Content uploaded by Fernando Bucheli
Author content
All content in this area was uploaded by Fernando Bucheli on Oct 06, 2022
Content may be subject to copyright.
Research Article
The influence of the ‘Fragmented City’on wellbeing: Do mixed
neighbourhoods affect young adults' life trajectories in Bogota?
Juan Fernando Bucheli
Affiliated lecturer Universidad Externado de Colombia, Calle 12 # 1-17 este, Bogota, Colombia
ARTICLE INFO
Keywords:
Young adults
Capability approach
Urban segregation
Propensity score matching
Bogota
ABSTRACT
This article investigates the effects that living in heterogeneous neighbourhoods may have on well-
being, in order to shed light on the potential consequences of promoting residential mix policies.
Using propensity score matching models, the study investigates whether 120 disadvantaged young
adults located in a heterogeneous neighbourhood in Bogota scored differently on a battery of
indicators of well-being compared to disadvantaged young adults located in a homogenous
neighbourhood. Results suggest that when young adults live close to better-off peers, those in
heterogeneous neighbourhoods are more likely to experience reduced capability scores than those
in homogeneous neighbourhoods.
1. Neighbourhood effects and informational spaces of wellbeing
Quality of urban life, as a measurement of people's achievement and well-being, is an outcome that varies not just between in-
dividuals but also between places. In the literature on neighbourhood effects, place attributes are likely to influence individuals'
trajectories and levels of quality of life. For instance, for those living in deprived neighbourhoods, place may have lasting conse-
quences in relation to how residents perform and achieve outcomes. Social norms, peer influence and role models, collective
socialisation, exposure to violence, and public services (Galster, 2012) are some of the mechanisms by which deprived and
non-deprived places experience differential outcomes, indicating that certain effects can be generated by the intrinsic characteristics
of the places where people live.
As neighbourhoods are considered to be the primary location where people develop their social lives, research has been over-
whelmingly focused on estimating the effects that neighbourhoods have on people's life changes. Since the publication of ‘The Truly
Disadvantaged’(Wilson, 1987), the literature on neighbourhood effects has tested the hypothesis that individuals living in disadvan-
taged neighbourhoods and contexts of urban poverty are more likely to experience worse outcomes than peers living in better-off urban
areas (Brooks-Gunn et al., 1997;Leventhal &Brooks-Gunn, 2000;R.J.Sampson, 2012). Factors that explain how the concentration of
poverty affects individual outcomes are diverse in nature, particularly if the multidisciplinary approach to understanding why space and
place affects the availability of opportunities for disadvantaged populations is taken into account. While the idea that neighbourhoods
matter may appear uncontroversial, Slater (2013), on the contrary, argues that it is people's life choices that affect where they live, and
not the other way around. From this point of view, neighbourhood effect studies should reframe their central claims, as poverty and
inequality within cities may be explained, for instance, by the cost of land and housing in specific areas. In line with this thinking, aspects
such as neighbourhood selection and residential mobility choices have become matters of discussion in order to explain the effects of
living in disadvantaged places.
E-mail address: juan.bucheli@uexternado.edu.co.
HOSTED BY Contents lists available at ScienceDirect
Journal of Urban Management
journal homepage: www.elsevier.com/locate/jum
https://doi.org/10.1016/j.jum.2022.08.003
Received 15 December 2021; Received in revised form 5 May 2022; Accepted 15 August 2022
Available online xxxx
2226-5856/©2022 The Authors. Published by Elsevier B.V. on behalf of Zhejiang University and Chinese Association of Urban Management. This is
an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Journal of Urban Management xxx (xxxx) xxx
Please cite this article as: Bucheli, J. F., The influence of the ‘Fragmented City’on wellbeing: Do mixed neighbourhoods affect young
adults' life trajectories in Bogota?, Journal of Urban Management, https://doi.org/10.1016/j.jum.2022.08.003
Within this debate, Galster (2012) points out that most of the scholarly work on neighbourhood effects has been focused on
researching the effects of the residential environment by looking at the impact on a set of either behavioural outcomes (Atkinson &
Kintrea, 2001;Ellen &Turner, 1997;Jencks &Mayer, 1990) or health outcomes (Pickett &Pearl, 2001;Steinmetz-Wood et al., 2017).
From a critical position, particularly regarding how and where the neighbourhood effect research has been carried out, studies have
mainly either inspected the identification of potential mechanisms through which neighbourhood effects take place, or focused on
quantifying causal pathways for specific individual outcomes. For the latter, it is quite remarkable that little interrogation of specialised
literature has taken place in order to scrutinise the type of outcomes against which effects are evaluated. Indeed, the choice of outcome
seems to have less relevance when it comes to understanding the mechanisms by which the neighbourhood effect operates. Contem-
porary literature on neighbourhood effects has focused widely on quantifying the relationship between place effects and individual
outcomes, and has been considerably less focused on dealing with which, and how, neighbourhood characteristics affect those outcomes
(Ellen &Turner, 1997;Galster, 2012;Levin et al., 2022).
This criticism becomes relevant when a human perspective is taken into consideration in the study of the literature on neigh-
bourhood effects. For instance, scholars (Wilson, 1987;Jencks &Mayer, 1990;R.J.Sampson, 2012) have agreed that urban poverty has
become more concentrated over the years; yet this does not address what sort of evaluative space is under scrutiny in making this claim.
Traditionally, a focus on individual outcomes has been linked to welfarist approaches where an exclusively utilitarian assessment of
well-being is considered, excluding other informative spaces of quality of life that can provide useful insights about how people develop.
At the urban policy level, the assessment of urban life takes an even more utilitarian perspective as outcomes are valued almost
exclusively in monetary terms. This is the case of urban quality of life indicators that place the emphasis of evaluation on counting the
number of goods provided by the urban policy in question (Ble
ci
c et al., 2013), without paying attention to the fact that individuals
differentiate in their personal characteristics and, therefore, in their ability to transform urban assets into human advantage. It is
important to consider that the selection of outcomes to assess differential quality of life between urban settings is not just a mere
methodological decision for researchers, but on the contrary, it constitutes a normative decision with clear paradigmatic consequences,
as the emphasis of evaluation is not only applied to the causes and mechanisms that are behind the productionof neighbourhood effects,
but also to the outcome variables where the effect is demonstrated.
To overcome the limitations of neighbourhood effects literature, which often employs a narrow focus on the availability of
resources as one of the principal measurements of people's advantages, this paper introduces Sen's capability framework (Sen,
1985). The capability approach (CA) considers that the emphasis of evaluation should be focused on a set of valuable ‘being and
doings’which can be measured by the available opportunities (capabilities) and the different combinations of outcomes (func-
tionings) that an individual can achieve. Under the capability framework, the ‘capability to function’is the outcome of interest,
which assesses the effective opportunities people have to undertake those actions they have reason to value. By looking at ca-
pabilities, the evaluative space of individuals' quality of life is expanded as the core of evaluation is no longer under the scrutiny of
utility-based approaches (e.g., happiness) or resourcism (e.g., income and commodities), but instead in assessing to what extent
domains of quality of life have been achieved and what kinds of obstacles may hamper individuals' freedom to choose effective
opportunities.
In this paper, the case of spatial polarisation and fragmentation in Bogota is used to examine how the capability approach can be
useful in assessing individual outcomes of wellbeing in the context of neighbourhood effects. The paper presents the case of hetero-
geneous neighbourhoods as a pattern of residential microsegregation in Bogota to test whether living in better off neighbourhoods
positively affects objective and subjective measurements of quality of life for young adults.
1
The remainder of this paper is organised as
follows. Section 1.2 ‘Urban Policy and Social Mix in Bogota: between polarisation and fragmentation’introduces the background of
social mix policies in Bogota. Section 1.3 ‘Data’presents the data, study area and variables used. Section 1.4 ‘Method and Empirical
Model’describes the method and the empirical model applied. Section 1.5 ‘Results’presents the empirical results. Section 1.6 ‘Dis-
cussion’looks at and interprets the findings. Section 1.7 examines the limitations of the study.
2. Urban Policy and Social Mix in Bogota: between polarisation and fragmentation
Social mixing policies are based on the rationale that contact between deprived households and socioeconomically better-off seg-
ments of the population can promote upward mobility for the former. Galster and Friedrich define the concept as “a state of relative
neighbourhood diversity according to socioeconomic status, in contrast to spatial concentrations of extremely high- or low-status
households”(2015, p. 176). In mixed neighbourhoods, it is expected that a process of microsegregation will emerge due to the pres-
ence of a variety of households from different socioeconomic backgrounds. In this context, worse-off individuals will benefit from
established economic activities at the local level, as they will have access to information flows that indicate where job and educational
opportunities are available (Barwick, 2017;Joseph et al., 2007). In this sense, social mix policies are a direct result of earlier studies on
neighbourhood effects which on the whole concluded that concentrated poverty has a clear effect on socioeconomic outcomes
(Brooks-Gunn et al., 1997;Duncan et al., 1993;Leventhal &Brooks-Gunn, 2000).
Social mix policies ultimately aspire to balance and diversify social environments through the creation of a “moderate heterogeneity”
1
The understanding of young adults' urban quality of life arises from the need to better understand the functionings and outcomes that need to be
achieved during the youth period. It has become increasingly important to understand the obstacles that young people are confronted with during
their transition from adolescence to adulthood, and to shed light on how urban contexts can facilitate/obstruct positive wellbeing outcomes. For a
more in-depth analysis of these issues, see Bucheli (2019,2020).
2
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
in urban spaces (Gans, 1961). Prior studies of social mixing have focused primarily on the case of the UnitedStates, the United Kingdom,
Canada, and Australia.
2
Social mixing policies have been conceived as urban planning theory and implemented through urban strategies
of both left and right wing political parties (Galster &Friedrichs, 2015). As an urban strategy, social mix policies have been implemented
by policy makers to advance towards urban social inclusion (Arthurson, 2010;Chaskin et al., 2012;Curley, 2010), to create more diverse
spaces by building social housing in areas where there are shortages (L
evy-Vroelant, 2014) or simply as a coherent urban strategy as
social mixing interventions operate through mediator factors that facilitate social integration. On the other hand, social mixing policies
have also been heavily criticised for their ability to enhance social diversity through legitimising the gentrification process (Davidson,
2010;Lees, 2008;Slater, 2013) in which entire disadvantaged communities are displaced or manipulated by more powerful external
agents who control community agendas (Bridge et al., 2014). Other authors have raised concerns regarding the lack of substantial
neighbourhood effects on labour markets in random assignment experiments such as Moving to Opportunity (Katz et al., 2000;Kling
et al., 2007;Ludwig et al., 2008;Sanbonmatsu et al., 2011).
