Embed Size (px)
Citation preview
Ethnic segregation and collective political action in urban England
Mariana Skirmuntt∗
Department of Government, University of Essex
Abstract
A number of studies in the �eld of political geography have focused on the ethnic pro�le of the
locality to examine its e�ect on political participation. Mainly focused on voter turnout and framed
in the �hyper-segregated� urban America, the research on the impact of the spatial distribution
of ethnic groups barely studied its e�ects on alternative forms of political action in Britain. This
paper claims that together with the economic features, the characteristics of the local distribution of
ethnic minorities can account for individuals' degree of collective political engagement. Considering
the three main ethnic groups in England (white, Black and Asian), I test competing theories
of segregation e�ects. I take into account the predictions elaborated by the �Ethnic community
model�, the �Racial threat models� and the mechanisms of inter-personal trust to explain the scope
of cooperative actions. For the empirical analysis, I pooled two rounds of the Citizenship Survey
conducted in England in 2005 and 2007 and run overall and separate logistic models to assess the
impact of the local authorities' dissimilarity index. I found that concentration of co-ethnic does not
a�ect the Asian's and Black's propensities to participate in collective political activities. However,
evidence suggests that environments with higher segregation of Asian population depress whites'
collective political activism. This research seeks to contribute to the current discussion on the causes
and consequences of the spatial patterns of residence of ethnic populations in urban England.
∗Electronic address: [email protected]
1
I. INTRODUCTION
Within the urban studies, the segregation of ethnic minorities in Britain has been subject
of heated debate on conceptual and measurement matters but most importantly, regarding
the scope of this social phenomenon. It was suggested that Britain was starting to su�er
from a ghettoisation of urban localities, a process comparable to the situation that exists in
many segregated American cities such as Chicago (Phillips 2005).[1] These statements were
severely challenged by the socio-geography literature which claimed that such a high degree
of segregation is not supported by the statistical evidence in Britain (Peach 1996, Peach
2009); in words of Simpson (2007), anxieties of other people's race or origin �are better seen
as ghettos of the minds rather than ghettos of realities�(pp.423).
The e�ects of the patterns of residence of ethnic minorities are currently under public and
scholarly debate. After the riots in the northern cities of Bradford, Burnley and Oldham
in 2001, it emerged a vivid interest to assess the causes and consequences of the spatial
distribution of ethnic groups across the boundaries of the localities. In the area of political
studies it remains under-explored the political consequences of the local distribution of
ethnic groups in Britain. Even though it is not my intention to delve into the discussion on
the level of segregation currently displayed in Britain, it is worth noting that the patterns
of spatial distribution vary widely among ethnic groups and across local authorities (Peach
1996, Peach 1998). Does this variability have speci�c political consequences to both majority
and ethnic minority groups?
In this paper, I shall focus on the individuals' political participation in collective actions.
I take into account the features of the contexts because they are crucial to understand
geographically speci�c social interaction and because they can be strong basis for the con-
struction of collective identity - requirement for any collective action (Miller 2000). In
particular, I investigate if the varying degree of clustering of ethnic minorities in the locality
can contribute to explain individuals' collective political engagement assuming that there
may be varying degrees of co-ethnic identi�cation, trust and inter-group con�ict in the ag-
gregation of preferences. Therefore, the aim of this paper is twofold. On the one side, I will
grasp on the degree of ethnic minorities' political involvement taking into account the local
segregation of individuals with similar ethnic backgrounds. On the other side, I attempt to
determine empirically whether local minorities' segregation can be a signi�cant predictor of
2
majority's' propensities to political action.
Since the seminal work of V.O. Key (Key 1949), we know that it necessary to inspect the
racial composition of the contexts in order to understand why people behave di�erently in
politics. Consistent with this line of research, a number of studies have focused on the ethnic
pro�le of the locality with the intention of examining its e�ects on mainstream political
participation (Uhlaner, Cain et al. 1989, Schlichting, Tuckel et al. 1998, Hill and Leighley
1999, Fieldhouse and Cutts 2008). However, the empirical investigation of the impacts of
the local ethnic pro�le on non-electoral participation is currently on a preliminary stage.
The available literature in British contexts gives little clues to understand, for instance, the
political behavior of Asian and blacks living among other co-ethnics.
The analysis that follows draws mainly on the premise that, taking into account for
economic factors, ethnic features of the localities impact di�erently on people with di�erent
ethnic backgrounds. The availability of a robust micro data set matched with aggregate data
from English local authorities allows me to begin to explore a topic - the linkages between
local ethnic make-up and the individuals' engagement in non-electoral participation - hardly
covered in the literature (Pattie, Seyd et al. 2004).
