ArticlePDF Available

People and places: The multilevel model as a general framework for the quantitative analysis of geographical data

Authors:
Kelvyn Jones at University of Bristol
Kelvyn Jones
Craig Duncan
Craig Duncan
Craig Duncan
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Download full-text PDFDownload full-text PDF
Read full-text
Download citation
Jones, Duncan - 1996 - People and places the multilevel model as a general framework for the quantitative analysis of geographical da
ta.pdf
Content uploaded by Kelvyn Jones
Author content
All content in this area was uploaded by Kelvyn Jones on Feb 20, 2017
Content may be subject to copyright.
Jones, Duncan_1996_People and places the multilevel model as a general framework for the quantitative analysis of geographical da
ta.pdf
Content uploaded by Kelvyn Jones
Author content
All content in this area was uploaded by Kelvyn Jones
Content may be subject to copyright.
A preview of the PDF is not available
... A form of the multivariate model in Chapter 7 can be used to simultaneously model occurrence (do you smoke?) and quantity (how many smoke?) (see Jones & Duncan, 1996), though this mixed multivariate model is not available in Mplus. ...
Understanding smoking behaviour in LGB populations across Great Britain: Quantitative analyses using secondary data
Thesis
  • Jan 2020
This thesis is the first to examine the associations between sexuality and smoking behaviour in Great Britain, using quantitative analyses of secondary data sources. Drawing on previous literature from geography, psychology and epidemiology, the thesis aims to understand the factors associated with greater smoking prevalence in lesbian, gay and bisexual (LGB) populations, how LGB smoking has changed over time, and whether different measures of sexuality impact smoking and co-behaviour likelihood. Whilst recent calls to reduce smoking prevalence in Great Britain have acknowledged inequalities between LGB and heterosexual populations, few studies in Great Britain have examined smoking trends and patterns in this population beyond prevalence rates. Much of the previous research looking at sexuality and smoking behaviour has been carried out in the United States, where contextual differences mean findings cannot be generalised. This thesis uses multilevel modelling to examine individual and environmental factors in smoking behaviour and patterns to draw comparisons across Great Britain between LGB and heterosexual populations. Findings suggest that, in Great Britain, inequalities in smoking persist between LGB people and heterosexual populations, and also within LGB groups. Across Great Britain, LGB people in certain areas may be more affected by smoking inequalities, and sexual minority women are amongst the most affected. Anti-smoking policies should address these inequalities by taking into account variations by place and recognising that sexual minority populations are more vulnerable.
View
... First, the residential environmental satisfaction survey data may have been influenced by other predictors of residential satisfaction (Van Duijn and Rouwendal, 2012); for example, wealthier and more educated individuals have a greater preference for living near urban heritage sites (Koster et al., 2016). A further in-depth analysis that considers the socioeconomic and demographic traits of residents is necessary to avoid aggregation errors (Jones and Duncan, 1996). Second, a cross-sectional dataset from 2016 was employed, without a time-series analysis, because longitudinal data were not available. ...
Urban Heritage and its Socioeconomic Impacts in Seoul, South Korea: An Empirical Study Using Residential Environmental Satisfaction and Housing Price as Indicators
Article
Full-text available
  • Jun 2022
Background and objective: Recent research has highlighted the need for urban heritage conservation due to rapid urbanisation, especially in Asian cities; however, few studies have investigated the socioeconomic impacts of heritage assets in urban contexts. This study examined urban heritage sites in Seoul Metropolitan City from the perspective of residents' environmental satisfaction and housing prices. Methods: A spatial regression model was developed to examine the associations between urban heritage sites and their corresponding protected areas as the independent variables (nationally assigned cultural heritage, city-assigned cultural heritage, nationally registered cultural heritage, nationally assigned cultural heritage protected area, and city-assigned cultural heritage protected area) and residential environmental satisfaction and housing prices as the dependent variables. The model investigated how urban heritage sites influence housing prices through the mediating effect of residential environmental satisfaction. Results: The results confirmed the impact of urban heritage sites on housing prices and the mediating effect of residential environmental satisfaction. Moreover, depending on their urban heritage classifications, noticeable differences were evident in the impact of urban heritage sites. Conclusion: These findings provide an intellectual foundation for public policies, offering insights into how they might achieve an optimum balance between private and public interests in matters of heritage conservation.