In the case of Bogota, residential segregation patterns have created a city divided by socioeconomic strata, where the most
economically depressed areas are located to the south and periphery of the city, while more affluent areas are located in northern areas
of the city (Alfonso, 2012). This urban tendency can be described as a process of macrosegregation due to the fact that urban differ-
entiation tends to produce a polarised city where households auto-segregate according to their urban preferences and socioeconomic
options. Fig. 1 shows the level of urban polarisation that the city has reached since the initial application of the stratification system.
3
This pattern of urban macrosegregation can be seen in maps A and B where a process of self-segregation of elite groups is observed in the
northern part of the city. For some commentators, the macrosegregation process is strongly supported by the existence of urban de-
velopers' rentier spirit and relaxed state housing provision policies that have contributed to widening the spatial distance between
affluent and worse off populations in the city (Alfonso, 2012). Nevertheless, the urban structure of Bogota has begun to be described as a
fragmented city, in the sense that a process of social mixing is evolving (Aliaga &
Alvarez, 2010). The fragmentation of the urban space
in Bogota represents a structural metropolitan transformation that has modified the scale of how socio-territorial segregation develops.
A result of this process is the transition towards cities that no longer display the spatial dichotomy –‘rich city’and ‘poor city’–typical of
polarised cities. Instead, cities with ‘enclaves’of poverty and wealth are more evident. Spatial segregation has evolved towards a greater
mix in terms of urban landscape, where the combination of income polarisation, and its effect on the urban structure, has consolidated
this process of fragmentation within the city –the emergence of ‘small units of wealth and poverty that are spatially contiguous but
socially isolated from one another’(Thibert &Osorio, 2014, p. 1325). The polarisation of previous years, where segregation took place
at the macro level and was visually evident, is now taking place at the micro level (Figs. 1–, Map C), where local spaces are in
competition with each other.
Mixed communities are now evident in the urban space of Bogota, reflecting a change in the geographic scale of segregation, which is
becoming more micro and localised than previously. These changes in the patterning of segregation in Bogota might give the impression
that processes of cohesion and integration among heterogeneous groups can be achieved more spontaneously, as degrees of segregation
tend to reduce with time. Indeed, a less polarised city contributes towards a greater social mix in overall terms (Arthurson, 2013;Bond
et al., 2011;Galster, 2007;Joseph, 2008;Kearns et al., 2013). However, the current process of de-polarisation and the intensification of
enclaves of social and class differentiation among communities has intensified residential segregation but now in smaller, and conse-
quently, more intimate spaces where social integration is jeopardised.
Urban policies aimed at social mixing and reduction of inequalities through spatial integration have traditionally been implemented
by using housing policy mechanisms. As urban land management has been used as a mechanism to regulate the production of social
housing, local governments have been challenged to address the housing shortage in cities. Since the ratification of Act No. 388 of 1997
(the Regional and Urban Development Act), the process of decentralisation has introduced a series of urban planning mechanisms to
allow municipalities and districts to regulate and intervene in the land market, particularly in the generation of land suitable for social
housing.
A more recent attempt to tackle this situation can be traced back to the administration of Gustavo Petro (2012–2016). The
‘Development Plan of Bogota Humana’(SDP, 2012) adopted as its general framework the ‘reduction of all kinds of social, economic,
spatial and cultural segregation by increasing the capabilities of the population’(p. 19). In the case of spatial inequalities, the devel-
opment plan advocated for increased affordable housing for poor households - in particular for internally displaced people arriving in
Bogota due to the internal conflict - in urban renewal projects in the central business district (CBD) of the city. The profile of
anti-segregation policies was raised even further when in November 2014, the mayor's office announced the plan to build social housing
in affluent neighbourhoods in Bogota, in a clear strategy to progress towards the creation of mixed communities in the city (Poor
households in affluent neighbourhoods, 2014). The announcement generated a fierce debate between policymakers and affluent res-
idents, who understood the strategy more as a populist decision than an incentive to increase the social mix of the city. Following the
public debate, the initiative was abandoned, and public intervention was focused on identifying areas close to the CBD to be developed
for social housing.
2
The perspective of the global south on social mix and mixed communities has been more limited. However, there is a emergent literature from
countries such as China, Chile, Singapore, South Korea, and Turkey (Bayram et al., 2012;Chu et al., 2019;Kumar, 2020;Tian et al., 2007;Vergara,
2021;Yang et al., 2018).
3
The ‘stratification system’is a socioeconomic mechanism that ranks residences from one to six strata, aiming to focalise subsidies to calculate
utility bill tariffs (Uribe &Pardo, 2006).The stratification system works not just as a mechanism to classify the urban population but also as an income
based spatial division that contributes to the polarisation of the city between poor and rich inhabitants (SDP, 2021).
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
3
(caption on next page)
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
4
To date, there is no deliberate urban policy in Bogota that looks at the creation of neighbourhoods in which populations are mixed, as
is the case with the initiatives documented in neighbourhood effects studies of mixed community strategies in the United States and
Europe (Andersson et al., 2007;Arthurson, 2013;Cheshire, 2009;Galster, 2007;Lupton &Tunstall, 2008;Musterd et al., 2014;
Ostendorf et al., 2001). The case of Bogota tends to be more atypical, as mixed communities are created more under the influence of
market-driven processes than state intervention. The institutional framework that has been introduced has left a significant gap in how
to move towards mixed communities. Although the objective of the legal framework promised better integration in the city, the
implementation of the norm was being captured by the segregationist logic of the city where affordable housing is pushed out to the
south and periphery of the city.
In the context of urban policy, ‘mixed community’policies are a controversial strategy. There is limited evidence that these kinds of
urban settings have a substantial effect on individual outcomes, particularly in reducing poverty. In fact, there is a tendency to overstate
the positive effects of this type of urban setting, where much of the literature tends to rely on reviews or summaries of the evidence
rather than on primary data (Bond et al., 2011). The assumption that social mixing is an appropriate strategy to address the negative
effects which are present in deprived neighbourhoods (Glossop, 2008;Kearns &Mason, 2007;Lupton &Tunstall, 2008;Wilson, 1987)
requires further investigation, particularly when social mix is not a deliberate policy implemented by the state or local authorities.
Therefore, this paper focuses on testing the hypothesis of whether living in a context of social mix improves the quality of life of
worse-off residents. To do this, the paper attempts to overcome two limitations in how neighbourhood effects are commonly investi-
gated. Firstly, the analysis looks at the integration of objective and subjective approaches to human advantage by using primary data
collected in an urban area where a process of fragmentation is present. Although several studies have focused on understanding which
objective and subjective variables need to be considered (Najafpour et al., 2014;Powell &Sanguinetti, 2010), studies often analyse them
separately and do not consider their associations and effects on the development of aggregated indices of quality of life.
Secondly, this paper investigates the effect of heterogeneous neighbourhoods upon capabilities and preferences using the method of
propensity score matching (PSM) to overcome the problem associated with selection into neighbourhoods (Ginther et al., 2000;Har-
ding, 2003;Morgan &Harding, 2006). As most studies of neighbourhood effect rely on observational data, researchers are hampered by
the difficulty of determining why people live in different neighbourhoods. The presence of unobserved conditions produces a selection
bias that makes it difficult to ascertain whether differences between neighbourhoods are caused by the neighbourhood context or simply
because the people who inhabit them are different. If residential self-selection bias is present, it is difficult to differentiate between the
effect of neighbourhood conditions on individual outcomes from the choice to live close to these features (Lamb et al., 2020). What we
have seen so far is that estimates of neighbourhood effects on individual outcomes have tended to vary widely between studies, sug-
gesting the omission of relevant individual and household characteristics. This can be confirmed by neighbourhood effect studies
focused on these particular methodological concerns (Dietz, 2002;Weinberg et al., 2004;R.Sampson, 2008;Burdick-Will et al., 2010;
Hedman &Ham, 2012;Lamb et al., 2020). To correct for selection bias, the matching method identifies comparable young adults living
in a context of mixed communities, comparing them to a counterfactual (disadvantaged neighbourhood) to estimate to what extent
living in a better-off neighbourhood affects quality of life. It is important to note that in order to respond to the challenge of selection
bias, the discussion in the literature has been significant, and several authors have addressed these issues with multiple approaches, with
the use of econometric methods to analyse observational data (Cao and Jason, 2010;Cao et al., 2010;Kawachi &Subramanian, 2007)
and experimental designs (Kling et al., 2007;Briggs, 1997;Burdick-Will et al., 2010;Clampet-Lundquist &Massey, 2008;R.Sampson,
2008;Sharkey, 2013;Chetty et al., 2015).
3. Data
3.1. Study area
In order to understand how mixed neighbourhoods affect young adults' life changes in Bogota, it was necessary to identify urban
areas where a process of microsegregation was in place. This would allow for the measurement of different indicators of wellbeing to
compare differences with similar peers living in urban areas with a contrasting process of segregation (homogeneous neighbourhoods).
A critical aspect in the research design was the selection of urban areas to ensure adequate comparisons between young adults who are
affected by different patterns of urban segregation. To do that, the selection of urban districts was based on two main criteria to detect
mixed income and homogeneous neighbourhoods. The first criterion consisted of identifying urban places where residential segregation
occurs according to a dynamic of intense/non-intense segregation. To account for this, the index of residential segregation (SRS) was
used to measure levels of diversity
4
in Bogota's urban districts, to identify macro and micro patterns of residential segregation. The index
of socioeconomic residential segregation is a measure calculated by the Planning Secretariat of Bogota which shows the relationship
Fig. 1. Process of urban fragmentation in Bogota
Note: Fig. 1 shows different processes of urban segregation in Bogota. Map A shows the distribution of all six socioeconomic strata across the city.
