The study aims to explore the dynamics of neighbourhood quality of life in urban residential neighbourhoods in Indian cities. Large scale urban data on various facets of neighbourhood become major stakeholders in such an analysis. The study utilizes data on prioritization of neighbourhood attributes for establishing a framework for optimization of neighborhood Quality of life. Qualitative research tools such as literature review and analysis is utilized initially to establish a theoretical framework for evaluation of quality of life at the neighbourhood level. A major chunk of the study relies on empirical studies with primary data collection to construct an empirical framework in conjunction with the theoretical base established earlier using SPSS software and Microsoft Excel for data visualization and analysis. Artificial neural networks analysis is used to decode the multivariate data and establish a predictive model towards neighbourhood quality of life. Grassroots level urban planning can be institutionalized using the framework along with crowd sourced data on resident's perception of their neighbourhoods.
According to the World Health Organization, Quality of Life(QoL) is defined as
“an individual's perception of their position in life in the context of the culture and
value systems in which they live and in relation to their goals, expectations, standards
and concerns.” WHO’s conceptualization of Quality of life comes across as a broad
ranging concept bearing complex relationships with the person's physical health,
psychological state, personal beliefs, social relationships and their interactions with
salient features of their environment. Research literature acknowledges that
neighbourhoods are acceptable unit of analysis to efficiently measure the local conditions that
impact various domains of human life.
From a planning perspective, a neighbourhood can be defined as a composition of people, place and identity. Consequently, Quality of life for the neighbourhood should be composed of people’s preferences, physical attributes which contribute to the place and community attributes which define the neighborhood’s identity. There is a clear research gap when it comes to the scale, context and conceptual expanse of the concept quality of life when applied to urban residential neighbourhoods of a thriving Indian city. Research literature appears to be severely conflicted when it comes to a comprehensive formulation of the concept of quality of life at the neighbourhood level. Most studies present a piecemeal view whereby they cover only one aspect of the people-place-identity triad. Most importantly, we find that the indicators used in these studies can be best evaluated at the city level and efforts to measure them at the neighbourhood scale may often give inconclusive results. Lastly, most of the studies originate in the global north where the socio cultural and urban form constraints are vastly different from the global south. It will perhaps be erroneous to apply the same in the context of dense, bustling neighbourhoods in Indian cities.
Urban Planning literature has abundant references to terms like Urban Quality of life, Liveability, area attractiveness, Social sustainability, neighborhood satisfaction. Each term in its own way tries to measure the desirability of living conditions in a given area. The variables included within each concept differ with the scope and the overall bent of the study.
Mulligan, Carruthers (2005) define QoL as the satisfaction that a person receives
from surrounding human and physical conditions which are scale-dependent and can
affect the behavior of individual people, groups such as households and economic
units such as firms. Marans, Stimson (2011) stress upon the importance of QoL in
estimating life satisfaction and happiness for individuals as well as communities. The
broad based nature of QoL was further summed up by
A review of literature related to the above three concepts suggest that though similar in overall intent there are significant differences between the concepts. While QoL is a broad based, multi dimensional concept, it is not necessarily place based. Liveability, on the other hand is an entirely place based concept which is usually employed for large urban areas. Liveability takes into account a large number of diverse indicators many of which may be slightly beyond the realm of urban planning itself. Social sustainability appears to be a community based concept which looks at both physical as well as social components of community life. A large number of diverse indicators have been suggested for measuring social sustainability and liveability in research literature.
There is an equal amount of confusion and contradictions when it comes to
quantitative measurement of QoL and its allied concepts. The following table highlights
some of the main methods specified in literature to quantify these concepts. The
indicator approach seems to be the most popular amongst researchers where the broader
concept is broken down into a series of quantifiable indicators
which focuses on comparing different urban areas according to a number of objective indicators assumed to reflect quality of life. Ad hoc weighting schemes were employed.