II. THEORIES ON SEGREGATION EFFECTS
Segregation of ethnic minorities: positive e�ects
Taking into account previous studies which pointed out that consciousness and identity
is created in place (Giddens 1984, Agnew 1989, Kirby 1989) I will start by explaining how
segregation can a�ect the collective political engagement of ethnic minorities.
The �ethnic community model� assumes that minority groups tend to face common prob-
lems and have interests that are speci�c and distinct from those of the majority of the
population. The will to express these interests or preferences by individuals belonging to
the same minority ethnic group often drives them to the development of political activities
of some sort. According to this approach, when a large number of individuals with the
similar ethnic background live and share the experiences that are generated in the same
geographical space the chances of political involvement among the group's members rise.
Recent research in Britain found that South Asians following the Muslim and Hindu
3
faiths tend to have a higher propensity of voter turnout and voting registration when living
among co-ethnics (Fieldhouse and Cutts 2008, Fieldhouse and Cutts 2008). Another inves-
tigation showed that there is not a signi�cant association between ethnic density and civic
engagement (Sta�ord, Bécares et al. 2010).
However, I would argue that the e�ects of the size of the minority ethnic group on political
participation would be neglectable if the minority is intermingled with the majority white
population. Consequently, it is the geographical distribution of ethnic minorities within the
locality what deserves further examination. The ethnic community e�ects can be stronger
if the ethnic group are spatially separated from the rest of the local population. In sum, it
is the spatial gathering of individuals from the same ethnic group in a given locality what
easier the communication among individuals with similar interests, needs and preferences.
This contact facilitates what the traditional literature called �group consciousness�, this is,
the awareness of the group's relative position in the larger society and, at the same time,
the commitment of its member to further actions to promote the group's interests (Miller,
Gurin et al. 1981).
Minority groups living in segregated places will create and reproduce the feeling of be-
longing to a particular ethnic community. Even if they belong to an ethnic group but they
do not live in the enclave, they can notice anyway the local segregation between own's group
and the majority. Moreover, spatial concentration can contribute largely to the creation of
their own organizations and leaders who in turn can evoke the ethnic identity for the mobi-
lization of the local ethnic minority residents (Ramakrishnan and Espenshade 2001, Uslaner
2011).
Segregation of ethnic groups can visualize the commonalities of its members and enhance
the in-group view that the status quo places them in a disadvantageous position in relation
to the rest of the population. Ultimately, the resulting in-group solidarity will likely generate
the organization of some political action. In brief, in-group cohesion or connectedness of
ethnic communities found in contexts of high residential segregation can potentially promote
political engagement.
In addition, rational models highlight the powerful force of identity and the bene�ts
derived from participation. According to the notion of relational goods proposed by Uhlaner
(1989) and Gui (1987), activities such as joining a group of volunteers or participating in a
political discussion are intangible outputs which people involved enjoy per se. The model of
4
relational goods developed by Uhlaner suggests that �the identity of individual participants
with the group is strengthened through the action; the relational portion of the bene�t
refers not to power and in�uence but to the actor's share of the enhanced identity� (pp.
257). Therefore, the identity and the interaction with members of the in-group constitute
an incentive to political action. Moreover, according to Leighley (2001), the interpersonal
identi�cation and interaction among members of a group can provide information which
leads to a reduction of the costs of participation.
Despite empirical investigation is not extensive, recent �ndings suggests that this ap-
proach holds for speci�c forms of political action such as voter turnout. For instance,
Schlichting et. al. (1998) found evidence to support this approach when they studied the
e�ects of hyper-segregation of two American cities on voter turnout. With certain caution
about the generalization of their case-study �ndings to other contexts, authors concluded
that ethnic communities exerts a mobilisation e�ects which leads to higher levels of turnout.
Interestingly, residential sorting by ethnicity may not only have implications for the be-
havior of ethnic minority groups. The size and segregation of these groups can also motivate
white majority to participate. It is reasonable to think though, that the consciousness of
being relative disadvantaged because of ethnic reasons is not the factor that stimulates po-
litical action in the case of the white majority. Therefore, an alternative approach should
be considered.
In a deep and critical review of the literature on racial threat mechanisms and political
mobilization (Key 1949to name some, Blumer 1958, Blalock 1967) Enos (2011) provides a
psychology-based approach to explain the link between threat and political action taking into
account a spatial dimension. Enos distinguishes two types of racial threats: the �material�
and the �psychological�. The material explanation of threats assumes that individuals react
rationally to an out-group which is challenging their social, economic or political position.