Map B shows the location of more affluent populations (strata 5 and 6), Map C presents the extremes of the stratification system, that is, strata 1 and 2
for the most disadvantaged populations and strata 5 and 6 for the location of better off inhabitants. Map D shows the tendency of better off pop-
ulations to move to more septentrional locations as worse off populations began to consolidate their social and cultural activities in the south (Aliaga
&
Alvarez, 2010;Dureau et al., 2007). Data shown has been provided by the Capital District's Spatial Data Infrastructure (IDECA).
4
Diversity between households can also be measured in terms of other variables of interest such as expenditure, and years of school attendance.
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
5
between households' ability to pay and the city's housing market. It takes into account the variance of factors such as the composition of
the household, employment levels and subsidies in order to create an index of dispersion that calculates levels of homogeneity and
heterogeneity between households (SDP, 2019, p. 35) .
5
According to results from the SRS (Table 1), the Chapinero and Ciudad Bolivar
urban districts are the spatial units which display the highest and lowest levels of segregation respectively in Bogota. In the case of the
Chapinero urban district, results for 2017 show a SRS score of 2.41 (SRS >1), indicating the lowest level of residential segregation
compared to other urban districts. Conversely, patterns of urban homogeneity were identified in the urban locality of Ciudad Bolivar.
Results from the SRS showed that Ciudad Bolivar has the highest levels of segregation in Bogota, 0.30 (SRS <1) (SDP, 2019, p. 35).
A second approach to the selection areas of analysis was to identify neighbourhoods which display similar economic conditions
within each identified urban locality. To do this, it was considered appropriate to use the Colombian stratification system to track similar
economic conditions across households in each urban locality. Bogota uses a socioeconomic stratification system, which classifies
households according to the conditions of buildings.
6
Taking this into account, the selection of comparable young adults was based on
prioritising those who were living in the urban districts of Chapinero and Ciudad Bolivar, and, additionally, who were classified within
strata 1 and 2. This meant that it was possible to compare young adults' outcomes across diverse (microsegregated) and homogeneous
(macrosegregated) urban settings. Following this criteria, in the case of the urban district of Chapinero, it was relatively easy to identify
neighbourhoods which were classified within strata 1 and 2 and which were closer to neighbourhoods classified within strata 5 and 6.
Finally, the neighbourhood of ‘Juan XXIII’was prioritised with respect to other similar neighbourhoods (for instance, Granada, Olivos
and Bosque Calderon) as it is closer to better-off neighbours and not isolated by main roads, rivers, large parks or other physical barriers
which may result in a distinct separation between groups.
The selection of the control group (the homogeneous neighbourhood) was carried out in the urban locality of Ciudad Bolivar. A set of
spatial and socioeconomic characteristics were considered in the selection of the comparison group in order to identify the neigh-
bourhoods which were most similar to Juan XXIII in the urban locality of Chapinero (Table 2). Firstly, a total of 21 neighbourhoods
matched the socioeconomic criteria. Following a second assessment, and considering exclusively spatial variables (area and distance)
and accessibility (security aspects), three neighbourhoods were identified as potential control groups: Primavera I, Perdomo Alto and
Quinta del Sur.
An active dialogue was established with local organisations and the public offices of the local mayor of Ciudad Bolivar with the aim
of selecting the appropriate neighbourhood. After receiving authorisation from local organisations and taking into consideration several
security issues, the neighbourhood of Perdomo Alto was selected as the control group to conduct the survey. Table 3 summarises the
differences between the two selected case studies, the neighbourhood of Juan XXIII in Chapinero, and Perdomo Alto in Ciudad Bolivar
(Fig. 2).
Table 1
Socioeconomic residential segregation index, Bogota (2017, 2014).
Urban district 2017 2014
Ciudad Bolivar 0.30 0.33
Bosa 0.34 0.3
Usme 0.33 0.25
San Cristobal 0.41 0.34
Rafael Uribe Uribe 0.43 0.38
Tunjuelito 0.4 0.36
Kennedy 0.62 0.5
Engativa 0.59 0.61
Puente Aranda 0.51 0.55
Los Martires 0.57 0.54
La Candelaria 0.77 1.32
Santa Fe 0.89 0.65
Fontibon 0.99 0.92
Antonio Nari~
no 0.66 0.61
Suba 1.07 1.41
Barrios Unidos 0.92 0.96
Teusaquillo 1.39 1.63
Usaquen 1.83 1.36
Chapinero 2.41 2.34
Source: Planning Secretariat of Bogota, 2019.
5
The SRS will be equal to or greater than 1 when the variability in a spatial unit of the city is equal to or greater than the variability of the whole
city. In this case, a SRS score greater than 1 will show that households present a low association and heterogeneity in the evaluated variables.
Conversely, a SRS score of less than 1 will indicate that the spatial unit being studied displays a higher level of association between variables,
corresponding to homogeneity among households.
6
This system has been developed to charge households different rates with regards to the consumption of electricity, water, gas, and drainage
services, among other utilities. As such, it is also a mechanism for geographic focalisation of social expenditure, since depending of the location of the
building the contribution to utility charges will be different.
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
6
3.2. Questionnaire design and sampling procedure
The selection of normative standards of quality of life among young adults followed a process of identification of relevant capabilities
by using Sen's Capability Approach. This identification of relevant capabilities among young adults in Bogota was based on the idea that
any list of capabilities should be developed through a process of public deliberation and participation. In contrast to more legal and
moral normative approaches (such as Nussbaum's capability list), the identification of capabilities for young adults in the context of
spatial inequality requires a context-specific approach where place-based perspectives are taken into account. In line with this, the
methodology used to identify the relevant capabilities for young adults followed four main stages and was based on Robeyns' (2003,
2005)five criteria to identify basic capabilities.
7
Robeyns' criteria were used as a starting point for identifying relevant
8
capabilities
during focus group discussions (FGD) and interviews with young adults in both selected urban settings. Framework analysis was used to
aggregate identified capabilities into domains of quality of life, which were then incorporated into a questionnaire through a
multiple-question Likert scale.
9
The questionnaire was tested for internal consistency using Cronbach's (
α
) alpha, ensuring the same
directionality of all variables through reverse coding. Results showed a highly reliable
α
(53 items;
α
¼0.806) for the entire capability
section of the questionnaire
10
. In addition to capability-based questions, the questionnaire administered in Chapinero and Ciudad
Bolivar urban districts also included questions regarding hedonic and cognitive subjective well-being (Diener, 1984;Kahneman et al.,
1999), as well as a 10-item personality inventory (TIPI) (Gosling et al., 2003), measuring personality traits and socioeconomic status
variables. Before the administration of the survey, a census survey of all households was conducted in order to identify the location and
total number of young adults in each urban setting.
11
In total, 300 young adults were identified in both urban settings of which 231 were
surveyed.
12
The size of the selected urban settings as well as the population density of young adults suggested that the administration of
a census survey was preferable to collecting data using a sample survey.
13
Table 2
Spatial and socioeconomic criteria for control group selection.
Neighbourhood characteristics Chapinero Ciudad Bolivar
Juan XXIII Perdomo Alto
Area (ha) 4.275 5.018
Distance (the closest) (km)
Main integral transport system 0.71430 0.5261
Green area 0.07040 0.0646
Health centre 0.58220 0.8059
Public school 0.35070 0.2097
Bus stop 0.06910 0.1004
Cadastre (no.)
Lots 353 250
Blocks 30 16
Houses 585 591
Socioeconomic
Stratum 1 and 2 1 and 2
Residential density (houses/ha) 136.82 117.77
Price mt
2
(average price, USD) 205.950 213.112
Note: Data from the Capital District's Spatial Data Infrastructure (IDECA).
7
The five criteria are: (i) explicit formulation: have an explicit, discussed and defended list of relevant capabilities; (ii) methodological justification:
justify the methodology that has been used to generate the list of relevant capabilities; (iii) sensitive to context: the identified list should seek to be both
abstract and practical in order to satisfy different audiences; (iv) different levels of generality: identify relevant capabilities that are not only
comprehensive of ideal domains of quality of life but that are also feasibly achievable; and (v) exhaustion and non-reduction: the list of relevant
capabilities should include all dimensions that are important to well-being. Equally important under this criterion is that no dimensions identified
should be reducible to other elements. By using Robeyns' criteria, the selection of domains of quality life attempts to reduce possible bias.
8
See forthcoming article (Bucheli, 2022).
9
The questionnaire is not included for the sake of brevity but is available upon request.
10
The questionnaire was also piloted with a self-selecting sample of 100 students from a private university in Bogota, helping to adapt the structure
of the questionnaire, amend the wording of some questions and test the time of application.
11
In most cases, the census and questionnaire were conducted at the same time in order to reduce the cost of administration and to ensure the whole
young adult population in each urban setting was covered.
12
In total, 231 young adults were surveyed of which 120 young adults were located in Perdomo Alto (control group) and 111 in Juan XXIII
(treatment group). The remaining 69 young adults were either not located or did not give consent to be interviewed.
13
A non-monetary incentive was used to stimulate participation among young adults and those who participated in the survey were asked to sign a
consent form to ensure full awareness regarding potential risks and benefits from their participation in the study. The young adult population census
survey was carried out with support of the local neighbourhood association in Juan XXIII and the Community Association ‘Cuyeca’in Perdomo Alto,
respectively.
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
7
3.3. Outcome variables
This paper conceptualises quality of life as a multidimensional construct where objective and subjective measurements account for
young adults' well-being and agency. In the case of subjective measures, the analysis uses emotional (experience) and cognitive
(evaluation) components of well-being (Diener, 1984). For the emotional component of well-being, which attempts to make a hedonic
assessment of feeling, desires and emotions, the questionnaire included the question: ‘Overall, how happy did you feel yesterday?’
(HAPP); and for the cognitive component, which assesses people's judgements related to expectations and ideal life (Hoorn, 2008)
(SATI), the question included was: ‘Overall, how satisfied are you with your life in general these days?’
Two additional subjective measures were also included: a variable to measure the level of achievement of well-being (Morrison,
2010) based on quality of life criteria (QoL), and a variable to assess perceptions of available opportunities (OPP). These last two
variables enquire directly about how much the place of residence affects well-being achievement and the availability of opportunities,
respectively. In order to develop a comprehensive multidimensional measure for young adults' advantage, the analysis incorporated two
continuous variables. SCALE is a variable that provides a global assessment of young adults' urban life, based on a scale of 0–100, where
0 represents the worst possible quality of life and 100 represents the best possible quality of life.