The market/resident approach in which housing price and/or wage differentials are theorized to compensate for quality-of-life differences between urban areas. Theoretical weighting based on resident’s preferences were used.
The subjective approach which is specifically designed to collect primary data at the disaggregate or individual level using social survey methods where the focus is on the peoples’ befrom official governmental data collections, including the census. This is an approach that is often associated with social indicators research.
of indicators –usually over time – derived from aggregated spatial data using official sources, such as the census, that are said to be related to perceived QOL
Marans, Stimson (2011)- Indicator based versus modeling approach haviors and assessments, or evaluations of aspects of QOL.
Modeling relationships between characteristics of the urban environment and measures of peoples’ subjective assessments of QOL domains, including their satisfaction with specific phenomena and with life as a whole. This approach typically involves data collected through survey research methods and analyzed using techniques such as regression analysis or structural equation models.
Blečić, Ivan, Talu. (2013)- Countability versus capability approach Countability approach: based on inputs or outputs
Several researchers have tried to assess the quality of life offered by urban
residential neighbourhoods. Research literature suggests that the neighbourhood attributes
that ascertain preference for one neighborhood above other branch out into distinct
categories. Social features such as community satisfaction
The multitudes of attributes which determine the character of a neighbourhood
have been well documented in literature.
Structural Type, scale, materials, design, state of repair, density, characteristics landscaping, etc. in the neighbourhood
Infrastructural Roads, sidewalks, streetscaping, utility services, etc. characteristics
Demographic Age distribution, family composition, racial, ethnic, and characteristics
Class status characteristics
Tax/public service package characteristics
Environmenta l characteristics
Proximity characteristics
characteristics religious types, etc. Of the resident population:
Income, occupation and education composition of the resident population
The quality of safety forces, public schools, public administration, parks and recreation, etc., in relation to the local taxes assessed
Degree of land, air, water and noise pollution, topographical features, views, etc.
Access to major destinations of employment, entertainment, shopping, etc., as influenced by both distance and transport infrastructure.
The degree to which local political networks are mobilised, residents exert influence in local affairs through spatially rooted channels or elected representatives
Local friend and kin networks, degree of inter household familiarity, type and quality of interpersonal associations, residents’ perceived commonality, participation in locally based voluntary associations, strength of socialisation and social control forces, etc.
Residents’ sense of identification with place, historical significance of buildings or district, etc.
With the exception of demographic, class status and political and sentimental characteristics, all other categories in the table shown above, fall into the realm of Urban Planning. However, when viewed at the neighbourhood scale we find that Environmental and Proximity characteristics are inconclusive since these are macro operators which depend on city scale and structure. Of the remaining characteristics, Infrastructural and Tax/public service (to a large extent) characteristics are mostly dependent on the whims of the government, often constrained by monetary considerations in the Indian scenario even though ideally they should be under control of the urban planner. Overall, the structural, socio interactive and infrastructural characteristics continue to be the areas of intervention from the point of view of urban planning in the context of existing urban residential neighbourhoods. An assessment of quality of life at the neighbourhood level necessitates an investigation of the above attributes along with their components and sub components.
People based dwelling unit size etc.
Housing typology Spatial character Density Development controls Visual character Roads, water supply, drainage, sewage systems, solid waste management systems, public transit stops etc.
Parks, Playgrounds, schools, health facilities, small retail, chemist shop etc.
