Flexible Spatial Multilevel Modeling of Neighborhood Satisfaction in Beijing

TLDR: Zhang et al. as mentioned in this paper developed an innovative and flexible Bayesian spatial multilevel model to examine the sociospatial variations in perceived neighborhood satisfaction using a large-scale household satisfaction survey in Beijing.

Abstract: This article develops an innovative and flexible Bayesian spatial multilevel model to examine the sociospatial variations in perceived neighborhood satisfaction, using a large-scale household satisfaction survey in Beijing. In particular, we investigate the impact of a variety of housing tenure types on neighborhood satisfaction, controlling for household and individual sociodemographic attributes and geographical contextual effects. The proposed methodology offers a flexible framework for modeling spatially clustered survey data widely used in social science research by explicitly accounting for spatial dependence and heterogeneity effects. The results show that neighborhood satisfaction is influenced by individual, locational, and contextual factors. Homeowners, except those of resettlement housing, tend to be more satisfied with their neighborhood environment than renters. Moreover, the impacts of housing tenure types on satisfaction vary significantly in different neighborhood contexts and spatial loc...

Figures

Table 2. Model comparison results

Table 3. Estimation results from the spatial multilevel model

Figure 2. Map displays the spatial pattern of the district-level random effects in urban Beijing

Figure 1. Map displays the spatial pattern of neighbourhood satisfaction (on a fivepoint scale) in urban Beijing

Table 1. Summary of socio-economic and demographic variables by homeownership types

Table 4. A sensitivity analysis of key model parameter estimation using different hyperpriors

Full text

This is a repository copy of Flexible Spatial Multilevel Modelling of Neighbourhood
Satisfaction in Beijing.
White Rose Research Online URL for this paper:
http://eprints.whiterose.ac.uk/112014/
Version: Accepted Version
Article:
Ma, J., Chen, Y. orcid.org/0000-0002-7694-4441 and Dong, G. (2017) Flexible Spatial
Multilevel Modelling of Neighbourhood Satisfaction in Beijing. Professional Geographer.
ISSN 0033-0124
https://doi.org/10.1080/00330124.2017.1298453
This is an Accepted Manuscript of an article published by Taylor & Francis in Professional
Geographer on 10 April 2017, available online:
http://www.tandfonline.com/10.1080/00330124.2017.1298453
eprints@whiterose.ac.uk
https://eprints.whiterose.ac.uk/
Reuse
Items deposited in White Rose Research Online are protected by copyright, with all rights reserved unless
indicated otherwise. They may be downloaded and/or printed for private study, or other acts as permitted by
national copyright laws. The publisher or other rights holders may allow further reproduction and re-use of
the full text version. This is indicated by the licence information on the White Rose Research Online record
for the item.
Takedown
If you consider content in White Rose Research Online to be in breach of UK law, please notify us by
emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request.

1
Flexible Spatial Multilevel Modelling of Neighbourhood Satisfaction in
Beijing
Jing Ma
Faculty of Geographical Science, Beijing Normal University, China
Yu Chen
School of East Asian Studies, University of Sheffield, UK
Guanpeng Dong
Department of Geography and Planning, University of Liverpool, UK

2
Flexible Spatial Multilevel Modelling of Neighbourhood Satisfaction in
Beijing
Abstract:
This paper develops an innovative and flexible Bayesian spatial multilevel model to
examine the socio-spatial variations in perceived neighbourhood satisfaction, using a
large-scale household satisfaction survey in Beijing. In particular, we investigate the
impact of a variety of housing tenure types on neighbourhood satisfaction, while
controlling for household and individual socio-demographic attributes and geographical
contextual effects. The proposed methodology offers a flexible framework for
modelling spatially clustered survey data widely used in social science research by
explicitly accounting for spatial dependence and heterogeneity effects. The results show
that neighbourhood satisfaction is influenced by individual, locational and contextual
factors. Homeowners, except those of resettlement housing, tend to be more satisfied
with their neighbourhood environment than renters. Moreover, the impacts of housing
tenure types on satisfaction vary significantly in different neighbourhood contexts and
spatial locations.
Key Words: spatial statistics, multilevel modelling, neighbourhood satisfaction,
housing tenure, Beijing