A second continuous variable, CAPA, calculates individual scores of capabilities. CAPA is a latent variable based on domains of active
functionings identified by young adults.
14
The questionnaire collected data in all relevant domains and calculated a score of capabilities
for each individual using principal component analyses. Domains used to calculate scores of capabilities were those capabilities that
were assessed as more influential for them in the context of residential segregation. A summary of all ouctome variables used is pre-
sented in Table 4.
Fig. 2. Study areas of social mix (Chapinero) and social homogeneity (Ciudad Bolivar)
Note: Data using shapefiles from Capital District's Spatial Data Infrastructure (IDECA).
Table 3
Selected case studies: Chapinero and Ciudad Bolivar urban settings.
Description Case 1 Case 2
Residential segregation process Polarisation Fragmentation
Scale of analysis Macrosegregation Microsegregation
Income-based composition Homogeneous Mixed/heterogeneous
Type of sorting Similarity (clustering) Dissimilarity (outliers)
Urban setting Ciudad Bolívar Chapinero
Neighbourhood setting Perdomo Alto Juan XXIII
Criteria 1 (level of diversity) SRS ¼0.30 (2017) SRS ¼2.17 (2017)
Criteria 2 (neighbourhood setting) Strata 1 and 2 Strata 1 and 2
Source: elaborated by the author.
14
The identification of CAPA domains took place following an inductive approach of prioritisation of domains of quality of life for young adults in
Bogota.
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
8
3.4. Independent variables
Table 5 presents a summary of the descriptive statistics for outcome and covariance variables. Independent variables were divided
into three different models. The first model corresponds to socioeconomic variables. Here, socioeconomic variables included young
adults' age, gender, ethnic group, income and spending level (monthly), educational attainment and household tenure. A second model
included neighbourhood conditions. In this model, variables included number of people living at home, time living in the neigh-
bourhood and commuting time to a place of work. A third model included young adults' personality traits, which included five per-
sonality dimensions: extraversion, openness, conscientiousness, neuroticism and agreeableness (Gosling et al., 2003).
Table 4
Outcome variables and questionnaire items.
Variable Code Item Type Domain of evaluation
Happiness HAPP How happy or unhappy did you feel yesterday? Categorical (ordinal) Affective well-being
Satisfaction SATI All things considered, how satisfied are you with your life in general
these days?
Categorical (ordinal) Cognitive well-being
Urban quality of life QoL What would you say that overall your quality of life is? Categorical (ordinal) Achievement well-
being
Available
opportunities
OPP Overall, what do you think your opportunities for living here are? Categorical (ordinal) Opportunities
Evaluation
satisfaction
SCALE Write a number from 0 to 100 that best describes your quality of life in
your neighbourhood
Continuous Life satisfaction
Capabilities/
functionings
CAPA Calculated using PCM Continuous (Capability
index)
Capabilities
Source: HAPP and SATI (Dolan et al., 2011), QoL (Morrison, 2011)
Table 5
Descriptive statistics of outcome and covariance variables.
Variable Treatment Control Total
(n ¼110) (n ¼120)
Outcome variables
HAPP (5 ordered outcomes) 4.17 3.86
(0.855) (0.888) (-2.892)**
OPP (5 ordered outcomes) 3.92 3.91
(0.65) (0.44) (-0.260)
QoL (5 ordered outcomes) 3.94 3.38
(0.865) (0.801) (-5.122)***
SATI (7 ordered outcomes) 5.97 5.71
1.12 (1.33) (-1.283)
SCALE 79.36 (13.77)*** 65.29 (18.01)***
CAPA 31.75 (0.164)*** 58.79 (0.181)***
Independent variables
Age 24.2 (3.13)*** 21.86 (3.06)***
Male (%) 48.18 (0.50)* 63.33 (0.48)*
Ethnicity (white) (%) 10.0 (0.30) 14.16 (0.35)
Time living in urban setting 208.49 (125.96)** 163.18 (95.63)**
Educational attainment
Primary education (%) 4.54 (0.20)*** 20.83 (0.40)***
Secondary education (%) 42.39 (0.49) 51.66 (0.50)
Vocational education (%) 34.54 (0.47) 23.33 (0.42)
University education (%) 16.36 (0.37)*** 4.16 (0.20)***
Postgraduate education (%) 1.81 (0.13) 0 (0)
Total net income 3.64 (1.15)*** 2.49 (1.41)***
Tenure (%) 64.54 (0.48) 52.5 (0.50)
TIPI (dimensions of personality)
Extraversion 4.43 (1.10 4.28 (1.27)
Agreeableness 4.24 (1.11) 4.45 (1.20)
Conscientiousness 4.25 (1.12) 4.43 (1.14)
Emotional stability 4.5 (1.02) 4.43 (1.34)
Open to new experiences 4.32 (1.13) 4.09 (1.40)
Violent assault in the neighbourhood 0.02 (0.16)** 0.14 (0.35)**
Commuting time to work (min.) 36.72 (25.36)*** 60.47 (39.75)***
Neighbourhood inclusion 0.93 (0.24) 0.85 (0.35)
House adaptability 0.81 (0.38)*** 0.55 (0.49)***
Note: Standard deviations in parentheses. t-test for continuous variables, and Fisher's exact test and Mann-Whitney UTest for categorical variables; *,
**, and *** indicate significance levels of 5%, 1% and 0.1%, respectively.
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
9
4. Method and Empirical Model
4.1. The counterfactual framework for estimation of treatment effects
Overcoming the problem of selection bias is a key consideration when assessing the effects of residential segregation on capability
scores and subjective measurements of well-being. In the context of neighbourhood effects literature, the selection problem is
considered to be one of the main difficulties in identifying causal effects on deprived neighbourhoods, as individuals do not randomly
choose where they live; with personal and family conditions playing an important role in this decision (Clampet-Lundquist &Massey,
2008;Jencks &Mayer, 1990).
The result is that it is very difficult to know whether being located in a deprived neighbourhood negatively affects people's life
trajectories, or whether the correlation between spatial poverty and lower individual outcomes is a pattern of residential selection,
where disadvantaged individuals self-select into or out of neighbourhoods (Clampet-Lundquist &Massey, 2008;Hedman &Ham, 2012).
From the perspective of evaluation standards and policy evidence, the simple correlation of mechanisms of residential mobility
(observed and unobserved individual characteristics) and deprived conditions can be wrongly assumed to constitute true neighbour-
hood effects (Hedman &Ham, 2012). Therefore, the lack of random assignment in observational studies forces researchers to control for
self-selection mechanisms and individual characteristics in order to estimate valid treatment effects.
Additionally, to calculate the effects that microsegregation has on young adults' capabilities and preferences it is necessary to know
the difference between the outcomes of young adults who are living in the context of microsegregation and thosewho are not. As it is not
possible to look at the difference in outcomes of the same young adults in both urban settings at the same time –a fundamental problem
of causal inference –an alternative approach is to find a separate group of young adults (control group) who have similar characteristics
to those who are living in the context of microsegregation (McDool, 2017). The existence of a control group allows for the possibility of
assessing what would have occurred if those young adults living in the context of microsegregation did not live there (counter-factual).
4.2. Model specification and matching variables
Matching techniques appeal to researchers because of their capacity to produce estimates with potential for causal interpretation
(Angrist &Pischke, 2009). In this section, the central proposal is to investigate whether being located in a heterogeneous urban setting
has an effect on how capabilities are perceived by young adults. By using matching as a strategy to determine causation, the research
aims to determine the average effect of treatment on the treated (ATT).
ATT E[y
i
1
y
i
0
|X,D
i
¼1] ¼E[y
i
1
|X,D
i
¼1] –E[y
i
0
|X,D¼1] (1.1)
Following Heckman et al. (1997) and Wendimu, Henningsen and Gibbon (2016),ATT is the average treatment effect on capability
scores of being located in a heterogeneous neighbourhood, y
i1
denotes the capability outcome for a young adult located in a hetero-
geneous neighbourhood and y
i
0
the outcome for the same young adult if they were not located in a heterogeneous neighbourhood; Xis a
vector of observed individual characteristics which may affect the probability of being located in a heterogeneous neighbourhood; and D
indicates participation in the programme or in this case, whether a young adult is located in the treatment area (D¼1) or in the control
group area (D¼0).
As the term E[y
i
0
|X,P¼1] cannot be inferred from observational data, it is necessary to substitute it (the expected capability value of
young adults of not being located in a heterogeneous neighbourhood) for the expected capability value of young adults who are not
located in heterogeneous neighbourhoods: E[y
i
0
|X,P¼0]. This means that unobserved outcomes from the treatment population (E[y
i
0
|X,
P¼0]) can be obtained from observed distribution for the matched nonparticipant group that is subject to the counterfactual condition
(Smith &Todd, 2005). In this analysis, young adults from heterogeneous (Juan XXIII) and non-heterogeneous (Perdomo Alto) neigh-
bourhoods are matched using observational characteristics that influence capability scores (outcome variable).
15
The average difference
in capability scores from both groups, after controlling for aspects that might yield differences between groups, can be inferred as the
effect of location (microsegregation) on the production of capabilities.
Therefore, the causal effect of being located in mixed neighbourhoods on capability scores for young adults can be described as:
ATT E[y
i
1
y
i
0
|X,P¼1] ¼E[y
i
1
|X,P¼1] E[y
i
0
|X,P¼0] (1.2)
Rosenbaum and Rubin (1983) suggest that in order to identify ATT, it is necessary to assume a strong ignorability (uncon-
foundedness), so independence between the treatment assignment and the outcome variable can hold. In this study, the matching
strategy consisted in adjusting for differences in a set of pre-treatment observable characteristics X, so that assignment to any of the two
conditions is independent of the potential outcomes once covariates, proposed theoretically, are kept constant (Guo &Fraser, 2014).
The variables in the survey that affect both the decision to participate and the outcome variable were considered simultaneously in
the matching process (Caliendo &Kopeinig, 2008). Bernal and Pe~
na (2011) suggest that researchers can be guided by econometric
models or studies that have been used previously to understand a particular phenomenon. To the best of the author's knowledge, there is
no previous study that researches the effects of mixed communities in Bogota. More importantly, the case study under investigation here
15
The estimated treatment effects of the model are evaluated in terms of their sensitivity “with respect to deviations from this identifying
assumption”(Becker &Caliendo, 2007).