View
... The advantage of using the multilevel structure is the ability to estimate the variability in results that can be attributed to neighborhood (e.g., community area) effects rather than only to individual station effects. By carefully controlling variable-inclusion at the appropriate level, the model considers correlations between observations within the same group (i.e., a given community area) as distinct from correlations between groups (Jones & Duncan, 1996). In contrast, a standard one-level regression model would ignore group-level distinctions (e.g., different commuting patterns in different communities) and group-level correlations (e.g., similar patterns of use among stations in the same community related to the income-level of riders). ...
Does ridesourcing respond to unplanned rail disruptions? A natural experiment analysis of mobility resilience and disparity
Preprint
Full-text available
  • Feb 2022
Urban rail transit networks provide critical access to opportunities and livelihood in many urban systems. Ensuring that these services are resilient (that is, exhibiting efficient response to and recovery from disruptions) is a key economic and social priority. Increasingly, the ability of urban rail systems to cope with disruptions is a function of a complex patchwork of mobility options, wherein alternative modes can complement and fill service gaps. This study analyzes the role of ridesourcing in providing adaptive mobility capacity that could be leveraged to fill no-notice gaps in rail transit services, addressing the question of distributional impacts of resilience. Using a natural experiment, we systematically identify 28 major transit disruptions over the period of one year in Chicago and match them, both temporally and spatially, with ridesourcing trip data. Using multilevel mixed modeling, we quantify variation in the adaptive use of on-demand mobility across the racially and economically diverse city of Chicago. Our findings show that the gap-filling potential of adaptive ridesourcing during rail transit disruptions is significantly influenced by station-, community-, and district-level factors. Specifically, greater shifts to ridesourcing occur during weekdays, nonholidays, and more severe disruptions, in community areas that have higher percentages of White residents and transit commuters, and in the more affluent North district of the city. These findings suggest that while ridesourcing appears to provide adaptive capacity during rail disruptions, its benefits do not appear to be equitable for lower-income communities of color that already experience limited mobility options.
View
... The advantage of using the multilevel structure is the ability to estimate the variability in results that can be attributed to neighborhood (e.g., community area) effects rather than only to individual station effects. By carefully controlling variable-inclusion at the appropriate level, the model takes into account correlations between observations within the same group (i.e., a given community area) as distinct from correlations between groups (Jones and Duncan, 1996). Instead, a standard one-level regression model would ignore group-level distinctions (for example, different commuting patterns in different communities) and group level correlations (for example, similar patterns of use among stations in the same community related to the income-level of riders). ...
Disparities in ridesourcing demand for mobility resilience: A multilevel analysis of neighborhood effects in Chicago, Illinois
Preprint
  • Oct 2020
Mobility resilience refers to the ability of individuals to complete their desired travel despite unplanned disruptions to the transportation system. The potential of new on-demand mobility options, such as ridesourcing services, to fill unpredicted gaps in mobility is an underexplored source of adaptive capacity. Applying a natural experiment approach to newly released ridesourcing data, we examine variation in the gap-filling role of on-demand mobility during sudden shocks to a transportation system by analyzing the change in use of ridesourcing during unexpected rail transit service disruptions across the racially and economically diverse city of Chicago. Using a multilevel mixed model, we control not only for the immediate station attributes where the disruption occurs, but also for the broader context of the community area and city quadrant in a three-level structure. Thereby the unobserved variability across neighborhoods can be associated with differences in factors such as transit ridership, or socio-economic status of residents, in addition to controlling for station level effects. Our findings reveal that individuals use ridesourcing as a gap-filling mechanism during rail transit disruptions, but there is strong variation across situational and locational contexts. Specifically, our results show larger increases in transit disruption responsive ridesourcing during weekdays, nonholidays, and more severe disruptions, as well as in community areas that have higher percentages of White residents and transit commuters, and on the more affluent northside of the city. These findings point to new insights with far-reaching implications on how ridesourcing complements existing transport networks by providing added capacity during disruptions but does not appear to bring equitable gap-filling benefits to low-income communities of color that typically have more limited mobility options.
View
... The MLM decomposed the total variance into different levels in the context phenomenon [53]. In addition, it operated on multiple scales or levels, so the overall model could include the microscale of the individual and the macroscale of the population [54]. ...