3
Chinese cities have experienced enormous housing and neighbourhood changes as the
country transits from a centrally-planned economy to a market one. Before 1978, the
majority of urban residents rented housing from their work units. The subsequent
housing reforms resulted in various housing tenure types and significant socio-
economic stratification. A large number of studies have examined the consequences of
the housing reforms, such as improved housing conditions and rising inequalities (Wang
and Murie 2000; Huang and Jiang 2009; Logan et al. 2010). Relatively few studies
focus on residents’ perceptions of residential environments as a result of significant
housing and neighbourhood changes. Nonetheless, it is important to research
neighbourhood satisfaction as it reflects neighbourhood quality, and has significant
impacts on overall life satisfaction (e.g. Ibem and Aduwo 2013).
This paper aims to fill the gap by examining the spatial patterns and
determinants of neighbourhood satisfaction, especially, the impacts of housing tenure
types. This is important because China’s housing reforms result in a variety of housing
tenure types that differ in neighbourhood environments, especially in terms of services
and facilities, access to transportation nodes, and geographic location relative to the city
centre. The study will enhance our understanding of how the housing reforms are
experienced by individuals through their subjective evaluation of residential
environment. It is also conducive to policies aimed at delivering better residential
environments.
We develop an innovative and flexible spatial multilevel modelling approach to
examine the determinants of neighbourhood satisfaction while controlling for potential
group dependence, spatial correlation and heterogeneity effects. Our data come from a
large-scale household satisfaction survey in Beijing. Similar to other surveys with
clusters presented by spatial units, our data are both hierarchical and spatial in nature

4
(Dong et al. 2016). Hierarchically, respondents nest into districts, potentially leading to
within-district dependencies. That is, neighbourhood satisfaction levels of individuals in
the same district tend to be more similar than those from different districts. This is often
termed group dependence effect and modelled using the multilevel approach (e.g.
Raudenbush and Bryk 2002;
Goldstein 2003). Spatially, the higher-level geographical
units (e.g. districts) might not be independent and thus their effects upon individuals
could be spatially correlated in a way that respondents in closer districts tend to report
similar levels of neighbourhood satisfaction (Haining 2003). Moreover, relationships
between certain variables might vary across geographical contexts because of either
generic contextual differences or unmodelled geographical unobservables. By using a
rigorous spatial multilevel modelling approach which accounts for both within-district
dependence and between-district spatial correlation and heterogeneity, we provide
robust evidence that neighbourhood satisfaction is influenced by individual, locational
and contextual factors. Meanwhile, neighbourhood satisfaction exhibits significant
spatial clustering patterns, and heterogeneous associations between housing tenure types
and neighbourhood satisfaction are found in urban Beijing.
In the following sections we first review previous studies on neighbourhood
satisfaction and then locate our study into the Chinese context by outlining the housing
reforms and different housing tenure types. This is followed by the introduction of the
spatial multilevel method and the Beijing survey. We then discuss the empirical
findings about spatial patterns and determinants of neighbourhood satisfaction, with
conclusions at the end.

Frequently asked questions

Q1. What have the authors contributed in "Flexible spatial multilevel modelling of neighbourhood satisfaction in beijing" ?

This paper develops an innovative and flexible Bayesian spatial multilevel model to examine the socio-spatial variations in perceived neighbourhood satisfaction, using a large-scale household satisfaction survey in Beijing. In particular, the authors investigate the impact of a variety of housing tenure types on neighbourhood satisfaction, while controlling for household and individual socio-demographic attributes and geographical contextual effects. 

Q2. What are the other hyperpriors used to test the sensitivity of the variance estimates?

For the two district-level variance parameters, other hyperpriors including logGamma (1,0.1), logGamma (1,0.01), and logGamma (1,0.001) are used to test the sensitivity of the variance estimates. 

Q3. What are the factors that influence the satisfaction of a person with their neighbourhood?

Individual factors include age, gender, marital status, education, family composition and household income, as they influence an individual’s needs and expectations of the neighbourhood environment. 

Q4. What is the effect of a fixed part of a housing tenure variable?

For a specific housing tenure variable (e.g., owner of commodity housing), its effect is divided into two parts: a fixed part βp and a random part θk,p that varies across districts. 

Q5. What is the importance of a careful consideration of geographical contexts in the analysis?

These findings demonstrate spatial heterogeneity between tenure types and neighbourhood satisfaction and the importance of a careful consideration of geographical contexts inthe analysis.