits. The implications for social security and family related benefits are examined with international comparisons. Most adults in the UK are not now married. The decline in marriage is tracked within the UK and other countries. Within the UK, comparisons are made between England and Wales, Scotland and Northern Ireland. The growth of single parenting is examined with attention to the numbers of children in age groups according to different family circumstances. Trends in expenditure on family related benefits are compared within these countries. Income tax is usually determined in relation to marital status. The implications for revenue from this tax resulting from the decline in marriage, is to be examined with comparison between the UK and other European countries. In the UK and Ireland there are universal benefits such as Child Benefit and also means tested benefits such as Child Credit. Income tax measures such as Child Tax Allowances as in Ireland are also considered. In France, there are particular measures to assist larger families with tax concessions and a variety of benefits. Trends and patterns in the cost of such benefits are compared in the UK and Ireland and France. Survivors benefits paid within National Insurance are paid with reference to marital status. The implications of the decline in marriage are considered.

An Introduction to the Scottish Longitudinal Study (SLS)
Susan Carsley, Lee Williamson, Fiona Cox, National Records of Scotland   

This poster will introduce the SLS and the datasets, the application process for researchers interested in using the SLS and outline research examples. The Longitudinal Studies Centre – Scotland (LSCS) was established in 2001 and hosts the Scottish Longitudinal Study (SLS). This study links together routinely collected administrative data for a 5.3% representative sample of the Scottish population (about 270,000 people). It currently includes a wealth of information from the censuses starting in 1991, vital events registrations (births, deaths and marriages), Scottish education data, and with appropriate permissions can be linked to NHS health data including cancer registry and hospital admission data. The size and scope of the SLS make it an unparalleled resource in Scotland for analysing a range of socio-economic, demographic and health questions. Additionally, the longitudinal nature of the SLS is particularly valuable, allowing an exploration of causality in a way that cross-sectional data collected at a single point in time does not. In this way, the SLS can provide insights into the health and social status of the Scottish population and, crucially, how it changes over time. The 2016 BSPS conference presents an excellent opportunity to highlight the data that is available and will help researchers decide whether the SLS is an appropriate resource for their research.

Levels and patterns of child stunting and underweight in Malawi: 2000 to 2010
Lana Clara Chikhungu 1, Nyovani Janet Madise 2, Sabu Padmadas 2, 1 University of Portsmouth, 2 University of Southampton   

Background Malawi is one of the countries that are worst affected by child under nutrition. Recent estimates reveal that among children under the age of five, 42.8% are stunted, 16.7% are underweight and 3.8% are wasted. This paper examines the levels and patterns of child stunting and under-weight in Malawi between the years 2000 and 2010 based on the 2006 WHO Growth standards. Methods Using the Malawi Demographic and Health Survey data sets of 2000, 2004 and 2010, chi square tests were carried out to investigate changes in the levels of child stunting and child underweight in Malawi between 2000 and 2010 and multilevel logistic regression analysis were employed to identify factors that were consistently associated with child stunting and underweight. Results: Stunting and underweight levels declined significantly from 54.1% and 21.4% (2000) to 47.1% and 12.7% (2010) respectively.

Explaining Fertility Decision-Making in Romania: Parity specific intentions and their determinants
Alexandra Andreea Ciritel, Max Plank Institute for Demographic Research & La Sapienza Universita di Roma,  Southampton University

This paper examines the parity-progression fertility intentions within the Romanian low-fertility context, using the Theory of Planned Behavior (Fishbein & Ajzen, 2010). Within the theoretical framework of planned behavior, I analyze how attitudes, subjective norms, and perceived behavioral control play in crystallizing fertility intentions among childless and one-child parents, using the 2005 wave of Generation and Gender Survey (GGS). Principal Axis Factor analysis confirms what items proposed by GGS, 2005 wave, act as valid and reliable measures of the suggested theoretical socio-psychological factors. Three logistic regression models compare the determinants of childbearing intentions among the childless and one-child parents. Objective measures of the socio-economic condition of the respondents, together with demographic variables are introduced step-wise to the first model, which is the reduced model, containing only the sociopsychological variables. Average marginal effects on the probability of having a child within the next three years are presented to facilitate comparisons across the two sub-samples and among the models. The aim is to observe how the psychological variables behave in the presence of other covariates and how these, in turn, influence childbearing intentions. Across all three models, for both sub-samples, the social pressure remains the strongest psychological variable which influences the childbearing intentions. Positive attitudes play a higher role as a childbearing predictor for parents than for childless, across all models. The result underlines the positive experience as a parent which influences the transition to the second child. Being employed counts as a predictor just for the childless group, while living in the poorest region of Romania, namely in the North-East development region, as compared with living in the Bucharest-Ilfov region (the capital region) increases the probability of planning the second child. The childless who belong to the age group of 30-35 are most likely to have a child compared with the other age categories. This supports the idea that delaying parenthood is a feature which recently embarked Romania on the journey of late starters of the Second Demographic Transition.

Impact of special populations on population estimates at data zones level within Scotland
Maya Clayton, National Records of Scotland   

This poster examines the issues of special populations and the effect they have on small area population estimates published annually by the National Records of Scotland (NRS). Different types of special populations are usually clustered within one or a few neighbouring data zones in the same area. These will be the areas with high proportion of students, those with high seasonal migration, those with hospitals which see a high number of non-Scottish residents’ births and death, those with high armed forces population as well as areas with other communal establishments. The population estimates include the use administrative data and are published at data zone level, a statistical geography which nests within local authorities’ boundaries and covers the whole of Scotland. The 6,976 data zones are built up from 2011 Census output areas and include populations of between 500 and 1,000 household residents. Due to a relatively small population dimension any significant changes brought about by special populations within specific data zones might have a considerable effect on overall estimates for these data zones and thus impact any planning or projections produced by the users of data such as local authorities. The poster highlights issues of identification of data zones that might require specific additional adjustments, and subsequently discusses the process of appropriate adjustment procedures for these areas.

