Applying Census Data for Small Area Estimation in Community and Social Service Planning

ObjectivesThis study aims to address, for the first time, the challenges of constructing small area estimates of health status using linked national surveys. The study also seeks to assess the concordance of these small area estimates with data from national censuses.SettingPopulation level health status in England, Scotland and Wales.ParticipantsA linked integrated dataset of 23 374 survey respondents (16+ years) from the 2011 waves of the Health Survey for England (n=8603), the Scottish Health Survey (n=7537) and the Welsh Health Survey (n=7234).Primary and secondary outcome measuresPopulation prevalence of poorer self-rated health and limiting long-term illness. A multilevel small area estimation modelling approach was used to estimate prevalence of these outcomes for middle super output areas in England and Wales and intermediate zones in Scotland. The estimates were then compared with matched measures from the contemporaneous 2011 UK Census.ResultsThere was a strong positive association between the small area estimates and matched census measures for all three countries for both poorer self-rated health (r=0.828, 95% CI 0.821 to 0.834) and limiting long-term illness (r=0.831, 95% CI 0.824 to 0.837), although systematic differences were evident, and small area estimation tended to indicate higher prevalences than census data.ConclusionsDespite strong concordance, variations in the small area prevalences of poorer self-rated health and limiting long-term illness evident in census data cannot be replicated perfectly using small area estimation with linked national surveys. This reflects a lack of harmonisation between surveys over question wording and design. The nature of small area estimates as ‘expected values’ also needs to be better understood.

Download Full-text

A Map of the Poor or a Poor Map?

Mathematics ◽

10.3390/math9212780 ◽

2021 ◽

Vol 9 (21) ◽

pp. 2780

Author(s):

Paul Corral ◽

Kristen Himelein ◽

Kevin McGee ◽

Isabel Molina

Keyword(s):

Small Area ◽

Census Data ◽

Mean Squared Error ◽

Small Area Estimation ◽

Selection Procedure ◽

Data Transformation ◽

Estimation Methods ◽

Simulation Experiments ◽

Area Estimation ◽

Level Information

This paper evaluates the performance of different small area estimation methods using model and design-based simulation experiments. Design-based simulation experiments are carried out using the Mexican Intra Censal survey as a census of roughly 3.9 million households from which 500 samples are drawn using a two-stage selection procedure similar to that of Living Standards Measurement Study (LSMS) surveys. The estimation methods considered are that of Elbers, Lanjouw and Lanjouw (2003), the empirical best predictor of Molina and Rao (2010), the twofold nested error extension presented by Marhuenda et al. (2017), and finally an adaptation, presented by Nguyen (2012), that combines unit and area level information, and which has been proposed as an alternative when the available census data is outdated. The findings show the importance of selecting a proper model and data transformation so that model assumptions hold. A proper data transformation can lead to a considerable improvement in mean squared error (MSE). Results from design-based validation show that all small area estimation methods represent an improvement, in terms of MSE, over direct estimates. However, methods that model unit level welfare using only area level information suffer from considerable bias. Because the magnitude and direction of the bias is unknown ex ante, methods relying only on aggregated covariates should be used with caution, but may be an alternative to traditional area level models when these are not applicable.

Download Full-text

Identification of Disaggregated Hotspots of Child Morbidity in Bangladesh: An Application of Small Area Estimation Method

10.1101/699538 ◽

2019 ◽

Author(s):

Sumonkanti Das ◽

Bappi Kumar ◽

Luthful Alahi Kawsar

Keyword(s):

Small Area ◽

Census Data ◽

Small Area Estimation ◽

Random Effect ◽

National Level ◽

Estimation Method ◽

Response Variable ◽

Area Estimation ◽

Child Morbidity ◽

Administrative Units

AbstractAcute respiratory infection (ARI) and diarrhoea are two major causes of child morbidity and mortality in Bangladesh. National and regional level prevalence of ARI and diarrhoea are calculated from nationwide surveys; however, prevalence at micro-level administrative units (say, district and sub-district) is not possible due to lack of sufficient data. In such case, small area estimation (SAE) methods can be applied by combining a survey data with a census data. Using a SAE method for dichotomous response variable, this study aims to estimate the proportions of under-5 children experienced with ARI and diarrhoea separately as well as either ARI or diarrhoea within a period of two-week preceding the survey. The ARI and diarrhoea information extracted from Bangladesh Demographic and Health Survey 2011 are used to develop a random effect logistic model for each of the indicators, and then the prevalence is estimated adapting the World Bank SAE approach for the dichotomous response variable using the 5% data of the Census 2011. The estimated prevalence of each indicator significantly varied by district and sub-district (1.4-11.3% for diarrhoea, 2.2-11.8% for ARI and 4.3-16.5% for ARI/diarrhoea at sub-district level). In a number of districts and sub-district, the proportions are found double the national level. District and sub-district levels spatial distributions of the indicators might help the policy makers to identify the vulnerable disaggregated and remote hotspots. Particularly, aid industries can provide effective interventions at the highly vulnerable spots to overcome the gaps between micro and macro level administrative units.

Download Full-text

SPATIAL ANALYSIS OF POVERTY AND HAPPINESS PROFILES IN SPECIAL REGION OF YOGYAKARTA USING SMALL AREA ESTIMATION METHOD

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xlii-4-w16-189-2019 ◽

2019 ◽

Vol XLII-4/W16 ◽

pp. 189-196

Author(s):

S. R. El-Yasha ◽

M. Rizky ◽

T. W. Wibowo ◽

Keyword(s):

Small Area ◽

Gini Index ◽

Census Data ◽

Small Area Estimation ◽

Estimation Method ◽

Poverty Line ◽

Poor People ◽

Area Estimation ◽

High Level ◽

Special Region

Abstract. In March 2017, the Province of Special Region of Yogyakarta (DIY Province) has poverty line of IDR 374,009, percentage of poor people (13.03%) and Gini index (0.432) above the national average (IDR 374,478; 10.64%; 0.393). The result of happiness index in 2017 shows the position of DIY Province (72.93%) is above average of national happiness index (70.69%). Scatterplot between happiness index and percentage of poor people in Indonesia in 2017 shows that DIY Province is on first quadrant. This marks the high level of happiness along with high percentage of poor people. Small area estimation method developed by Elbers et al (known as ELL method) is used to determine spatial characteristics of poverty and happiness profiles in DIY Province. This study used village census data (Podes) 2018; Susenas March 2017 and SPTK 2017 as survey data. There are twenty three household variables and another five variables that are significant to poverty and happiness models at urban and rural provincial level. Rural regency areas dominates high poverty profile (FGT0 0.0491 – 0.1076), low happiness profile (FTG0 0.0087 – 0.0124), and inequality of happiness profile (Gini index 0.0847 – 0.0923). Urban regency areas dominates low poverty profile (FTG0 0.0082 – 0.0491), high happiness profile (FTG0 0 – 0.0087), and perfect equality of both income (Gini index 0.3048 – 0.3604) and happiness profiles (Gini index 0.0624 – 0.0847). Yogyakarta City has happiest and wealthies profiles, whereas Gunung Kidul regency urban area has perfect equality of both income and happiness profiles.

Download Full-text