Modelling and prediction of global non-communicable diseases
Article
Full-text available
  • Jun 2020
  • BMC PUBLIC HEALTH
Background: Non-communicable diseases (NCDs) are the main health and development challenge facing humankind all over the world. They are inextricably linked to socio-economic development. Deaths caused by NCDs should be different in different socio-economic development stages. The stratified heterogeneity of NCD deaths is currently not fully explored. Methods: Countries were classified according to their socio-economic types and development stages, which were illustrated as a tree-like structure called Geotree. NCD deaths were linked to the countries and so were attached to the Geotree, which was modelled by a multilevel model (MLM) approach. Accordingly, the levels of NCD death indexes were predicted for 2030. Results: Through the Geotree structure constructed in the study, it can be seen that the NCD death index has obvious stratified heterogeneity; that is, the NCD death index shows different trends in different country types and socio-economic development stages. In the first-level branches (country type), as national income increases, NCD mortality rate decreases and the proportion of NCD deaths to total deaths increases. In the secondary-level trunks (socio-economic development stage), as a country's development stage rises, the NCD mortality rate decreases and the proportion of NCD deaths to total deaths increases. In addition, combined with the hierarchical nature of the evolution tree model, the MLM was used to predict the global NCD death index for 2030. The result was that by 2030, the global average age-standardized NCD mortality rate would be 510.54 (per 100,000 population) and the global average mortality for NCD deaths of the total number of deaths would be 75.26%. Conclusions: This study found that there is a significant association between socio-economic factors and NCD death indicators in the tree-like structure. In the Geotree, countries on the same branch or trunk can learn from countries with higher development stages to formulate more effective NCD response policies and find the right prevention and treatment path.
View
... Scale is a key geographical-cartographical concept; the realisation that relationships are not necessarily consistent across spatial scales stimulated methods of spatial decomposition (Haggett, 1964;Moellering and Tobler, 1972;Jones and Casetti, 1992). A major extension came with adoption of multilevel modelling (Jones et al., 1992;Jones and Moon, 1993), developed in educational research for examining the nature and strength of relationships between variables at several scales (Jones, 1991;Jones and Duncan, 1996), with geographers involved in the software developments at the Centre for Multilevel Modelling 4 and their application in studies of, for example, scalar variations in residential segregation (Jones et al., 2015(Jones et al., , 2018 and political polarisation ) -modified to uncover interactions among variables in multivariate models , or intersectionality (Green et al., 2017;Evans et al., 2018). ...
Quantitative methods II: How we moved on – Decades of change in philosophy, focus and methods
Article
  • Aug 2019
The first of these three reports reprised human geography’s theoretical and quantitative revolutions’ origins, covering the philosophy, focus and methods that dominated their early years. Over the subsequent decades the nature of work categorised as quantitative human geography changed very considerably – in philosophy, focus and methods. This second report summarises those changes, highlighting the main features of the extensive volume of work published over the last five decades, as a prelude to the final report that will focus on the contemporary nature of quantitative human geography and its likely futures.
View
... On the other hand, there are main differences in terms of the structural behaviour between higher-level units (Arbués, Baños, Mayor, & Suárez, 2016;Bhat, 2000;Fazio & Piacentino, 2010). Furthermore, it is necessary to disentangle the heterogeneity among higher-level units and individual heterogeneity (Jones & Duncan, 1996). Summing up, multilevel modelling allows one to more reliably analyse the extent to which differences in employment probability are due to differences in jobseeker characteristics and/or differences in the spatial environment, which are based on the characteristics of specific postal codes and jobcentres. ...
The spatial structure of the labour market across public jobcentres. Does their accessibility matter?
Article
  • Oct 2018
  • PAP REG SCI
The objective is to analyse how one’s place of residence affects the probability of finding a job and to measure the definition of the public jobcentre catchment area, which contributes to improving labour outcomes in the most deprived areas. We propose a multilevel model to estimate the probability of finding a job controlling for individual characteristics and discerning the effect of the place of residence and the contribution of public employment centres. We use an administrative register of jobseekers (70,379) grouped by 384 zip codes and 24 jobcentres. The econometric results confirm the hypothesis that there is a strong residence effect that is not sufficiently mitigated by public employment services.
View
View
View
View
ResearchGate has not been able to resolve any references for this publication.