Occupational Patterns of Asian Birthplace Groups in Australia and their Ancestry Subgroups
Sheruni De Alwis, Nick Parr*, Macquarie University

Asia-born migrants, who mostly enter Australia as skilled migrants, contributed half of permanent settler movements into Australia in 2014–15. Most of the larger Asian birthplace groups are ethnically heterogeneous. However, existing literature on the occupational outcomes of migrants in Australia has tended to overlook the diversity exhibited between and within individual migrant groups. This study analyses data on the 2-digit code occupations of the Australian and New Zealand Standard Classification of Occupations from the 2011 census. An index of dissimilarity and a measure of average occupational status are used in the analysis of occupational attainment patterns of the ten largest Asian birthplace groups in Australia, as well as the diversity within birthplace groups by ancestry and sex. The results reveal heterogeneous occupational patterns, with concentrations of persons in high-skilled occupations in most of the birthplace groups; concentrations of persons in low-skilled occupations in the Vietnam and Philippines birthplace groups; and bimodal occupational concentrations in the China and South Korea birthplace groups. All birthplace groups, except Vietnam, South Korea and the Philippines have higher occupational status scores than the Australia-born. Compared to other ancestry groups from the same birthplace, those of European and Australian ancestries have higher propensities to be in managerial positions, while the Chinese and Indian ancestry groups attain the highest percentages in professional occupations and overall occupational status. With the primary focus of Australia’s immigration policies being on addressing labour skills shortages, understanding the occupational outcomes of migrants is a key consideration in assessing migration policy success.

Exploring the geographical relationship between area deprivation and health: An output area level analysis using Geographically Weighted Regression
Emily Dearden, University of Liverpool   

Given persistent and widening health inequalities in the UK, this investigation aims to explore the extent and spatial distribution of disparities in health between more and less disadvantaged populations in England and Wales. Using 2011 Census data for England and Wales, this quantitative investigation has an explicit geographical focus and provides a comprehensive contextual, area-based account of how health inequalities are generated, reproduced and spatially manifested. A spatially explicit approach is employed by drawing on an under-utilised geographically weighted regression (GWR) approach to explore how health inequalities and their explanatory factors manifest spatially. GWR is a spatial statistics tool that expands standard regression by allowing for spatial variance in parameters. Preliminary GWR results confirm that place matters for health and the relationship between contextual variables and poor health prevalence varies substantially across England and Wales. Within an international policy context, this research suggests that a ‘one-size fits all’ policy strategy is not appropriate to address inequalities in health outcomes. Health policies need to be spatially adaptive, based on the contextual characteristics of each individual area. 

Census & Administrative data: LongitudinaL Studies Hub
Allan Findlay 1, Oliver Duke-Williams 2, Alan Marshall 1, Fiona Cox 1, 1 University of St Andrews, 2 University College London 

The Census & Administrative data LongitudinaL Studies Hub (CALLS Hub) has been commissioned by the ESRC to support, promote and harmonise the work of the three LS Research Support Units (CeLSIUS, NILS-RSU, SLS-DSU), with the aim of providing a more streamlined experience for users. The three UK census Longitudinal Studies provide a unique and powerful research resource for a range of academic disciplines. They also form a powerful source of research evidence for policy-makers, practitioners and third sector bodies. CALLS Hub exists to help researchers find the information and resources they need in a straightforward way, and to promote the work and impact of the RSU’s to a wider audience. By bringing together the three studies, we can also highlight the potential benefits and possibilities of using more than one LS, either to allow regional comparisons or to build a national population. The aims of CALLS Hub are: to enhance the research potential of the LSs by co-ordinating the development of new resources and methodologies, to enhance and streamline the user experience of obtaining information about the LSs and applying to use them for research, to increase academic impact by developing communication strategies to raise awareness of the LSs, promoting their outputs and facilitating their impact strategies, to increase the economic and societal impact of the LSs by working together with key external stakeholders to develop research projects meeting their evidence needs, to facilitate and encourage the use of multiple LSs for UK-wide research.

HIV and timing of antenatal care in Zimbabwe: Evidence from Multivariate multilevel Modelling of Population-based Surveys
Martin Marufu Gazimbi, School of Social Sciences, University of Hull    

Women are argued to initiate antenatal care (ANC) early and to have HIV pregnancy related test in the 1st trimester (WHO, 2004). The rationale for an early ANC visit and HIV test is to identify HIV positive mothers early and introduce them to prevention of mother to child HIV transmission programmes. Despite recommendations, women still initiate ANC very late. Recent data shows that 90% of pregnant women had at least one antenatal care (ANC) visit, but only 19 % visited in the 1st trimester (ZDHS, 2011). Based on (ZDHS, 2005-2011) pooled datasets, the study used multilevel binary logistic regression analysis to identified individual and community determinants of early ANC visit and the independent effects of HIV. The results suggest that, urban, older and wealthier women were more likely than their counterparts to initiate ANC in the 1st trimester. Similarly, women exposed to mass media were more likely to have an early ANC visit compared with women who were not. Regarding HIV factors, women not concerned with HIV status disclosure were more likely than those concerned to have an early ANC visit. Those who had no prior HIV test when they initiated ANC had higher likelihood of having an early ANC visit. At the community level, those residing in communities with many people who knew someone with HIV and those from higher HIV prevalence areas were more likely to have an early ANC visit, compared with their counterparts. The study has identified where action is needed to reduce late ANC visits. 

Scotland’s Demography: how do we compare with the rest of the UK and Europe?

Jay Gillam, Sarah Mohammed, National Records of Scotland


Scotland is a small European country, with a population roughly a tenth of the size of its closest neighbour, England. The topology of Scotland is distinguished by the Highland Boundary Fault, which separates the mountainous Highlands in the North and West from the lowland areas in the South and East where the majority of the population resides. The Highlands are the most sparsely populated part of the UK, and are among the most sparsely populated areas of Europe. The majority (90%) of Scotland’s population live in concentrated areas: settlements that make up less than 3% of Scotland’s area. Historical economic, political and cultural differences have also contributed to some of the demographic peculiarities that are still present today. How do Scotland’s demographics compare with the rest of the UK and Europe?