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
10
does not constitute an example of the deliberate policy of moving well-off households to areas of heterogeneous conditions,
16
as is
usually the case with studies of mixed communities.
Residential self-selection for the treatment and control groups can be controlled by socioeconomic conditions such as housing, and
preferences are likely to be similar between them. Thus, one potential way to deal with this is to assume exclusively socio-demographic
and economic variables to control for in the matching process, as they are likely to simultaneously affect the participation of a mixed
community project and the outcome variable under investigation. It is important to note that these matching variables were at play in
conjunction with the spatial and socioeconomic criteria for selecting the control group to ensure that the matched controlled group
could provide an unbiased treatment effect estimate. Therefore, once differences between the treatment and the control group are
controlled, both groups are considered homogeneous, so the fact of being located in a mixed neighbourhood can be regarded as a
random event.
5. Results
5.1. Propensity score estimation
To estimate propensity scores, a binary logistic regression equation was calculated as the treatment can only have two possible values
(i.e. D ¼1, D ¼0). A probit function was selected to indicate the probability of being located in a heterogeneous neighbourhood,
regardless of whether a given young adult was or not. Earlier t-test results showed some variables do not approach significance level,
suggesting that differences between the treatment and the control group might not necessarily be accounted for by matching techniques.
It is important to highlight that differences between groups are produced by the multiplicative effects between independent variables
towards the outcome variable of interest, and not just by the bivariate relationship between them. Therefore, it is still relevant to use
matching techniques to account for possible differences in outcome variables between treatment and control once a set of possible
covariances is at play.
Results of the probability of participation in heterogeneous neighbourhoods reveal that young adults with a higher level of education
tend to be more likely to live in heterogeneous neighbourhoods. Older young adults with higher incomes and shorter commuting times
are also more likely to live in a heterogeneous neighbourhood. Young adults who have lived in the neighbourhoods for longer, with
respondents defining their ethnicity as white and renters also more likely to receive the treatment. Personality trait results were not
significant in the probit regression; however, the sign of the coefficients indicates that young adults who are more self-confident,
assertive and who seek a high level of external stimulation (extraversion), as well as those who are more intellectually curious and
more unconventional (openness), are more likely to receive the treatment. Conversely, those young adults who display behaviours
towards control and self-regulation (conscientiousness) tend to be less likely to participate. The estimated results from the probit
regression are presented in Table 6.
The analysis then progressed to calculate a propensity score which estimated the probability that each young adult would be exposed
to a heterogeneous urban setting, given the control variable determined earlier. For each matching strategy, all confounding variables
displayed better balancing after matching since all covariants obtained a much lower level of bias. Standardised differences in means for
matched variables were less than 0.25 as is recommended by Stuart and Rubin (2008).
The analysis was restricted to the region of common support, which indicated that for each treated observation there is a comparison
observation with a close propensity score, so an optimal overlap can be ensured.
17
Finally, different matching strategies were performed
to estimate the average effect of treatment (ATT).
18
Table 7 presents unmatched and matched values for outcome variables using
nearest-neighbour (N1) matching. ATT is positive and statistically significant for SCALE and OPP, and negative and statistically sig-
nificant for CAPA. This result suggests that young adults living in heterogeneous neighbourhoods on average feel more satisfied with life
than those who inhabit homogeneous settings. Young adults in mixed communities tend to have a higher level of affective well-being
(HAPP) than young adults located in homogeneous settings. However, the effects for QoL and HAPP do not reach statistical significance.
Other matching techniques were used to test for robustness of findings. Caliper and radius matching were used to improve the risk of
potential bad matches if neighbours are far away between treated and controls.
19
Table 8 reports the results of all different matching strategies employed. Estimated ATTs in each matching strategy show the positive
effect of heterogeneous neighbourhoods on young adults' subjective well-being, measured by the level of satisfaction and the level of
16
During the administration of Gustavo Petro (2012–2016), the construction of affordable housing in mixed neighbourhoods was part of the urban
strategy to tackle levels of residential segregation in the city. The political agenda of ‘Bogota Humana’consisted of reducing spatial inequalities
through building integration spaces of different parts of the city. After Petro's administration, public policy in this regard was abandoned, or at least
public debate on the subject has been reduced (Hernandez &Becerra, 2017).
17
The ‘trimming level’was calculated following Smith and Todd's (2005) suggestion of defining the region of common support by including only
those values where there is a positive density.
18
The standard errors of ATT estimators were obtained by bootstrapping based on 1000 replications.
19
A caliper is set to impose a maximum propensity score distance, so bad matches are not taken into consideration in the analysis. A possible
drawback of this matching technique is that treated individuals are excluded from the analysis if no matches can be found within the caliper (Smith &
Todd, 2005). Kernel matching was also employed in the analysis. In this type of matching, weighted averages of all untreated individuals are used
within the bandwidth (0.01) ensuring a lower level of variance, as much information is considered in the analysis (Caliendo &Kopeinig, 2008). The
analysis also used stratification matching to allow the common support to be partitioned in a set of intervals in order to calculate the effects within
each interval by looking at the mean difference between treated and control (Caliendo &Kopeinig, 2008, p. 42).
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
11
opportunities perceived in the urban environment. Equally, there is a negative effect of living in heterogeneous settings on the level of
capabilities. Results show that CAPA has a much lower sensitive response among the different matching methods in comparison to OPP
and SCALE. Nevertheless, the effects of heterogeneous settings on CAPA, OPP and SCALE remain significant and robust in each matching
strategy.
Affective (HAPP) and cognitive (SATI) measurements of well-being have a positive relationship with being located in a heteroge-
neous setting, however the effect is not statistically significant. The significant correlational claim arising from HAPP and treatment does
not survive the test of causal effects. Similarly, there is no evidence that living in heterogeneous neighbourhoods has an effect on
perceptions of quality of life, however it raises the question of whether the relationship between quality of life and heterogeneous
settings can produce detrimental effects.
If there are unobservable factors that were correlated to the treatment variable, it is recommended to use instrumental variables to
replace them. As variables used in the analysis come from primary data, there are no additional variables that can serve as instrumental
variables or that have the property to be uncorrelated to unobserved factors and that can be provide information on the process of
Table 7
Average treatment effect among the treated based on nearest-neighbour matching (1-NN).
Variable Sample Treated Control Difference Std. Err. T-stat
HAPP Unmatched 4.150 3.840 0.309*** 0.123 2.51
ATT 4.154 3.976 0.178 0.202 0.88
OPP Unmatched 3.935 3.353 0.581*** 0.117 4.95
ATT 3.940 3.428 0.511*** 0.193 2.64
SATI Unmatched 5.967 5.672 0.294 0.178 1.65
ATT 6.000 5.841 0.158 0.264 0.60
QoL Unmatched 3.892 3.902 0.010 0.077 0.13
ATT 3.892 4.035 0.142 1.119 1.20
SCALE Unmatched 79.032 64.65 14.377*** 2.278 6.31
ATT 79.059 62.76 16.297*** 4.194 3.89
CAPA (%) Unmatched 0. 319 0.590 0.271*** 0.023 11.33
ATT 0.311 0.490 0.179*** 0.035 5.06
Note: *, **, and *** indicate significance levels of 5%, 1% and 0.1%, respectively.
Table 8
Average treatment effect among the treated on subjective and objective measures of well-being. Results from different matching methods.
Estimation method HAPP OPP SATI QoL SCALE (%) CAPA (%)
t-test 0.30** 0.56*** 0.26 0.01 14.07*** 0.27***
Regression with dummy 0.08 0.54*** 0.32 0.06 14.73*** 0.23***
Nearest-neighbour (NN ¼1) 0.18 0.51*** 0.16 0.14 16.29*** 0.18***
Caliper 0.06 0.50** 0.33 0.10 13.29** 0.20***
Kernel 0.08 0.53** 0.35** 0.10 14.40*** 0.19***
Stratification 0.16 0.50*** 0.41** 0.04 14.30*** 0.19***
Note: Regression models for categorical variables used ordered probit regressions to consider the order of integers rather than the distance. *, **, and
*** indicate significance levels of 5%, 1% and 0.1%, respectively.
Table 6
Probit regression for propensity score matching.
Dependent variable Heterogeneous neighbourhood
Independent variable Coef. Std. Err. Z P >z 95% conf. Interval
Age 0.072 0.034 2.11 0.035* 0.005, 0.140
Male 0.293 0.213 1.37 0.169 0.711, 0.124
Ethnicity (white) 0.003 0.003 1.01 0.314 0.003, 0.011
Time living in the urban setting 0.001 0.001 1.79 0.074 0.4269, 0.142
Education attainment 0.426 0.142 3.00 0.003** 0.147, 0.706
Income level 0.285 0.086 3.30 0.001*** 0.115, 0.454
Tenure (rent) 0.504 0.233 2.16 0.031* 0.046, 0.962
Extraversion 0.088 0.090 0.98 0.327 0.088, 0.265
Agreeableness 0.031 0.101 0.31 0.755 0.229, 0.166
Conscientiousness 0.206 0.105 1.96 0.050* 0.412, 0.000
Emotional stability 0.064 0.107 0.60 0.545 0.275, 0.145
Open to new experiences 0.099 0.100 0.99 0.545 .0975, 0.295
Cons 3.709 1.322 2.80 0.005 6.302, 1.116
Note: *, **, and *** indicate significance levels of 5%, 1% and 0.1%, respectively.
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
12
microsegregation. An alternative option to test for effects of unobserved variables is to carry out a sensitivity analysis to account for
some of the covariance that the analysis was not able to identify.
20
6. Discussion
The aim of this paper was twofold. First, the analysis attempts to extend the literature on neighbourhood effects by looking at how
divergent dynamics of residential segregation in Bogota produce effects on young adults' well-being and agency. The second aim was to
use matching techniques with primary data to compare results of subjective and objective measures of well-being. This method allowed
us to investigate to what extent mixed neighbourhoods affect quality of life, by controlling for factors affecting both residential
segregation and objective and subjective domains of quality of life.
