Ethnic and cultural demography

Time, generation or migration? Exploring the convergences of differences in household characteristics by ethnic group. 
Gemma Catney, Ludi Simpson (University of Manchester) 

Differences in the socio-demographic characteristics of ethnic groups are of policy relevance; the persistence or otherwise of these differences has implications for planning, and for addressing issues of persistence in inequality and discrimination. Of interest, for example, is family size and structure, fertility levels, employment and occupational status, and how these may be converging, diverging or remaining stable over time, dependent on the influences of immigration itself, of cultural or traditional preferences, and of structural disadvantage. In this paper, we explore the household characteristics of ethnic groups in England and Wales, to consider how household size (number of adults and dependent children) differs between each minority group and the White population over time, and between types of area, including areas that are immigrant settlement areas and other areas to which minorities have subsequently migrated. A well-documented case of variation in this regard relates to traditionally larger household sizes – attributable in particular to the number of dependent children and the prevalence of extended households – in the South Asian population compared to other ethnic groups. We are concerned with how such differences may be reducing over time, generation and migration. We provide a solution to the methodological issues of changing geo-statistical units in the census over time, and the small populations of minorities in areas of dispersal which are nonetheless of substantive interest. 


The intergenerational transmission of Welsh: an analysis of the 2001 Census results. 
Hywel Jones (Bwrdd yr Iaith Gymraeg / Welsh Language Board) 

The intergenerational transmission of language within the family is recognised as a crucial feature which must be assured if language shift is to be reversed (Fishman, 1991). Wales has some tradition of attempting quantitative analysis of this topic. A research project into the attitudes of bilingual mothers who had chosen to rear monolingual English-speaking children (Harrison et al., 1981) identified relationships between transmission and family composition and social class. The published 1991 Census results and a major official survey in 1992 provided further statistical sources. The Welsh Language Board released an analysis of the published 2001 Census results in 2003. Recent sociological research (Gathercole (ed), 2007; Irvine et al., 2008; Morris and Jones, 2007) has looked at the effect of the factors identified by Harrison, and others, on transmission within the family, with varying conclusions. Although the 2001 Census results could provide data related to some of these factors, the published results did not do so. Special tabulations were commissioned to provide the relevant data. Variables available for analysis included family composition (including whether married or cohabiting, sex of adults present and their ability to speak Welsh) and socio-economic classification. The analysis included the use of log odds ratios for exploratory analysis, and logistic and spatial modelling, including grappling with the implications of the Census's small cell count adjustment routine. The paper will report on the analyses and draw some comparisons with Scotland and Ireland. 


Ethnicity, country of birth and the Scottish escalator effect. 
Maarten van Ham1, Allan Findlay2, David Manley1 (1University of St. Andrews, 2University of Dundee) 

A number of powerful forces have produced uneven opportunities for occupational advancement in Scotland. Edinburgh as capital of a devolved nation, hub for financial service activities and regional head office location for many public sector bodies certainly boasts many of the characteristics that one would expect to find in an escalator region. However, there are also many individual level factors that can influence success in the labour market. This paper seeks to unpick the complex relationships between occupational mobility, migration, ethnicity, country of birth and place of residence in Scotland. We do this using longitudinal Census data linking individual records from the 1991 and 2001 Censuses obtained via the Scottish Longitudinal Study. From the data we create logistic regression models assessing the probability that individuals in social classes 3 or 5 move upwards, and individuals in social classes 1 or 2 manage to maintain their position. Particular attention is given to the labour force experience of English-born residents of Edinburgh, whom the cross sectional literature suggests are more likely to achieve high occupational status than their Scottish counterparts. 


The estimation and application of ethnic group specific fertility rates in a projection of sub-national populations in the UK.
Paul Norman (University of Leeds) 

As inputs to a projection model, fertility rates by ethnic group have been estimated for local authorities in the UK using a combination of data sources: Vital Statistics, Census, Labour Force Survey and Mid-Year Estimates. To reduce the complexity of developing future fertility scenarios, local authorities have been 'clustered' into a classification of locations sharing similar fertility levels and trends over the last 25 years. This paper will describe these developments and will show how the adoption of different fertility assumptions impacts on projected births by ethnic group.  