Students: where do they come from, where do they go? The student demography problem in Scotland
Jay Gillam 1, Rosie Seaman 2, 1 National Records of Scotland, 2 MRC/CSO Social and Public Health Sciences Unit, University of Glasgow    

Highly mobile demographic groups are among those most difficult to capture in population estimates. Young adults, especially young males, are less likely to interact with the data sources we currently use to estimate population, for example GP registrations. This research compares migration estimates from GP practice registrations with those calculated from the 2011 Census “address one year ago”. Council areas with the biggest differences are those where student populations are known to contribute to migration, and the biggest differences are seen for 18 to 24 year olds. What impact does this have on migration estimates, and what can we do about it?

The household structure in Hungary according to the census of 1869
Réka Gyimesi, Faculty of Humanities, University of Pécs, Hungary   

My research project (The demographic processes and the household forms of the 19–20th centuries on the example of Bonyhád (1850-1941) is concerned with historical demography. In my dissertation I am focusing on census materials and parish registers, which contribute to becoming familiar with the composition of historical households and the demographical characteristics of their members. Bonyhád was a multi-ethnical (Hungarians, Germans, Slovaks, Serbs) and multi-confessional (Roman Catholics, Lutherans, Calvinist, Israelites) Transdanubian town, with various distributions by occupation, with a considerable number of inhabitants. The reconstruction of the co-residential forms is based on Peter Laslett’s household typology. This method helps to differentiate households by structure.  However, there are some disadvantages of this technique; one is that it ignores the “life cycles of families”, because it is based only on census materials, a cross-section from a given point in time. To eliminate this problem, I employ Lutz Berkner’s synthetic cohort method to demonstrate the family dynamics, where social differences are also stressed. With the same methodology I analysed the individual data of census 1869 from the town of Mohács in my master thesis and gave an overview of the household structure and its determinants. I mean to compare these two nearby, southern Hungarian towns in the poster section employing the above mentioned methods. The results of this analysis will represent the household structures according to the census of 1869 in a country which is right on the famous Hajnal-line. My main question is which aspects most influenced the forming of household structures in the 19th century Hungary?

Parental work hour demands and ‘quality time’ with children
Stefanie Hoherz, University of Southampton 

This research analyses how the work hour demands, and especially long working hours, of mothers and fathers affect the time they spend with their children in activities that reflect a certain level of interaction. Especially, spending time together in structured leisure activities, eating dinner together, and talking about important matters. The statistical analysis is based on waves 1, 3 and 5 of the UK Household Longitudinal Study using Ordered Random Effects Models. I find that parents who work relatively long hours (more than 30 hours per week for mothers/more than 48 hours per week for fathers) spend less structured leisure time with their children than parents who work less or do not work. Fathers often eat less with the family when they work long hours but for mothers this is only true in households with a lower educational level. However, the frequency of talking about important matters is not affected by parents’ working hours. Interestingly, including interactions with different household income groups show that the negative effects of long working hours on time with children can mainly be found for parents with a lower household income. This indicates that parents in lower income households face particular difficulties combining work and family demands, which negatively affect the quality time they spend with their children.

Genetic programming through foetal starvation in chronically under-developed region of Sunderban, India: Foetal origin hypothesis versus predictive adaptive response
Zakir Husain 1, Mousumi Dutta 2, 1 Humanities & Social Sciences Department, Indian Institute of Technology, 2 Economics Department, Presidency University, India   

Despite India’s impressive growth performance, malnutrition has remained a major challenge. While studies have focussed on the impact of nutritional deficiency of the mother on the birth outcome, the foetal origin hypothesis argues that nutritional deficiency at the foetal stage may have a more long -term effect on physical development of the child. This will reduce productivity of the workforce, divert resources from productive use and lowers the potential growth rate of the economy. A major methodological problem in empirically verifying this hypothesis is the long time gap between foetal starvation and the onset of ill-health. The common approach is to identify a disaster year and assume that all children born in this year have been exposed to foetal starvation. But being borne in a disaster year is neither a necessary nor sufficient condition to ensure foetal starvation. Realizing this, recent studies have focussed on the Muslim community as they undertake a month-long fasting every year. The assumption is that all Muslim women fast, even when pregnant. To overcome this methodological problem, we have studied both Muslim children and their mothers, verifying whether they had really fasted during their pregnancy. A primary survey of 450 youths aged between 18-22 years and their mothers in the chronically backward region of Sunderban, India was undertaken in 2014-15. Econometric analysis of 27 indicators (height, weight, fat composition, muscle composition, etc.) revealed that foetal starvation, particularly during the first trimester, affected physical development of the study group vis-à-vis the control group.

Analysing cohorts using the ONS Longitudinal Study
Lorraine Ireland, Julian Buxton, Office for National Statistics   

The ONS Longitudinal Study (LS) contains linked census and life event data for 1% of the population of England and Wales. Information is linked from five successive censuses (1971 to 2011) and life events data, including birth, death and cancer registrations. The LS contains information on households, economic activity, qualifications and marital status from five successive censuses that will support studies on the life course. The marital status question was updated in 2011 to include civil partnerships. Questions asked for the first time in 2011 included: passports held, month and year of arrival in the UK, intended length of stay in the UK and main language. The 2016 BSPS conference presents an opportunity to highlight the LS and help researchers decide whether it is appropriate for their research. This poster will give an overview of the LS and key variables it contains and its potential use in furthering understanding of the life course. Using the Longitudinal Study, it is possible to follow a wide variety of groups or cohorts over time. These are usually defined by the characteristics of the group. For example, birth cohorts, groups of women looking at their fertility, different age groups for survival analysis, young people and older people at different time points, or groups defined by characteristics such as partnership status or economic activity status. As an example, this poster will look at working mothers and changes in their economic activity and living arrangements over time. We will also look at children born in the 1960s and the 1970s and compare their outcomes 40 years later, including housing, educational attainment and occupation.

