Abstract
This paper draws on the author's PhD research on developing strategies for estimating demographic rates for small areas and ethnic groups for use in population projections. Specifically this paper presents and evaluates six strategies for estimating small area fertility rates for ethnic groups in Bradford.
In the UK, at subnational and subdistrict levels there is a growing need for plausible and reliable population projections for efficient planning and resource allocation by Local Authorities. Ideally projections are required with sex and single year of age (SYOA) detail and often according to ethnic group.
The first part of the paper introduces the Bradford fertility data (which can be likened to Hospital Episode Statistics) and the problem presented by the small number of births when broken down by ward and ethnic group.
The second part gives an overview of the strategies through which the Bradford wards may be grouped together to overcome the problem of small numbers when deriving Age Specific Fertility Rates (ASFRs). This starts by simply grouping wards according to the Total Fertility Rate (TFR), then grouping wards by urban-rural measures, then ONS classifications of wards and various commonly-used deprivation indices. These are considered along with cluster analysis undertaken on the wards using a range of Census variables which encompass indicators of different dimensions of deprivation.
The final part considers six sets of ASFRs based on different estimation strategies, assessing their performance against actual births both by ethnic group at ward-level and at the ward-level. The measurement of error presented is the mean average percentage error.
In the UK, at subnational and subdistrict levels there is a growing need for plausible and reliable population projections for efficient planning and resource allocation by Local Authorities. Ideally projections are required with sex and single year of age (SYOA) detail and often according to ethnic group.
The first part of the paper introduces the Bradford fertility data (which can be likened to Hospital Episode Statistics) and the problem presented by the small number of births when broken down by ward and ethnic group.
The second part gives an overview of the strategies through which the Bradford wards may be grouped together to overcome the problem of small numbers when deriving Age Specific Fertility Rates (ASFRs). This starts by simply grouping wards according to the Total Fertility Rate (TFR), then grouping wards by urban-rural measures, then ONS classifications of wards and various commonly-used deprivation indices. These are considered along with cluster analysis undertaken on the wards using a range of Census variables which encompass indicators of different dimensions of deprivation.
The final part considers six sets of ASFRs based on different estimation strategies, assessing their performance against actual births both by ethnic group at ward-level and at the ward-level. The measurement of error presented is the mean average percentage error.
Original language | English |
---|---|
Publication status | Unpublished - 2006 |
Event | British Society for Population Studies - Southampton, United Kingdom Duration: 18 Sept 2006 → 20 Sept 2006 |
Conference
Conference | British Society for Population Studies |
---|---|
Country/Territory | United Kingdom |
City | Southampton |
Period | 18/09/06 → 20/09/06 |