Overall, the findings suggest conflicting results between objective and subjective measurements of well-being. On the side of sub-
jective measures of well-being, results show that there is no evidence that living within diverse communities has an effect on emotional
well-being. Results show a positive effect on HAPP but effects do not reach statistical significance, meaning that it is not possible to
sustain the thesis that being located or living in diverse communities explains young adults' happiness. This result suggests that more
research is needed to collect better variables to account for ‘geographies of subjective well-being’, particularly those that can encap-
sulate the relationship between the spatial organisation of urban amenities and individual satisfaction and that can be assessed by using
construct and criterion validity. With regard to cognitive measures of well-being, results are significant. Young adults in heterogeneous
settings report a positive level of satisfaction on measurements of SCALE and SATI. For the case of SCALE, statistically significant
differences were found. Young adults in Juan XXIII (microsegregated) report that they are 16% more satisfied with their lives than their
peers in Perdomo Alto (macrosegregated). Conversely, normative measures of well-being based on capabilities and functionings report
an opposite result. Here, a statistically significant decrease of 18–20 perceptual points of CAPA is reported for young adults living in
heterogeneous neighbourhoods compared to their peers in homogeneous settings.
The greater adaptation of young adults in Juan XXIII may be a sign of the lack of autonomy and the difficulty in defining substantive
freedoms compared with their peers in Perdomo Alto. This finding is in line with previous studies that point to the problem of adaptation
when people's aspirations are modified in adverse contexts (Clark, 2009;Qizilbash, 2006). Adaptability formation is linked to ways of
coping with the frustration of not having the material resources to enhance their quality of life at the same rate as their well-off
neighbours. Therefore, the constant process of catching up with peers in social and economic domains, and the fact that, as a minor-
ity, they are less targeted by public policy due to dispersal, young adults from heterogeneous neighbourhoods tend to accept the given
order and adapt their perceptions and social arrangements more easily than young adults in homogeneous settings. The fact that people
adapt preferences when they face uneven situations makes them “implicit accomplices”as they legitimise unequal orders (Watts, 2009,
p. 430). This serves to explain why young adults in Juan XXIII show a rather lower level of political engagement and social participation
than their peers in Perdomo Alto. Based on the qualitative results of previous studies (Bucheli, 2019), worse-off young adults from Juan
XXIII described themselves as more apathetic and indifferent towards change, showing their preference for assuming an adaptive po-
sition that is represented by their behaviour of following the status quo prevailing in the urban setting in which they are located. In
comparison, young adults from Perdomo Alto tend to adapt their preferences at a lower rate. For them, contestation and struggle are
mechanisms to preserve autonomy, so they are able to ask for and demand the fulfilment of basic rights, which represents less
acceptance of the context of inequality they live in and less adjustment of young adults' aspirations.
The acceptance by young adults in Juan XXIII of the need to lower their expectations with regards to inaccessible options
(employment, acquisitions, status) is also represented by the high levels of perceived availability of opportunities. Treated individuals
reported a positive effect on the variable OPP which measures to what extent opportunities are more available to them due to their
location in heterogeneous neighbourhoods. The significant level of this variable might indicate that restrictions to identifying certain
choices can be at play, as opportunities are in most cases provisional or do not provide long-term solutions.
21
Living close to better-off
neighbours does not represent a real case for upgrading opportunities but a deliberative manipulation of wants where the accessibility of
alternatives and options is mainly conditioned by the ability to pay and the existence of influential social relations.
7. Conclusions
This paper shows the detrimental impact of segregation on capabilities of young adults in two urban districts in Bogot
a at a micro
scale. This is in line with previous research, which arrives at similar conclusions using different well-being information spaces (Luttmer,
2004;Morrison, 2010). The paper also complements findings that argue that heterogeneous neighbourhoods have a modest effect on the
amount of social mixing and social integration that they promise to achieve (Chaskin et al., 2012;Tach, 2014). There is conflicting
evidence regarding the effects of social mixing on reducing social inequality, as several studies have found that social mixing does not
translate into meaningful interactions when people are living close to each other (Curley, 2010;Goetz &Chapple, 2010). Other studies
show that there is no social contact at all between different communities despite their proximity (Arthurson, 2012;Kleit, 2001,2013).
20
Sensitivity analysis assesses the extent to which matching estimates are robust to the presence of unobserved variables that simultaneously have
an effect on treatment assignment and outcome variables (Rosenbaum, 2002).
21
For instance, young adults in Juan XXIII identify that living close to better-off families has increased access to job options such as domestic work,
dog walking, babysitting, home maintenance and so on, but they are not durable solutions to unemployment as these types of occupations tend to
exacerbate informality and job insecurity.
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
13
From an evidence-based policy perspective, the findings presented here provide an important contribution to the question of whether
place has an effect on how urban life is shaped. More importantly, the finding of adverse effects of heterogeneous neighbourhoods
provides a clear rationale to assess mixed neighbourhood initiatives critically from the perspective of capabilities. If an integrative
approach of objective and subjective measurements of well-being is taken into account, heterogenous neighbourhoods can be
considered as a potential cause of detrimental impacts on young adults' opportunities and autonomy.
This finding adds to the current literature on neighbourhood effects in two directions. Firstly, the use of capabilities as an infor-
mational space to assess well-being allows us to deviate from paternalistic approaches in the analysis of urban public policies that take
for granted predefined domains to quantify urban life. The fact that the analysis included an outcome variable (CAPA), the scores of
which reflect achievement outcomes in the domains that young adults have reasons to value, implies an expansion of the evaluative
space within which quality of life is assessed. Secondly, in methodological terms, the analysis presented here contributes to strength-
ening the quantitative evidence on the effects of social mixing policies. However, as the analysis and method employed in this study do
not allow us to completely claim causal effects in the results, the detrimental effects of heterogeneous neighbourhoods on capabilities
compared to homogeneous settings raise doubts concerning arguments that residential segregation exclusively reflects social and
economic inequality rather than causing it (Cheshire, 2012;Ostendorf et al., 2001). Residential social mixing in Bogota are far from
being harmless in producing additional disadvantages for worse-off young adults. However, this conclusion does not preclude the idea
that urban poverty and social exclusion are strongly influenced by other structural and agency factors that also produce inequality. The
importance of better understanding residential mobility in Bogota is relevant, not just because it can explain contemporary patterns of
urban segregation in the city, but because it can also advance towards understanding whether neighbourhood effects are constant
among deprived neighbours or whether the transmission of low capability scores is neighbourhood-specific.
Research interests
Young adults, youth, operationzalisation of the capability approach and measurement of well-being, subjective well-being.
Funding details
Funding for this study was provided by the Colciencias Cambridge Scholarship (Cambridge Commonwealth, European and Inter-
national Trust).
Conflict of interest
The author declares that there is no conflict of interest with any financial organisations regarding the materials reported in this
manuscript.
References
Alfonso, O. (2012). Bogot
a segmentada: Reconstrucci
on hist
orico-social de la estructuraci
on residencial de una metr
opoli latinoamericana. U. Externado de Colombia.
Aliaga, L., &
Alvarez, M. J. (2010). Residential segregation in Bogot
a across time and scales. Lincoln Institute of Land Policy.
Andersson, R., Musterd, S., Galster, G. C., & Kauppinen, T. M. (2007). What mix matters? Exploring the relationships between individuals' incomes and different
measures of their neighbourhood context. Housing Studies, 22(5), 637–660. https://doi.org/10.1080/02673030701474602
Angrist, J. D., & Pischke, J.-S. (2009). Mostly harmless econometrics: An empiricist's companion (1 edition). Princeton University Press.
Arthurson, K. (2010). Questioning the rhetoric of social mix as a tool for planning social inclusion. Urban Policy and Research, 28(2), 225–231. https://doi.org/10.1080/
08111141003693117
Arthurson, K. (2012). Social mix and the city: Challenging the mixed communities consensus in housing and urban planning policies. CSIRO Publishing.
Arthurson, K. (2013). Mixed tenure communities and the effects on neighbourhood reputation and stigma: Residents' experiences from within. Cities, 35, 432–438.
https://doi.org/10.1016/j.cities.2013.03.007
Atkinson, R., & Kintrea, K. (2001). Disentangling area effects: Evidence from deprived and non-deprived neighbourhoods. Urban Studies, 38(12), 2277–2298. https://
doi.org/10.1080/00420980120087162
Barwick, C. (2017). Social mix revisited: Within- and across-neighborhood ties between ethnic minorities of differing socioeconomic backgrounds. Urban Geography,
1–19. https://doi.org/10.1080/02723638.2017.1405690, 0(0).
Bayram, N., Bilgel, F., & Bilgel, N. G. (2012). Social exclusion and quality of life: An empirical study from Turkey. Social Indicators Research, 105(1), 109–120. https://
doi.org/10.1007/s11205-010-9767-4
Becker, S. O., & Caliendo, M. (2007). Sensitivity analysis for average treatment effects. STATA Journal, 7(1), 71–83.
Bernal, R., & Pe~
na, X. (2011). Guía pr
actica para la evaluaci
on de impacto. Universidad de los Andes.
Ble
ci
c, I., Cecchini, A.‘B.’, & Talu, V. (2013). The capability approach in urban quality of life and urban policies: Towards a conceptual framework. In S. Serreli (Ed.),
City project and public space (pp. 269–288). Springer Netherlands. http://link.springer.com/chapter/10.1007/978-94-007-6037-0_17.
Bond, L., Sautkina, E., & Kearns, A. (2011). Mixed messages about mixed tenure: Do reviews tell the real story? Housing Studies, 26(1), 69–94. https://doi.org/10.1080/
02673037.2010.512752
Bridge, G., Butler, T., & Le Gal
es, P. (2014). Power relations and social mix in metropolitan neighbourhoods in north America and Europe: Moving beyond
gentrification? International Journal of Urban and Regional Research, 38(4), 1133–1141. https://doi.org/10.1111/1468-2427.12125
Briggs, X. de S. (1997). Moving up versus moving out: Neighborhood effects in housing mobility programs. Hous. Policy Debate, 8(1), 195–234. https://doi.org/
10.1080/10511482.1997.9521252
Brooks-Gunn, J., Greg, D., & Aber, J. L. (1997). Neighborhood poverty. In Context and consequences for children, ume 1. Russell Sage Foundation.