Positive Mental Health and Mortality
Christopher Jacobi, University of Oxford   

This study examines the association between positive mental health (PMH) and all-cause mortality in a nationally-representative sample of the United Kingdom via the new Warwick-Edinburgh-Mental-Well-Being-Scale (WEMWBS). The WEMWBS combines emotional, psychological and subjective well-being into an integrated PMH measure. The coherence and usefulness of the WEMWBS as an indicator of PMH is tested by comparing and contrasting it to an established measure of negative mental health (NMH), the GHQ, and to a unidimensional measure of cognitive mental health, life satisfaction. Discrete-time survival models (complementary log-log) are used to analyse if PMH is associated with a lower risk of mortality over a 4-year follow-up period (N=28,662), when adjusting for chronic health conditions and common sociodemographic factors. The results indicate that better PMH has a strongly protective effect against mortality

The Hampshire Acute Kidney Injury Study
Matt Johnson 1, Hilda Hounkpatin 2, David Culliford 1, Simon Fraser 3, Paul Roderick 3, 1 NIHR CLAHRC Wessex Methodological Hub, Faculty of Health Sciences, University of Southampton, 2 NIHR CLAHRC Wessex Public Health and Primary Care Theme, Faculty of Medicine, University of Southampton, 3 Primary Care and Population Sciences, University of Southampton 

Acute Kidney Injury (AKI) is a rapidly occurring decline in kidney function associated with poor clinical outcomes and high burden to the health system. To facilitate detection in a clinical setting and improve patient outcomes, NHS England has mandated use of an early warning algorithm in hospital laboratories. The algorithm detects AKI based upon a rising level of creatinine in the blood, and provides an assessment of the severity of the condition at the point of testing (an ‘AKI alert’).

We used the Hampshire Health Record Analytical database, an anonymised database of linked clinical and demographic data, to reproduce the algorithm at a regional population level in a secondary data environment. Although most AKI research has focussed on patients in hospital, a significant proportion of people with AKI cross the primary/secondary care divide. Use of linked data to identify AKI acquired in hospital and in the community is therefore essential to fully describe the epidemiology of the condition.

We demonstrated the feasibility of replicating the algorithm at a population level, and of identifying risk factors and incident AKI both in hospital and the community. Of the 643,039 eligible study population, 5,724 generated at least one AKI alert during the observation year. We also found a strong social and age gradient, with older people and those resident in more deprived areas experiencing greater risk of AKI. These findings were used to inform AKI harm-prevention efforts.

The spatial and temporal dimensions of the recent refugee and migration crisis in Greece Stamatis Kalogirou, Artemis Tsiopa, Harokopio University of Athens

The aim of this paper is to study the temporal and spatial dimension of the current refugee and migration crisis in Europe focusing in the area of Greece. While the average illegal immigration to Greece was at the scale of one hundred persons per year during the 2000s and early 2010s, it increased tenfold in 2015 approaching a million people entering Greece mainly by sea. The first two months of 2016, despite the bad weather conditions, 130,000 people came to Greece by sea compared to nearly 9,000 people entering Greece in January and February 2015. By using data from the Hellenic Police and Coast Guard as well as the UNHCR, this paper attempts to map the entry points and temporary settlement of refugees and immigrants in Greece over time. Because of their proximity to the west coast of Turkey, the islands of Chios, Lesvos, Samos, Leros, Kos and Kastellorizo receive most of the immigrants. In some cases, such as in Kastellorizo, the refugees/immigrants significantly outnumber the local population who are not able to provide for the former as it would be culturally expected in Greece. Unfortunately, the Greek Government initially underestimated the scale and urgency of the problem and only recently, under international pressure, decided to act. With the western Balkan route to Western Europe currently being closed after unilateral decisions being made, most of the recently arrived refugees/immigrants have been forced to reside in Greece and gradually they are distributed across the country. 

The social and economic impact of international female labour migration on left-behind children in East Java, Indonesia
Amie Kamanda1, Saseendran Pallikadavath1,Keppi Sukesi2, Faishal Aminuddin2,Henny Rosalinda2, Ngianga II Kandala1, Kieron Hatton3, 1Portsmouth-Brawijaya Centre for Global Health, Population and Policy, University of Portsmouth, 2Portsmouth-Brawijaya Centre for Global Health, Population and Policy, University of Brawijaya,3 School of Health Sciences and Social Work, University of Portsmouth 

Context: An estimated 80% of individuals migrating officially outside of Indonesia to work are women. International female labour migration leads to the temporary separation of mothers from their children. Previous research showed that the separation led to the reduced wellbeing of the child, as being a homogeneous group. This paper explores the impact of this separation on both young children as well as adolescents; the impact is conceptualised to significantly vary across these physically and emotionally growing sub-groups. Data and methods: The study was carried out in two villages in Malang and Ponorogo Regencies in East Java. A census (178 households in two villages) was followed by in-depth interviews (71) with left-behind family members, including adolescents. Descriptive analysis was applied to the household dataset and thematic analysis was used to triangulate the views of the household members. Results and conclusions: Both young children and adolescents benefited from migration: while young children mainly benefited from better quality education, health care, consumption; older children had greater access to pocket money, and extracurricular activities. Some mothers migrated after childbirth, leaving their children in the care of their husbands or grandmothers for a longer duration. Such children reported feeling a sense of loneliness and neglect leading to emotional separation from their mothers. Adolescent girls reported difficulties in managing their physical changes without mother, leading to stress and discomfort. A contribution of this study is identifying the challenges faced by adolescent girls as they started puberty in the absence of their mothers. 

Mothers on the Move: Maternal Health among Female Migrants in Ghana
Samantha Lattof, London School of Economics and Political Science 

Background: Ghana’s migrant kayayei (girls and young women who work as head porters) are exposed to sexual and reproductive health risks including rape, unintended pregnancies, and unsafe abortion. Does the agency and independent decision making that kayayei have demonstrated by migrating in search of economic opportunities extend to their maternal healthcare decisions? Methods This mixed-methods study on female migration and health in Accra involved surveying 626 kayayei and conducting in-depth interviews with 48 purposively selected kayayei. Quantitative and qualitative analyses explore kayayei’s health status and health seeking behaviours, including the motivations for use of maternal health services. Results One-third of mothers did not want to get pregnant at the time of their last pregnancy. Yet, only 11.8% currently use family planning. In a largely illiterate and uneducated population, kayayei’s first introduction to family planning is most likely to occur at infant weighing visits after giving birth. Qualitative analyses reveal that unintended pregnancy is a concern for kayayei in Accra and also in their home villages in northern Ghana; unintended pregnancy can actually serve as a driver of migration as mothers move in search of employment to support their babies and pay children’s school fees. Conclusions Kayayei’s lack of education, combined with their vulnerability to sexual violence, impedes this population’s ability to make their own healthcare decisions and access maternal health services. The paper concludes with policy and programmatic recommendations focused on health education and access to care among female migrants in Accra and northern Ghana from where this population originates.