Quality and quantity of public space Community interaction
A glance at the neighbourhood attributes and their multiple relationships with quality of life in the neighbourhood shows that there is a need for a clear empirical framework to evaluate QoL. Though we cannot undermine the impact of qualitative attributes, it is the quantitative attributes which can be directly included in the master planning process. It is clear from the review of literature that Housing characteristics, spatial character, Density, development controls; Infrastructural characteristics and socio interactive characteristics are necessary ingredients in formulation of any framework to evaluate QoL at the neighbourhood level. Density appears to be a dominant factor and though it has clear links with QoL, the exact nature of the relationship (whether positive or negative) is inconclusive in literature. Density also finds itself as a backdrop for most QoL studies because it is in stressed conditions that QoL studies find their real relevance. The findings suggest that perhaps High density environments would be the best context to carry out Quality of life studies in the urban setting. Visual character and housing typology are often the perceptual and physical manifestations of density. Hence these can also be treated as context for carrying out QoL studies. Of the remaining attributes, the infrastructural (social) and socio interactive attributes need a tool for empirical evaluation and quantification. Overall we can conclude that, Quality of life at the neighborhood level may be expressed as an aggregate of the impact of structural, infrastructural (social) and socio interactive characteristics. Overall satisfaction with the neighbourhood as reported by the residents may be treated as a surrogate for the overall quality of life offered by the neighbourhood. Aggregated Manifestation of Neighbourhood Quality of life In High density environments categorized by specific visual character and typology =
Structural Characteristics Housing Characteristics P Structural Quality of Life
Typology
Urban Form P + Social
Physical
+ P Social Quality of Life People based P
Place Based P Fig 1. Method and Tools employed for Formulation of NQI. SOURCE: Author
Bringing back our initial conceptualization of neighbourhood quality in terms of people, place and identity, we find that spatial character and development control impacts give a true representation of the place. The identity/community aspect is more or less revealed in the socio interactive characteristics and the access to social infrastructure. An examination of most of these attributes from the resident’s opinion facilitates the fulfilment of the people aspect. Most of the studies in literature attempt to visualize neighbourhood quality of life using either one or two of the people-placeidentity triad. An attempt at consolidating all the attributes mentioned above into an empirical framework can be a significant contribution of this study.
Neighbourhood Quality Index is proposed as a composite index that aggregates the structural, social infrastructural and socio interactive characteristics of the neighbourhood. Where, Pi- Normalized value of neighbourhood quality parameter
Wi- Normalized weightage of neighbourhood Quality parameters based on its relative contribution towards overall satisfaction with neighbourhood. The following indicators were identified for evaluating neighbourhood social quality after review of literature
These indicators formed the basis for preparation of structured questionnaires for an expert opinion survey (EOS). The EOS questionnaire asked the experts to rate the listed given indicators on a scale of 1 to 5 based on the importance of the given indicator in determining the social quality of an urban residential neighborhood. A total of 52 surveys were conducted each with ratings for a set of 38 indicators. In order to make the sample variable ratio more focused for further analysis, an initial screening of the indicators was carried out on the basis of mean values of importance ratings as given by the experts. Indicators which scored less than 3.5 as mean importance rating were removed from the matrix put forward for further analysis. Furthermore indicators related to travel times to social infrastructure were excluded in favor of indicators which judged the qualitative aspects of the social infrastructure.
Four High density neighbourhoods in Bangalore namely Mattikere, Mahalakshmipuram, Gurappanapalya and Kammanahalli were selected as case study areas for data collection regarding the individual indicators. These 4 neighbourhoods have several common characteristics in terms of homogeneity in population density, area, plotted development(non slum) and primarily residential landuse. A reconnaissance survey during the initial stages of the research had shown that despite their commonalities the neighbourhoods offered varying quality of life to its residents. A total of 270 household surveys were conducted using random sampling to collect data regarding the shortlisted neighbourhood attributes. The final data set with 8 indicators (52 X 8=416 data points) was further put through SPSS for statistical data reduction through factor analysis.
Fig 2. Factor Analysis results generated in SPSS
SPSS was used to generate a correlation matrix where it was seen that several
correlations in the matrix were above the minimal thumb rule value of ±0.3 and above.