The mechanism assumes that if the size of the out-group increases, the resource competition
raises, leading individuals to political action. This is particularly understandable in those
localities where tensions may arise as consequence of gaps in resources in housing, education
and employment provision. In the local context, the material threat is palpable among
whites since they have the belief that ethnic minorities `were using resources � housing,
education, health care � which rightly belonged to white people.� (CRE 1998)
However, Enos observes some limitations in this approach and explores the Social Identity
5
Theory to ground his own understanding of the relationship between threat and political
action. Building on this literature, he indicates that individuals can behave in a non-rational
way (since the threat can be �real� or just �perceived�) when sacri�cing their own individual
bene�t for the sake of their own group's utility. Since individuals con�ate individual and
group utility, he argues, group identity is a powerful in�uence on political behavior. In line
with this premise, Enos argues that �larger groups are more salient and more salient groups
stimulate political participation� (pp.8). The salience of a group is a key element which leads
to an increasing stereotypization of the out-group and di�erentiation of policy preferences
that foster political engagement among in-group members.
But it is not only about the size of the out-group. The spatial separation of ethnic
groups is another factor which contributes to explain the salience of the out-group. Enos
(2010) suggests that the study of threat e�ects should also take into account the way that
groups are related spatially. In relation to this, he states that �the greater salience of the
more concentrated groups is because the more concentrated out-group is more distinct and,
therefore, more salient than a group that that is spatially integrated with other groups�. For
instance, he has recently found through a natural experiment in Chicago a decrease of white
voter turnout rates after the removal of highly concentrated African American neighbors
as a result of a project of reconstruction of public housing. In sum, Enos proposes that
concentration delineates the spatial impact on political behavior.
The implications for the white majority living in more ethnically segregated environments
would be greater participation in collective action to protect the social values and norms
they consider as traditional. For instance, places with increasing larger Asian enclaves may
constitute potential targets for mobilizing groups such as the English Defence League which
have been promoting in several occasions protests about planning issues over the construction
of local mosques.
Segregation of ethnic minorities: negative e�ects
A contrasting account of political (dis)engagement among population in segregated con-
texts considers the concepts of trust and expectancies over other ethnic minority's members.
When a group organizes a political activity it will try to get as many people to support the
cause and join it to succeed. If the di�erent groups of people that make up the locality do not
6
know each other because they live apart, the chance of interaction and contact is minimal
and therefore the level of interpersonal trust decreases. Uslaner (2008) found that nations
with a signi�cant level of ethnic segregation are less trusting because of the lower chances
of inter-group interaction. More recently, he reports that in the UK residential segregation
constitutes the root of the low trust (Uslaner 2011).
Speci�cally, trust between people is central to explain the development of collective activ-
ities since it reduces the participation costs. It diminishes the uncertainty on others' values,
attitudes and behavior. Dennis Chong (1991) explains that engagement is a function of the
amount of people individuals expect will get involved. So, if you start or join a protest, you
will do so because you expect to �nd a good amount of people in your local area that shares
the similar values and interests and that are willing to get engaged. As Uslaner (2002)
explains, �when others share our basic premises, we face fewer risks when we seek agreement
on collective action problems� (pp. 18).
The ethnic separation damages the chances o breeding a local collective consciousness
required to breed collective actions to address local issues. White majority and ethnic mi-
norities living in segregated contexts may encounter di�culties to �nd this common base to
start a collective political enterprise. In segregated contexts, there are deeper perceptions of
social, religious or cultural di�erences between ethnic groups. Geographical ethnic segrega-
tion may exacerbate the social distance between groups which can reduce the interpersonal
trust required to develop cooperative political actions. For instance, Asian minority segre-
gation can represent the type of separation which can harm the most the chances of the
development of a joint venture: separate cultures, language and religious beliefs discourage
successful collective actions. In brief, for local residents to live in segregated context may
imply a decrease in the amount of interpersonal trust and consequently a reduction of the
expectation to gather a signi�cant sum of participants in political activities likelihood of
success in political action.