The timing of life events and stopping smoking: How do English men and women behave?
Maria Herica La Valle, University of Southampton and ESRC Centre for Population Change    

This work focuses on health behaviours as a cause of health outcomes in late life, and investigates the factors influencing them. In particular, it aims at verifying and identifying the life events that may lead or prevent people to stop smoking and at exploring the relationship between the timing of these life events and the decision to stop smoking at a certain point of the life course. The expectation is that a particular life event, such as cohabitation, marriage, childbearing, abortion, interruption of work, job and residential mobility, kinship ties’ rupture, affects the decision to stop smoking differently on the basis of the time when it happens. By drawing data from the third wave of English Longitudinal Study of Ageing (ELSA), which provides retrospective information on individual health, employment and family and fertility histories, the purpose is to conduct a cohort analysis by a gender perspective. The work will use Survival Analysis approach (with constant time), in order to analyse the risk to stop smoking at a certain age, considering specific life course events that are thought of potentially having an effect on it depending on the age when men and women experience them.

An unexpected situation for the Western European Marriage Pattern: repopulating Spanish rural areas after the Moors’ expulsion in the 17th century
Francisco J. Marco-Gracia, University of Zaragoza, Spain   

In 1610, the last Moors were expelled from Spain. The Moors tended to be concentrated in certain areas and populations. After the expulsion, dozens of villages were empty and had to be repopulated. I created a database using the Family Reconstitution method (Louis Henry) for 7 Aragonese villages (in the northeast of the Iberian Peninsula). With this database, we can compare the demographic behaviour of individuals who repopulated 4 villages, and individuals from 3 neighbouring villages that were composed exclusively of Christians and, therefore, its population remained stable. We will use the age at first marriage and number of children born alive as indicators. The aim is to compare whether there are different behaviours among villages (repopulation vs. stable population). The results show that the age at first marriage is low (around one year) in low population pressure villages and the number of children born alive is higher (although less than 1 additional child). Finally, we can conclude that the low population pressure favours the relaxation of the Western European Marriage Pattern.

The challenges of including sexuality education in the Life Orientation programme offered by schools: the case of Mahikeng, South Africa
Akim J. Mturi 1, Andre L. Bechuke 2, 1 School of Research and Postgraduate Studies, Faculty of Human and Social Sciences, North-West University (Mafikeng Campus), Mmabatho, South Africa, 2 Faculty of Education and Training, North-West University (Mafikeng Campus), Mmabatho, South Africa 

Teenage pregnancy among schoolgirls is persistently high in South Africa, despite the introduction of sexuality education in the schools in 2002. This paper uses a qualitative approach to investigate the challenges facing the schools offering sexuality education in Mahikeng, South Africa. Four school principals, seven Life Orientation teachers and 39 male and female learners were interviewed. The curriculum of sexuality education indicates that sexual matters are introduced to learners only when they reach Grade 8, whereas pregnancy is already found amongst girls in Grade 3. All types of participants indicated that Life Orientation is not taken seriously in schools like other subjects. Although learners are interested in the subject matter, there is a lack of qualified teachers and there is no depth to the content of sexuality education curriculum. In addition, schools pay little attention to this subject because Life Orientation does not count in terms of university entrance. This study recommends that the government needs to revise the curriculum content and presentation to see improvement of sexuality education in schools. The appropriate age-specific topics on sexual matters for learners should be devised and qualifications of teachers need to be taken seriously.

Childbearing preferences among postpartum women in Uganda
Patricia Ndugga, University of Southampton

Abstract Background: Studies on the drivers of demand for children among postpartum women in Uganda are limited and yet such studies would inform population reduction policies in a country whose fertility has remained persistently high. Despite this, there is increasing evidence that women in Uganda are having more children than they actually desire. For example, their desired fertility is 4.8 and yet actual fertility is 6.2. The aim of this paper was to examine the drivers of women’s fertility preferences in Uganda. Methods: I used the 2011 Uganda Demographic and Health Survey (UDHS) data, and selected a weighted sample of 2, 789 married and unmarried women who had a live birth in the last two years prior to the survey and were not pregnant. The main dependent variable was child preference (want more/ want no more). I used chi-squared tests and binary logistic regressions to investigate the factors associated with childbearing preference among postpartum women. Results: Almost two thirds (63%) of women want to have another child soon or later. The odds of reporting preference for another child were higher among women from the North/ Karamoja region (OR = 3.65; 95% CI: 1.88-7.08) and among married women (OR = 3.09; 95% CI: 2.25-4.24). Conclusion: Interventions addressing fertility should place more emphasis on married women and those women in North/ Karamoja region of Uganda and assisting them realise their fertility preferences

The power of a spouse and fertility of a couple – a comparison of France and Poland
Beata Osiewalska, Kracow University of Economics   

The theory and previous literature suggest that various fertility levels observed nowadays among European countries are correlated with the country’s institutional support for a family, as well as with the gender equality and power relation between partners. This study aims to investigate couples’ procreative behaviour by the multidimensional partners’ power distribution, which includes educational and age gap between partners, female age at union formation, the division of housework and decision-making responsibilities. The main research question is whether and, if yes, how the balance or imbalance in the division of power between partners influence their reproductive behaviour regarding both, tempo and quantum effects of fertility, and whether this impact depends on the country specific-context. Consequently, France and Poland are compared. Poland gives a context of a low level of gender equity, in both family institution and individual layer. France, in turn, provides a case of a higher level of gender equity and well-adjusted family institutions. The first wave of GGS data is used. Since behavioural drivers might differ among parents and childless couples, both groups are considered under the hurdle Poisson model. The analysis confirmed that couple’s family model is driven by the gender power distribution. The results differ by country. In post-socialistic Poland, the male advantage in power relations has the strongest positive impact on fertility. In France, in turn, the positive effect of both male and female empowerment on fertility occurs. Interestingly, in both countries, equal power division between partners is associated with a relatively low level of fertility.