Bucheli, J. F. (2019). Before entering adulthood: Developing an index of capabilities for young adults in Bogota. Appl. Res. Quality. Life.https://doi.org/10.1007/
s11482-019-09781-6
Bucheli, J. F. (2020). Marginal youth: Mapping spatial capability exclusion in Bogota. Geojournal, 85(3), 593–615. https://doi.org/10.1007/s10708-019-09970-8
Bucheli, J. F. (2022). Quality of life in segregated places: What does it mean for young adults in Bogota ? University of Cambridge (Unpublish manuscript).
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
14
Burdick-Will, J., Ludwig, J., Raudenbush, S., Sampson, R., Sanbonmatsu, L., & Sharkey, P. (2010). Converging evidence for neighborhood effects on children's test scores: An
experimental, quasi-experimental, and observational comparison*.Undefined https://www.semanticscholar.org/paper/Converging-Evidence-for-Neighborhood-
Effects-on-An-Burdick-Will-Ludwig/fbdef14f61ab6192d943f33b2a2aba0a426fec34.
Caliendo, M., & Kopeinig, S. (2008). Some practical guidance for the implementation of propensity score matching. Journal of Economic Surveys, 22(1), 31–72. https://
doi.org/10.1111/j.1467-6419.2007.00527.x
Cao, X., & Jason). (2010). Exploring causal effects of neighborhood type on walking behavior using stratification on the propensity score. Environ. Plann.: Economy and
Space, 42(2), 487–504. https://doi.org/10.1068/a4269
Cao, X., Jason), Xu, Z., & Fan, Y. (2010). Exploring the connections among residential location, self-selection, and driving: Propensity score matching with multiple
treatments. Transportation Research Part A: Policy and Practice, 44(10), 797–805. https://doi.org/10.1016/j.tra.2010.07.010
Chaskin, R., Khare, A., & Joseph, M. (2012). Participation, deliberation, and decision making: The dynamics of inclusion and exclusion in mixed-income developments.
Urban Affairs Review.https://doi.org/10.1177/1078087412450151, 1078087412450151.
Cheshire, P. (2009). Policies for mixed communities faith-based displacement activity? International Regional Science Review, 32(3), 343–375. https://doi.org/10.1177/
0160017609336080
Cheshire, P. (2012). Are mixed communities policies evidence based? A review of the research on neighbourhood effects. In Neighbourhood effects research: New
perspectives (pp. 269–291). Dordrecht: Springer. https://doi.org/10.1007/978-94-007-2309-2_4.
Chetty, R., Hendren, N., & Katz, L. F. (2015). The Effects of Exposure to better Neighborhoods on children: New Evidence from the Moving to opportunity experiment (working
paper No. 21156). National Bureau of Economic Research. http://www.nber.org/papers/w21156.
Chu, C., Nomura, R., & Mori, S. (2019). Actual conditions of mixed public–private planning for housing complexes in beijing. Sustainability, 11(8), 2409. https://
doi.org/10.3390/su11082409
Clampet-Lundquist, S., & Massey, D. S. (2008). Neighborhood effects on economic self-sufficiency: A reconsideration of the moving to opportunity experiment.
American Journal of Sociology, 114(1), 107–143. https://doi.org/10.1086/588740
Clark, D. A. (2009). Adaptation, poverty and well-being: Some issues and observations with special reference to the capability approach and development studies.
J. Human. Dev. Capabilities, 10(1), 21–42. https://doi.org/10.1080/14649880802675051
Curley, A. M. (2010). Relocating the poor: Social capital and neighborhood resources. Journal of Urban Affairs, 32(1), 79–103. https://doi.org/10.1111/j.1467-
9906.2009.00475.x
Davidson, M. (2010). Love thy neighbour? Social mixing in London's gentrification frontiers. Environ. Plann.: Economy and Space, 42(3), 524–544. https://doi.org/
10.1068/a41379
Diener, E. (1984). Subjective well-being (SSRN scholarly paper ID 2162125). Soc. Sci. Res. Netw., 95(3), 542–575. http://papers.ssrn.com/abstract¼2162125.
Dietz, R. (2002). The estimation of neighborhood effects in the social sciences: An interdisciplinary approach. Social Science Research, 31(4), 539–575. https://doi.org/
10.1016/S0049-089X(02)00005-4
Dolan, P., Layard, R., & Metcalfe, R. (2011). Measuring subjective Wellbeing for public policy: Recommenda tions on measures (CEP special paper No. 23). Centre for Economic
Performance, LSE https://ideas.repec.org/p/cep/cepsps/23.html.
Duncan, C., Jones, K., & Moon, G. (1993). Do places matter? A multi-level analysis of regional variations in health-related behaviour in britain. Social Science &
Medicine, 37(6), 725–733. https://doi.org/10.1016/0277-9536(93)90366-C
Dureau, F., Barbary, O., & Lulle, T. (2007). Dinamicas metropolitanas de poblamiento y segregacion [Metropolitan dynamics of demography and segregation]. In
Ciudades y sociedades en mutaci
on: Lecturas cruzadas sobre Colombia [Cities and societies in mutation: Cross readings about Colombia]. U. Externado de Colombia.
Ellen, I. G., & Turner, M. A. (1997). Does neighborhood matter? Assessing recent evidence. Hous. Policy Debate, 8(4), 833–866. https://doi.org/10.1080/
10511482.1997.9521280
Galster, G. C. (2007). Neighbourhood social mix as a goal of housing policy: A theoretical analysis. Int. J. Housing Policy, 7(1), 19–43. https://doi.org/10.1080/
14616710601132526
Galster, G. C. (2012). The mechanism(s) of neighbourhood effects: Theory, evidence, and policy implications. In M. van Ham, D. Manley, N. Bailey, L. Simpson, &
D. Maclennan (Eds.), Neighbourhood effects research: New perspectives (pp. 23–56). Springer Netherlands. https://doi.org/10.1007/978-94-007-2309-2_2.
Galster, G. C., & Friedrichs, J. (2015). The dialectic of neighborhood social mix: Editors' introduction to the special issue. Housing Studies, 30(2), 175–191. https://
doi.org/10.1080/02673037.2015.1035926
Gans, H. J. (1961). The balanced community: Homogeneity or heterogeneity in residential areas? Journal of the American Institute of Planners, 27(3), 176–184. https://
doi.org/10.1080/01944366108978452
Ginther, D., Haveman, R., & Wolfe, B. (2000). Neighborhood attributes as determinants of children's outcomes: How robust are the relationships? Journal of Human
Resources, 35(4), 603–642. https://doi.org/10.2307/146365
Glossop, C. (2008). Housing and economic development: Moving forward together. Centre for Research and Market Intelligence.
Goetz, E. G., & Chapple, K. (2010). You gotta move: Advancing the debate on the record of dispersal. Hous. Policy Debate, 20(2), 209–236. https://doi.org/10.1080/
10511481003779876
Gosling, S. D., Rentfrow, P. J., & Swann, W. B. (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37(6), 504–528.
https://doi.org/10.1016/S0092-6566(03)00046-1
Guo, S., & Fraser, M. W. (2014). Propensity score analysis: Statistical methods and applications (2 edition). SAGE Publications, Inc.
Harding, D. J. (2003). Counterfactual models of neighborhood effects: The effect of neighborhood poverty on dropping out and teenage pregnancy. American Journal of
Sociology, 109(3), 676–719. https://doi.org/10.1086/379217
Heckman, J. J., Ichimura, H., & Todd, P. E. (1997). Matching as an econometric evaluation estimator: Evidence from evaluating a job training programme. The Review of
Economic Studies, 64(4), 605–654. https://doi.org/10.2307/2971733
Hedman, L., & Ham, M. van (2012). Understanding neighbourhood effects: Selection bias and residential mobility. In Neighbourhood effects research: New perspectives
(pp. 79–99). Dordrecht: Springer. https://doi.org/10.1007/978-94-007-2309-2_4.
Hernandez, F., & Becerra, A. (Eds.). (2017). Marginal urbanisms: Informal and formal development in cities of Latin America. Cambridge Scholars Publishing.
Hoorn, Andre van (2008). A short introduction to subjective well-being: Its measurement, correlates and policy uses. Prepared for the international conference. https://
www.semanticscholar.org/paper/A-Short-Introduction-to-Subjective-Well-being%3A-and-Hoorn/d0b4ca1dad8989378034bd4f2af6031e73d1d5f7.
Jencks, C., & Mayer, S. (1990). The social consequences of growing up in a poor neighborhood. In Inner-city poverty in the United States. National Academies.
Joseph, M. L. (2008). Early resident experiences at a new mixed-income development in chicago. Journal of Urban Affairs, 30(3), 229–257. https://doi.org/10.1111/
j.1467-9906.2008.00394.x
Joseph, M. L., Chaskin, R. J., & Webber, H. S. (2007). The theoretical basis for addressing poverty through mixed-income development. Urban Affairs Review, 42(3),
369–409. https://doi.org/10.1177/1078087406294043
Kahneman, D., Diener, E., & Schwarz, N. (1999). Well-Being: Foundations of hedonic psychology. Russell Sage Foundation.
Katz, L. F., Kling, J. R., & Liebman, J. B. (2000). Moving to Opportunity in boston: Early Results of a randomized mobility experiment (working paper No. 7973). National
Bureau of Economic Research. https://doi.org/10.3386/w7973
Kawachi, I., & Subramanian, S. V. (2007). Neighbourhood influences on health. Journal of Epidemiology &Community Health, 61(1), 3–4. https://doi.org/10.1136/
jech.2005.045203
Kearns, A., & Mason, P. (2007). Mixed tenure communities and neighbourhood quality. Housing Studies, 22(5), 661–691. https://doi.org/10.1080/
02673030701474628
Kearns, A., McKee, M. J., Sautkina, E., Cox, J., & Bond, L. (2013). How to mix? Spatial configurations, modes of production and resident perceptions of mixed tenure
neighbourhoods. Cities, 35, 397–408. https://doi.org/10.1016/j.cities.2013.03.005
Kleit, R. G. (2001). The role of neighborhood social networks in scattered-site public housing residents' search for jobs. Hous. Policy Debate, 12(3), 541–573. https://
doi.org/10.1080/10511482.2001.9521418
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
15
Kleit, R. G. (2013). Commentary: False assumptions about poverty dispersal policies. Cityscape, 15(2), 205–209.
