PPubMed27 Sep 2023Cited 18×

Identifying deprived “slum” neighbourhoods in the Greater Accra Metropolitan Area of Ghana using census and remote sensing data

Robert MacTavish, Honor Bixby, Alicia Cavanaugh, Samuel Agyei-Mensah, Ayaga Bawah, George Owusu, Majid Ezzati, Raphael Arku, Brian Robinson, Alexandra M. Schmidt, Jill Baumgartner

Abstract

5.3.1#### Spatial patterns of deprived ‘slum’ areas in GAMA Our predictive model identified 899 urban EAs (19% of all urban EAs) in GAMA with a high predicted probability (≥0.80) of being deprived, nearly half of which were located in EAs outside of AMA (n = 423). These EAs were geographically dispersed across most of GAMA except for peri-urban areas in the north which are administratively considered ‘rural’ in the census and thus excluded from prediction. Based on population data from the 2010 census, we estimated that approximately one in five residents (745,714 people) in urban GAMA lived i

A figure from Identifying deprived “slum” neighbourhoods in the Greater Accra Metropolitan Area of Ghana using census and remote sensing data — fig. from the paper

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

A Bayesian spatial model using publicly available census and satellite data identified 899 (19%) urban enumeration areas in Greater Accra as deprived, affecting about 22% of residents.

What’s new

i.Integrated publicly available census and remote sensing data to map deprived neighborhoods.
ii.Developed a Bayesian spatial modelling framework that provides uncertainty estimates and accounts for spatial random effects.
iii.Generated fine‑grained enumeration‑area level classifications for the Greater Accra Metropolitan Area, scalable to other regions.

How they did it

The authors combined UN field‑mapped slum data from Accra’s core with housing, population, and environmental variables from the latest census and satellite imagery. They fitted a Bayesian logistic regression model on 2,414 enumeration areas to learn predictors of deprivation, then applied the model to predict probabilities for all 4,615 urban enumeration areas in Greater Accra.

What they found

899 urban enumeration areas (19% of 4,615) had a high predicted probability (≥80%) of being deprived.
These areas contain an estimated 745,714 residents, representing 22% of the urban population.
Deprived areas were characterized by higher population density, lower elevation, less vegetation, and reduced access to indoor piped water and sanitation.

Caveats

✕The study did not extend predictions to rural/peri‑urban enumeration areas, which may have different characteristics.
✕A conservative probability cut‑off (≥0.80) may have missed some deprived neighborhoods; lowering the threshold increased the count.
✕Residual spatial confounding between neighbourhood‑level random and fixed effects cannot be ruled out.

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Identifying deprived “slum” neighbourhoods in the Greater Accra Metropolitan Area of Ghana using census and remote sensing data

The paper, in five minutes

Ask a question about this paper

Similar papers in all four archives

Similar papers in all four archives

The paper, in five minutes