Childlessness in Colombia: Exploring the trends, associated factors, and meanings of non-parenthood since the 1980s
Cristina Perez, University College London   

Between 1965 and 2010, Colombia experienced a ‘spectacular’ fertility decline from a TFR of 7 children per woman to 2.1. The country’s women now outperform men at every level of education, and between 1985 and 2005, female labour-force participation rose faster than in any other country in the region. Against a backdrop of profound sociodemographic change, this two-part mixed methods study aims to improve our understanding of childlessness, beyond ‘infertility’, in a less-developed country context. The first, quantitative part (presented here) analyses a series of Demographic and Health Surveys, carried out in 1986 and every five years since 1990. Using descriptive statistics and generalised linear modelling, it explores the factors associated with female childlessness (as compared to motherhood) at age 30 and 45. It asks who is most likely to be childless; whether this is a voluntary or involuntary outcome; and whether this changes across the life-course. It focuses particularly on the stratification of reproduction (including ‘childlessness’) according to socioeconomic stratum, education, region, urban/rural residence, occupation, and partnership status. Finally, it explores the hypothesis that a ‘dual’ demographic transition is taking place, where different sub-groups are simultaneously experiencing the changes associated with the First and Second Demographic Transitions. The second part of the study will complement this quantitative analysis with 12 months of ethnographic fieldwork in Colombia, in order to better understand women’s and men’s contextualised experiences of childlessness and parenthood. One of the study’s ultimate goals is to contribute to extending anthropological demography to Latin America.

Do young people not in education, employment or training (NEET) experience long term occupational scarring? A longitudinal analysis over 20 years of follow up Kevin Ralston, Zhiqiang Feng, Dawn Everington, Chris Dibben, University of Edinburgh   

NEET is a contested concept. However, it is consistently used by policy makers and shown in research to be associated with negative outcomes. In this paper we use the Scottish Longitudinal Study (SLS) to examine whether NEET status is associated with subsequent occupational scarring. The SLS provides a 5.3% sample of Scotland, based on the censuses of 1991, 2001 and 2011. We model occupational position, using CAMSIS, controlling for the influence of sex, limiting long term illness, educational attainment and geographical deprivation. We find the NEET categorization to be a marker of a subsequent negative outcome at the aggregate level. This appears to be redolent of a Matthew effect, whereby disadvantage accumulates to the already disadvantaged. Our results also show that negative NEET effects are variable when stratifying by educational attainment and are different for men and women. These findings confirm that there are negative effects on occupational position associated with prior NEET status but that outcomes are heterogeneous depending on levels of education and gender.

Outsourcing Housework and the Female Labor Supply – A Natural Experiment 
Liat Raz-Yurovich; Department of Sociology and Anthropology and the School of Public Policy, The Hebrew University of Jerusalem   

Why is it important that welfare states provide incentives for households to outsource housework? The outsourcing of housework has the potential to help women better reconcile work and family life; this process can also affect the labor supply of women who purchase such services. We use the enactment of the Belgian Service Voucher Scheme and its extended intervention period as a natural experiment to examine whether the growing ability of highly skilled women to employ domestic workers, due to this policy, has increased their employment rates. Using time-series analyses as well as fixed-effect models, we find both short- and long-term positive changes in the employment rates of these women in Belgium, which corresponds to the growth in the adoption of the policy. Moreover, the results of our fixed-effects models suggest that the primary driving mechanism behind this observed growth is the ability of highly skilled women to outsource housework.

Characteristics of those approaching retirement and those at retirement age using the ONS Longitudinal Study
Nicola Rogers, Vicky Field, Lorraine Ireland, Office for National Statistics 

The ONS Longitudinal Study (LS) contains linked census and life event data for 1% of the population of England and Wales. Information is linked from five successive censuses (1971 to 2011) and life events data, including birth, death and cancer registrations. The LS contains information on households, general health, caring and economic activity and many other variables that will support studies on the life course. The 2016 BSPS conference presents a timely opportunity to show researchers an example project using the LS and how it can be used to inform wider policy. The aim of this project is to gain a better understanding of the circumstances of those approaching retirement age and during retirement. Comparisons will be made across two cohorts and then used to make inferences about the experiences of future generations. The results will be used to help inform the State Pension Review that is being undertaken by John Cridland. We will present results in trends in those aged 45-64 approaching retirement and how this has changed over time (1991-2011). Social demographic information will include whether individuals are in work and whether they are reducing their working hours as they approach retirement. It will include living arrangements, marital status and whether they have children living at home. Comparisons will also be made for those aged 65+ in 2011 and those aged 65+ in 2001. Looking at whether they have caring responsibilities, their health and comparing living arrangements

Adolescent drinking in Chile: does it matter which school they go to?
M. Francisca Roman, Noriko Cable, Yvonne Kelly, Owen Nicholas, Department of Epidemiology and Public Health University College London 

Taking Chile as an example of a socially stratified schooling system, this study aims to examine how school socioeconomic environment would relate to pupil’s drinking patterns over and above individual and parental factors. Multilevel analyses were conducted on cross-sectional data with four-level nested structure: pupils within classes within schools within municipalities. Individual-level information was extracted from a nationally representative survey (N=58,148, aged 13 to 18) conducted in 2013 (School Population National Substance Use Survey) and linked to school-level data (N=1,687) obtained from the Ministry of Education. Patterns of non-drinking (having never drunk), non-binge drinking (having drunk but not bingeing) and binge drinking (consumption of five or more drinks in the past month) were identified at the individual level. School socioeconomic status comprised five socioeconomic groups: low, lower middle, middle, upper middle and high. Multilevel logistic regression models of non-binge drinking (vs. non-drinking) and binge drinking (vs. non-binge drinking) were stratified by gender and adjusted for individual-level variables. About half the participants reported drinking alcohol but not bingeing. Thirty percent of participants reported binge drinking in the past month. Results obtained from multilevel analysis showed that pupils attending socially disadvantaged schools were less likely to drink (Boys: OR=0.46, 95%CI=0.39-0.55; Girls: OR=0.73, 95%CI=0.60-0.88), but if they happen to drink, they were likely to binge drink (Boys: OR=1.24, 95%CI=1.07-1.42; Girls: OR=1.75, 95%CI=1.51-2.03) compared to those from advantaged schools. Results suggest that risks of alcohol-related harm caused by binge drinking can be more prevalent in pupils from socially disadvantaged schools.