The research reported is supported by ESRC Research Award RES-163-25-0032, "What happens when international migrants settle? Ethnic group population trends and projections for local areas." 


Choice-based letting, ethnicity and segregation in England. 
David Manley, Maarten van Ham (University of St. Andrews) 

Concerns have been expressed that offering people choice in residential locations through choice-based letting has the potential to increase neighbourhood segregation along ethnic lines. It is well established that there are differences between ethnic groups in the desired ethnic mix of the neighbourhoods in which they live. Over time, these (sometimes small) differences can lead to marked patterns of ethnic segregation. Alternatively, it has been argued that the lack of real choice under choice-based letting, not self-segregation by choice, might be a cause of (sustained) social and ethnic segregation in social housing. Currently, very little is known about consumer responses to choice-based lettings and choice-based letting outcomes in terms of their spatial or distributional effects of social housing applicants over neighbourhoods. In this paper we analyse whether households who let their property under choice-based letting end up in neighbourhoods with different levels of socio-economic deprivation and ethnic concentrations than households who are matched to a dwelling using non-choice-based letting systems. Special attention is paid to how the effect of choice-based letting differs for ethnic minority households and non-ethnic minority households. We use unique data representing the majority of all lettings made in the social housing sector in England in 2006/2007 collected for the Housing Corporation and model the effect of ethnicity and choice-based letting on the type of destination neighbourhood using a multinomial multilevel logistic regression model. 


Ethnic density and social cohesion in the United Kingdom. 
Laia Becares1, James Nazroo2, Mai Stafford1 (1University College London, 2University of Manchester) 

Recent studies suggest that ethnic density undermines a sense of community and social cohesion, arguing that as ethnic concentration increases in an area, social capital decreases (Alesina & Ferrara, 2002; Costa & Khan, 2003; Putnam, 2007). However, the bulk of this literature has been conducted in the US, and the association between ethnic diversity, social capital and social cohesion is not yet clear in the UK, and particularly among ethnic minority groups. This study hypothesises that residence in areas of greater ethnic minority concentration is associated with increased social cohesion and mixed-ethnicity friendships, but that this association depends on whether ethnic density is measured as co-ethnic, overall ethnic minority, or perceived ethnic density. To test this, multilevel logistic regressions adjusted for age, sex, individual socioeconomic position and area deprivation were applied to a combined dataset from the 2005 and 2007 Citizenship Surveys to obtain odds ratios of reporting social cohesion, proportion of friends of same ethnicity, and neighbourhood sense of belonging as ethnic density (co-ethnic, overall ethnic minority, and perceived) increased. This study found that social cohesion is greater in areas of higher ethnic density for all ethnic minorities on all three ethnic density measures, with the exception of Caribbean people. Feelings of belonging to the neighbourhood and having friends of the same ethnic group were significantly associated with all measures of ethnic density, although the effect was stronger for perceived ethnic density. White people were more likely to have mixed ethnicity friendships as overall ethnic minority increased. 


Racial and ethnic residential segregation and socioeconomic status in the United Kingdom and United States.  Pablo Mateos1, John Iceland2 (1University College London, 2Pennsylvania State University) 

Immigration over several decades has increased racial and ethnic diversity in the United States, the European Union, and indeed in many countries around the globe. Commentators have struggled to understand the social, political, and economic implications of this diversity. In this paper we develop a cross-national comparison of the residential or "spatial" incorporation of ethnic minority groups between the United Kingdom and the United States. More specifically, we compare the role that socioeconomic status (SES) plays in shaping residential patterns for different ethnic groups in both countries, an issue that has been under-explored in existing cross-comparative studies. We examine the extent to which higher SES minority ethnic groups are less segregated from others than lower-SES ones. Our analyses rely on data from the 2001 UK census and the 2000 US census. We compute levels of ethnic segregation, using the dissimilarity and isolation indexes, for a set of comparably defined ethnic groups and small areas, as well as a few additional groups that are demographically important in each of the two national contexts. In doing so we aim to arrive at a better understanding of ethnic incorporation in the UK and the US, and the ways in which economic and residential patterns help shape one another.