The results of KMO and Bartlett test for sampling adequacy revealed a KMO measure
of 0.55 and significance <0.05 which verified the adequacy of the data for proceeding
with factor analysis
The analysis revealed that a total of 3 factors (components) account for around 69.612% of variance in the data. The above factor analysis gave us the indicators which are deemed necessary for defining neighborhood quality. Based on the authors’ understanding each of the factors has been allocated a name viz. Access to Space, Community Linkage, Urban Form. To reduce the multitudes of components into a list of prioritized components and allocate weightages to each component, the procedure shown in Table 8 has been followed. The structural validity for the index has been further reinforced on the basis of artificial neural networks based modeling.
A neural network is a powerful computational data model that is able to capture and represent complex input/output relationships. The motivation for the development of neural network technology stemmed from the desire to develop an artificial system that could perform "intelligent" tasks similar to those performed by the human brain such as: 1. A neural network acquires knowledge through learning. 2. A neural network's knowledge is stored within inter-neuron connection strengths known as synaptic weights.
The true power and advantage of neural networks lies in their ability to represent both linear and non-linear relationships and in their ability to learn these relationships directly from the data being modeled. The most common neural network model is the Multilayer Perceptron (MLP). This type of neural network is known as a supervised network because it requires a desired output in order to learn. The goal of this type of network is to create a model that correctly maps the input to the output using historical data so that the model can then be used to produce the output when the desired output is unknown.
Artificial Neural networks analysis has been used to generate a Predictive model that determines the relationship between overall satisfaction with neighborhood and parameters of neighborhood quality. The ANN analysis also helps in Estimation of relative importance of each parameter in determining overall satisfaction with neighborhood. 2.2
parameters of neighborhood quality with neighborhood and
The neighbourhood quality parameters selected through statistical analysis on expert opinion survey data manifest themselves in the neighbourhood in form of overall satisfaction with the neighbourhood. The parameters selected are a hybrid mix of physical and social components of neighbourhood quality of life. In order to assess the selected parameters and their relative contribution towards overall satisfaction drawn from the neighbourhood we need to carry out multivariate analysis and data modeling. The model proposes that Overall satisfaction with neighbourhood is a function of the neighbourhood quality parameters. Here, the Dependent variable is Overall satisfaction with neighbourhood derived from household survey data. Neighbourhood quality parameters from Household survey data constitute the Independent variables. A 3-layer feed forward Artificial Neural networks analysis employed to verify the validity of the proposed model. The ANN analysis studies the underlying data structure and derives the structural relationship for use in predictive modeling. A total of 239 x 7=1673 data points were input the neighbourhood quality parameters. The ANN analysis is a two stage analysis where it was reported that the model was able to predict with an accuracy of 84.8% in the training phase. In the testing phase, the model achieved an accuracy of prediction amounting to 76.7%. The ANN analysis also generates normalized importance for the independent parameters based on their relative contribution towards the Dependent variable. These values may be used as weightages for formation of Neighbourhood Quality Index. perception of neighborhood convenience perception of neighborhood attractiveness Living space -average floor area per person Perception of neighborhood convenience Perception of neighborhood attractiveness
The study contributes in a twofold way to the knowledge and practice of urban planning. On the theoretical level, the major contributions of the study would be to propose a new paradigm for evaluation of quality of life offered by a neighbourhood in the context of Indian cities. A neighbourhood is composed of people, place and social life within the place. An evaluation of each of these components is necessary in order to present a holistic picture of the quality of life offered by the neighbourhood. The study introduces a new paradigm for the same, namely- Neighbourhood Quality. The concept of neighbourhood quality aims at an empirical formulation of an otherwise subjective concept. The second contribution of the study is towards the practice of urban planning at the neighbourhood as well as city level. Quantification of neighbourhood quality and its various sub components can then be used as a guiding tool towards optimization of quality of life in the city. The urban planning guidelines which emerge out of the study can be active contributors towards ensuring well being and quality of life at the neighbourhood level despite rapid intensification in population and building. The Neighborhood Quality concept described here can become an active tool for micro level planning and allocation of city resources towards targeted development of the disadvantaged neighbourhoods.