The socio-economic consequences of international female labour migration on left-behind family members in East Java, Indonesia
Henny Rosalinda 1, Saseendran Pallikadavath 2, Amie Kamanda 2, Keppi Sukesi 1, Faishal Aminuddin 1, Kieron Hatton 3, 1 Portsmouth-Brawijaya Centre for Global Health, Population and Policy, University of Brawijaya, Malang, Indonesia 2 Portsmouth-Brawijaya Centre for Global Health, Population and Policy, University of Portsmouth, 3 School of Health Sciences and Social Work, University of Portsmouth 

Context: International female labour migration leads to the temporary separation of women from their husbands, children and family members. One of the countries in the South East Asia region which has witnessed the feminisation of migration is Indonesia. Eighty percent of Indonesia’s official international labour migrants are female, originating mainly from Java. A comprehensive analysis on the impact of female migration on the left-behind family members at the household level is needed to provide an evidence base for policies. Data and methods: Data were collected using a sequential mixed method strategy; household census followed by interviews of left-behind family members from two villages. Household census (178 households) data and 71 interview transcripts were analysed using descriptive analysis and thematic analysis, respectively. Results and conclusions: There is clear evidence of migration leading to economic benefits to the family. Realising the pride of constructing a new home was reported as a significant achievement, as it symbolises the economic and social status. A prosperous home with finances to meet basic necessities, school fees, and medical emergencies were enabled by remittances. On the social side, the impact was both positive and negative. Reversal of gender roles was noted; men doing chores which were traditionally performed by women. While this is a welcome change, this reversal discontinues when the women return. On the negative, migration posed social challenges from marital dissolution, child neglect and family instability to the burden of childcare for grandparents. Support mechanisms for left-behind family members should be developed as well as advice on effective management of remittances.

“Stop the killer in the kitchen”: inter-linkages between women’s intrahousehold bargaining power and clean fuel adoption in Sub-Saharan Africa: evidence from Senegal
Soazic Elise Wang Sonne, United Nations University, Merit 

In Sub-Saharan Africa, 76% of the population, mostly in rural areas uses firewood as fuels for cooking and heating, generating a great amount of pollution through the release of black carbon and CO2 (IEA 2015). Fuel switching models to understand the high dependency of Sub Saharan African households on traditional fuels have been blind beneath and above the household level, undermining the role played by intra-household (women’s) bargaining and the household’s embeddedness in its external environment (informal institutions). As a result, there is still a lack of understanding of the region’s high dependency on solid fuels. Using two barely used statistical approach, namely a Latent Trait Model and a Simultaneous Equation Model on Demographic and Heath Survey (DHS) data from Senegal, this paper suggests that intra-household bargaining and informal institutions in which households are embedded in, also affect the adoption and the transition from traditional to clean fuels. We found in line with the literature that while some socio-economic characteristics matter in the adoption of clean fuel (age, household size, land and house ownership, wealth, earnings, religion, region, type of residence, education); an increase of woman’s intrahousehold bargaining power leads to an increase of clean fuel adoption. Reversely, households using a clean fuel are the ones with woman having a high level of bargaining power. Therefore, policies aiming to enhance clean fuel uptake, ‘stopping the killer in the kitchen’ should also focus on enhancing women’s empowerment through for instance, Cash Transfer Program.

The Washington State Census Board and Its Demographic Legacy
David A. Swanson, University of California Riverside

This presentation covers two main topics: (1) an historical overview of the Washington State Census Board from its establishment in 1943 to its abolition in 1967, when its operations were transferred from the University of Washington into an executive agency of state government; and (2) a description of the Board's legacy within the state of Washington and beyond it.

Forecasting Using Modified Cohort Change Ratios
Jeff Tayman, University of California San Diego David A. Swanson, University of California Riverside,  Lucky M. Tedrow, Western Washington University Jack Baker, University of New Mexico

The Hamilton-Perry (HP) method, which uses cohort change and related ratios (CCRs), has gained acceptance as research has demonstrated its practical value and accuracy in forecasting population. Assessments of this method have been based on the usual assumption that CCRs developed over the base period are held constant over the forecast horizon. We are not aware of any studies that have evaluated the accuracy of the HP method using modified CCRs. In this chapter, we propose several approaches for modifying CCRs over the forecast horizon. These approaches are averaging and trending CCRs and a synthetic method that bases local CCR changes on state-level changes in CCRs. We evaluate their errors against the forecast errors holding the CCRs constant for counties in Washington State and for census tracts in New Mexico. The evaluation suggests that averaging or trending CCRs are not worthwhile strategies, but the synthetic method reduces errors compared to holding CCRs constant over the forecast horizon.