Kling, J. R., Liebman, J. B., & Katz, L. F. (2007). Experimental analysis of neighborhood effects. Econometrica, 75(1), 83–119. https://doi.org/10.1111/j.1468-
0262.2007.00733.x
Kumar, V. (2020). When heritage meets creativity: A tale of two urban development strategies in kampong glam, Singapore. City &Community, 19(2), 398–420.
https://doi.org/10.1111/cico.12427
Lamb, K. E., Thornton, L. E., King, T. L., Ball, K., White, S. R., Bentley, R., Coffee, N. T., & Daniel, M. (2020). Methods for accounting for neighbourhood self-selection in
physical activity and dietary behaviour research: A systematic review. International Journal of Behavioral Nutrition and Physical Activity, 17(1), 45. https://doi.org/
10.1186/s12966-020-00947-2
Lees, L. (2008). Gentrification and social mixing: Towards an inclusive urban renaissance? Urban Studies, 45(12), 2449–2470. https://doi.org/10.1177/
0042098008097099
Leventhal, T., & Brooks-Gunn, J. (2000). The neighborhoods they live in: The effects of neighborhood residence on child and adolescent outcomes. Psychological
Bulletin, 126(2), 309–337.
Levin, I., Santiago, A. M., & Arthurson, K. (2022). Creating mixed communities through housing policies: Global perspectives. Journal of Urban Affairs, 44(3), 291–304.
https://doi.org/10.1080/07352166.2021.1983442
L
evy-Vroelant, C. (2014). Contradictory narratives on French social housing: Looking back and looking forward. Housing Studies, 29(4), 485–500. https://doi.org/
10.1080/02673037.2014.882498
Ludwig, J., Liebman, J. B., Kling, J. R., Duncan, G. J., Katz, L. F., Kessler, R. C., & Sanbonmatsu, L. (2008). What can we learn about neighborhood effects from the
moving to opportunity experiment? American Journal of Sociology, 114(1), 144–188. https://doi.org/10.1086/588741
Lupton, R., & Tunstall, R. (2008). Neighbourhood regeneration through mixed communities: A ‘social justice dilemma. Journal of Education Policy, 23(2), 105–117.
https://doi.org/10.1080/02680930701853013
Luttmer, E. F. P. (2004). Neighbors as negatives: Relative Earnings and well-being (No. 10667; NBER working papers). National Bureau of Economic Research, Inc. https://
ideas.repec.org/p/nbr/nberwo/10667.html.
McDool, E. (2017). Neighbourhood Effects on educational attainment: Does family background Influence the relationship? (No. 2017002; working papers). The University of
Sheffield, Department of Economics. https://ideas.repec.org/p/shf/wpaper/2017002.html.
Morgan, S. L., & Harding, D. J. (2006). Matching estimators of causal effects: Prospects and pitfalls in theory and practice. Sociological Methods &Research, 35(1), 3–60.
https://doi.org/10.1177/0049124106289164
Morrison, P. S. (2010). Local expressions of subjective well-being: The New Zealand experience. Regional Studies, 45(8), 1039–1058. https://doi.org/10.1080/
00343401003792476
Morrison, P. S. (2011). Local expressions of subjective well-being: The New Zealand experience. Regional Studies, 45(8), 1039–1058. https://doi.org/10.1080/
00343401003792476
Musterd, S., Murie, A., & Kesteloot, C. (Eds.). (2014). Neighbourhoods of poverty: Urban social exclusion and integration in Europe (1st ed. 2006 edition). Palgrave
Macmillan.
Najafpour, H., Bigdeli Rad, V., Lamit, H. B., & Rosley, M. S. F. B. (2014). The systematic review on quality of life in urban neighborhoods. Life Science Journal, 11(7),
355–364 (Scopus).
Ostendorf, W., Musterd, S., & Vos, S. D. (2001). Social mix and the neighbourhood effect. Policy ambitions and empirical evidence. Housing Studies, 16(3), 371–380.
https://doi.org/10.1080/02673030120049724
Pickett, K. E., & Pearl, M. (2001). Multilevel analyses of neighbourhood socioeconomic context and health outcomes: A critical review. Journal of Epidemiology &
Community Health, 55(2), 111–122.
Poor households in affluent neighbourhoods. (2014, November 8). El tiempo. http://www.eltiempo.com/archivo/documento/CMS-14808564.
Powell, A., & Sanguinetti, P. (2010). Measuring quality of life in Latin America's urban neighborhoods: A summary of results from the city case studies. In The quality of
life in Latin American cities: Markets and perception (pp. 31–64). Inter-American Development Bank.
Qizilbash, M. (2006). Capability, happiness and adaptation in sen and J. S. Mill. Utilitas, 18, 20. https://doi.org/10.1017/S0953820805001809, 01.
Robeyns, I. (2003). Sen's capability approach and gender inequality: Selecting relevant capabilities. Feminist Economics, 9(2–3), 61–92. https://doi.org/10.1080/
1354570022000078024
Robeyns, I. (2005). Selecting capabilities for quality of life measurement. Social Indicators Research, 74(1), 191–215. https://doi.org/10 .1007/s11205-005-6524-1
Rosenbaum, Paul, & Rubin, Donald (1983). Biometrika, 70(1), 41–55.
Rosenbaum, P. R. (2002). Observational studies (2nd ed.). Springer-Verlag www.springer.com/gb/book/9780387989679.
Sampson, R. (2008). Moving to inequality: Neighborhood effects and experiments meet social structure. American Journal of Sociology, 114, 189–231.
Sampson, R. J. (2012). Great American city: Chicago and the enduring neighborhood effect. University of Chicago Press.
Sanbonmatsu, L., Ludwig, J., Katz, L., Gennetian, L., Duncan, G., Kessler, R., Adam, E., McDade, T., & Lindau, S. T. (2011). Moving to opportunity for fair housing
demonstration program -final impacts evaluation. Department of Housing &Urban Development. Office of Policy Development and Research.
SDP. (2012). Plan de desarrollo, Bogota Humana 2012-2016. Secretaria Distrital de Planeaci
on. http://www.sdp.gov.co/portal/page/portal/PortalSDP/ciudadania/
PlanesDesarrollo/BogotaHumana/2012_2016_Bogota_Humana_Plan_Acuerdo489_2012.pdf.
SDP. (2019). Segregacion residencial socioecon
omica (SRS) en Bogot
a 20017 (No. 108; Estudios macro. Secretaria de Planeaci
on Distrital de Bogota.
SDP. (2021). Hacia una nueva metodología de estratificaci
on socioecon
omica urbana. Secretaria Distrital de Planeaci
on de Bogota. https://www.sdp.gov.co/sites/default/
files/hacia_una_nueva_metodologia_de_estratificacion.pdf.
Sen, A. (1985). Well-being, agency and freedom: The dewey lectures 1984. The Journal of Philosophy, 82(4), 169. https://doi.org/10.2307/2026184
Sharkey, P. (2013). Stuck in place: Urban neighborhoods and the end of progress toward racial equality. University of Chicago Press.
Slater, T. (2013). Your life chances affect where you live: A critique of the ‘cottage industry’of neighbourhood effects research. International Journal of Urban and
Regional Research, 37(2), 367–387.
Smith, J., & Todd, P. (2005). Does matching overcome LaLonde's critique of nonexperimental estimators? Journal of Econometrics, 125(1), 305–353. https://doi.org/
10.1016/j.jeconom.2004.04.011
Steinmetz-Wood, M., Wasfi, R., Parker, G., Bornstein, L., Caron, J., & Kestens, Y. (2017). Is gentrification all bad? Positive association between gentrification and
individual's perceived neighborhood collective efficacy in montreal, Canada. International Journal of Health Geographics, 16(1), 24. https://doi.org/10.11 86/
s12942-017-0096-6
Stuart, E. A., & Rubin, D. B. (2008). Best practices in quasi–experimental designs: Matching methods for causal inference. In J. Osborne (Ed.), Best practices in
quantitative methods (pp. 155–176). SAGE Publications, Inc. https://doi.org/10.4135/9781412995627.d14.
Tach, L. (2014). Diversity, inequality, and microsegregation: Dynamics of inclusion and exclusion in a racially and economically diverse community - ProQuest.
Cityscape, 16(3). http://libsta28.lib.cam.ac.uk:2150/docview/1628955098/fulltextPDF?accountid¼9851.
Thibert, J., & Osorio, G. A. (2014). Urban segregation and metropolitics in Latin America: The case of Bogot
a, Colombia. International Journal of Urban and Regional
Research, 38(4), 1319–1343. https://doi.org/10.1111/1468-2427.12021
Tian, Y., Bi, X., & Li, D. (2007). Feasibility analysis of mixed-income housing in China. Frontiers of Architecture and Civil Engineering in China, 1(3), 371–377. https://
doi.org/10.1007/s11709-007-0050-y
Uribe, C., & Pardo, C. (2006). La ciudad vivida: Movilidad espacial y representaciones sobre la estratificaci
on social en Bogot
a. Universitas Humanística, 62(62). https://
doi.org/10.11144/2214
Vergara, L. (2021). Convivencia y conflictos en barrios de ingresos mixtos [1]. Revista Bit
acora Urbano Territorial, 31(1), 41–52.
Watts, M. (2009). Sen and the art of motorcycle maintenance: Adaptive preferences and higher education. Studies in Philosophy and Education, 28(5), 425–436. https://
doi.org/10.1007/s11217-009-9127-5
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
16
Weinberg, B. A., Reagan, P. B., & Yankow, J. J. (2004). Do neighborhoods affect hours worked? Evidence from longitudinal data. Journal of Labor Economics, 22(4),
891–924. https://doi.org/10.1086/423158
Wendimu, M. A., Henningsen, A., & Gibbon, P. (2016). Sugarcane outgrowers in Ethiopia: “Forced”to remain poor? World Development, 83(Supplement C), 84–97.
https://doi.org/10.1016/j.worlddev.2016.03.002
Wilson, W. J. (1987). The truly disadvantaged: The inner city, the underclass, and public policy (New edition edition). University Of Chicago Press.
Yang, S., Kim, H., Kim, S.-N., & Ahn, K. (2018). What is achieved and lost in living in a mixed-income neighborhood? Findings from South Korea. Journal of Housing and
the Built Environment, 33(4), 807–828. https://doi.org/10.1007/s10901-017-9586-x
J.F. Bucheli Journal of Urban Management xxx (xxxx) xxx
17