Age at first birth and life outcomes: the case of teenage mothers in the ONS Longitudinal Study
John Tomkinson, French Institute for Demographic Studies (INED) / University of Strasbourg (SAGE)   

Based upon the hypothesis that a teenage birth represents a “major disruption” in the lifecourse (Charbonneau, 2000), much interest has been devoted to the question of what becomes of teenage mothers in later life? If in later life women having had a child as a teenager suffer poorer life outcomes, the direct effect of the timing of entry into motherhood is hard to establish and poses a methodological challenge – the question being whether the life trajectory observed post teenage birth is a continuation of a pre-existing pathway or represents a veritable turning point. Several studies have used innovative methods to try to evaluate the effect of a teenage birth on ulterior life outcomes (Geronimus et Korenman, 1992; Ermisch et Pevalin, 2003; Hotz et al., 2005). This study attempts to respond to this question by following the familial, professional and housing trajectories of women using the ONS Longitudinal Study (LS). The results of this analysis indicate associations between a teenage first birth and poorer life outcomes compared to women having entered into motherhood between the ages of 25 and 29 years old. Amongst others, those women having experienced a teenage birth are more likely to have never worked (OR = 8.4), to be a single parent (OR = 2.3) and to live in a deprived household (OR = 1.9) whilst being less likely to be married (OR = 0.4) or to be a homeowner (OR = 0.2). Some of these associations (inactivity, difficulties to enter the housing market) are stronger for those teenage mothers being from backgrounds of lower social origin.

Planning for the future – Census data, Population data, Geographical data
Jo Wathan, University of Manchester   

The UK Census is changing from being a monolithic and self-contained information gathering exercise to become part of a wider in combination with a range of other information sources to meet the UK’s future information requirements as effectively and economically as possible. It is important that at this time of change, we have a clear understanding of research users’ data needs. The Census Support Service (part of the UK Data Service) and its precursors have provided access to complementary data from non-census sources for many years, and we are keen to ensure that this work is expanded to maintain and improve data utility in a period of census transformation. This poster will cover the following: services provided by Census Support, work underway to improve the accessibility of data and the flexibility of our tools, to enable users to query multiple data types and to generate linked outputs, datasets already supported by Census Support that are from sources other than the census, current priorities for integrating additional data sources into our systems to improve ease of use, information on our work to establish user needs going forward, and how you can make your views known.

Three generations of young people Not in Education, Employment or Training (NEETs) in England and Wales - household composition and care responsibilities Wei Xun, Christopher Marshall, Rebecca Lacey, Stephen Jivraj, Nicola Shelton, Department of Epidemiology and Public Health, University College London 

Worklessness at early stages in the life-course can have long-term effects on health status in later life through the accumulation of socio-economic disadvantages. There is extensive evidence on which individual characteristics are associated with youth worklessness, but fewer studies focused on the role of the wider picture of the household and caring responsibilities may affect economic activity statuses in young people. This study of people aged 15/16-24 explores the associations between the status of not in Education, Employment or Training (NEET) and baseline household composition, in three representative, cross-sectional samples of 1% of the England and Wales census population drawn at 1971, 1991 and 2011 from the Longitudinal Study (ONS LS). Three aspects of household composition and NEET status are investigated: child, spousal and parental characteristics. Preliminary findings from age-adjusted logistic regression analysis showed that for women, having childcare responsibilities verses those who did not, had increased odds of NEET status for both married- and lone-mothers, verses married couples without children as reference. Unmarried men were more likely to be NEETs compared to their married counterparts, although this is inversed in women. For those with parents present in the same household, having parents who work reduced the odds of being NEETs in young people of both genders. Further results will also be presented on the influences of spousal and parental caring responsibilities. 

Health-selective migration and its role in the spatial distribution of psychiatric morbidity, with reference to England and Wales
Sam Wilding, David Martin, Graham Moon, Department of Geography and Environment, University of Southampton   

The spatial distribution of psychiatric morbidity and its relationship with environmental risk factors (e.g. overcrowding) is well established. Yet, the movement of individuals with varying degrees of wellbeing within the country are often overlooked as drivers of mental health geographies. Past research has shown that young healthy migrants moving into urban university towns for instance exaggerate regional health inequalities in physical health, and 50% of these inequalities can be explained by migration away from region of birth. Less is known about the impact on mental health geographies of migration. This is important because migration flows may exaggerate or mask the association between environmental risk factors and mental health. Often analyses fail to account for influences on migration behaviour at the origin and destination, which we aim to address. Drawing on the British Household Panel Survey and Understanding Society from 1991 until 2013 (23,546 individuals), we assess changes in the mental health (GHQ-12) and migration patterns of individuals. Longitudinal logistic models are used to predict the probability of moving between waves of the surveys, comparing the probabilities for those at risk of psychiatric morbidity and those not at risk. The analysis extends methodology outlined in Thomas et al (2015) to control for influences at the origin and the destination for migrants and characteristics known to influence migration behaviour. We test the hypothesis that psychiatric morbidity is associated with more frequent residential moves and moves into deprived inner city neighbourhoods. References Thomas, M., Stillwell, J., & Gould, M. (2015). Modelling multilevel variations in distance moved between origins and destinations in England and Wales. Environment and Planning A, 47(4), 996–1014.

Couples’ parental leave strategies and continued childbearing: The importance of economic bargaining and gender roles
Jonas Wood, Karel Neels, University of Antwerp, Belgium 

In tandem with increases in women’s labour force participation and the decline of fertility in Western European countries, the late 20th Century witnessed the development of work-family policies such as parental leave schemes. Although available research assesses the determinants of parents’ leave-taking and the association with continued childbearing, particularly for Scandinavia, this article aims to enhance our knowledge in two respects. First, we use longitudinal couple data allowing to distinguish economic rationality (e.g. the partner with the lowest pre-birth wage will be most likely to exit the labour force or take parental leave) from socially embedded practices and gender norms. Second, although previous studies acknowledge that self-selection into different leave strategies may influence reported associations with continued childbearing, this selectivity is rarely documented. Using a register-based panel for 77,901 couple-quarters, this study takes a two-step approach by modelling couples’ parental leave strategies and the impact on second birth hazards with particular attention to a wide range of pre-birth employment characteristics and couple-level unobserved heterogeneity. Preliminary results support economic rationality as couples with a female main earner before childbearing are more likely to display a strategy in which he opts out of the labour force. However, with respect to parental leave use, this mechanism does not seem to hold, suggesting that parental leave strategies reflect unobserved gendered norms. Currently, this project implies the estimation of second birth hazard models with particular attention to parental leave strategies, pre-birth employment characteristics, and couple-level deviances in parental leave strategies due to unobserved characteristics.