In the case of the ethnic minorities, the e�ect of segregation can be even stronger be-
cause in-group interactions may dampen the contact with the broader society. Interactions
among individual from the same ethnic group may provide valuable social networks which
facilitate family organisation, economic enterprises, etc. Indeed, some point out the negative
consequences of a potential dispersal of these residents since it would remove the vital social
solidarity (Rex 1981). Living in such context can reduce the chances of political contest
7
since family and close social networks can compensate the lack of welfare state interventions
in these areas (Ireland 2008). The social networks created in places with high concentration
of speci�c minorities will promote a separate �ow of information, generally more focused
on the native or ancestors' countries rather than the country of residence. Tam Cho (1999)
highlighted the consequences of the di�erentiated informational �ow and social networks of
the minorities indicating that they �may not derive the same sort of satisfaction from a�rm-
ing allegiance to the political system or have the same sense of responsibility for preserving
the democratic process� (pp. 1144).
In fact, segregation will likely lead to a gathering of a signi�cant proportion of people who
do not have a good level of �uency in English (Anwar 2010) and population with modest
information about local politics and how the democratic institutions work. Local networks
su�ering from this lack of abilities and information may tend to reduce the likelihood of
political discussion and political action.
The works of Huckfeldt and his associates have been repeatedly demonstrating the im-
portance of social networks for providing political cues which can derive in political action
(Huckfeldt 1986, Huckfeldt and Sprague 1995, Huckfeldt 2001). It can be argued that in seg-
regated contexts, social networks' exchange of politically-relevant information is not frequent
and therefore individuals are neither motivated to hold political conversation nor stimulated
to translate it into political action (McClurg 2003).
If political information or knowledge does not circulate among co-ethnics, individuals
have less resources and motivation to engage in local politics. In sum, residential ethnic
segregation could be associated to lower rates of political engagement as concentration is
frequently related to a lack of contact with the broad society. Segregation could prevent
minorities from a political learning (when minorities are newcomers) and acquisition of
participatory norms.
Accounting for the socio-economic conditions of the locality
The socio-economic characteristics of the context play a major role for determining the
political behavior of local residents. An extensive body of contextual e�ects research argues
that people are placed in social environments which a�ect their exposure to certain political
information and condition the social interactions among individuals (Carmines and Huck-
8
feldt 1996). It has been argued that a number of political activities involve speci�c resources
(money, skills, time) to be carried out. These resources are not evenly distributed across the
di�erent segments of the population; rather they are usually found among a�uent people.
At the same time, Huckfeldt (1979) suggested that engagement in politics can be learned
and stimulated. For that reason, he concludes that �higher status contexts populated by
more politically activated individuals can encourage participation� (pp.581). Those residents
living in high-status localities will tend to feel the pressure to participate and will enjoyed
more opportunities to get involved. Therefore, we can expect that the richer the locality,
the more it prevails participatory norms among its residents and the greater the chances to
engage in some political activity.
In every local area, there is a speci�c set of values and norms which tend to be followed
by its residents. Norms are �expectations about action -one's own action, that of others, or
both- which express what action is right or what action is wrong� (Coleman 1987). In a�uent
areas people are in�uenced by the participatory norms; conversely, the theory indicates that
in localities su�ering from high levels of poverty, civic norms prescribing social cooperation
do not prevail among local residents. Empirical investigations found evidence to support
this contextual theory.
In order to avoid the risk of con�ating the e�ects of local deprivation and ethnic segre-
gation e�ects, I will take into account for both in the models. If we departure the analysis
considering the complex nature of the spatial distribution of ethnic minorities across the ter-
ritory, it must be acknowledged that the level of a�uence of the locality where the individual
is situated should be also considered. The introduction of this compositional e�ect makes
it possible to assess the independent e�ects of the residential patterns of ethnic minori-
ties. Thus, despite socio-economic features explain some variance in participation, ethnic
characteristics also condition participation.
III. RESEARCH DESIGN
In order to assess the di�erent theories on the impact of contextual e�ects on political
participation of individuals, I pooled two rounds of the Citizenship Survey conducted in
2005 and 2007. This repeated cross-sectional survey has an outstanding core sample size
in each round (9,691 and 9,336 respectively) and a boost sample of ethnic minority respon-
9
dents (4,390 and 4,759 respectively). In this research, in addition to the white majority,
I will focus on the political behavior of the largest minority ethnic groups in Britain: the
Asians (Bangladeshi, Pakistani, Indian and other Asian respondents) and blacks (Caribbean,
African or other black backgrounds).
The individual data was merged with aggregate data corresponding to the local author-
ities where the respondent resides.[2] While the territorial units in question, namely the
local authorities, cannot be equated with smaller geographical units such as the neighbor-
hoods (traditionally used in the contextual e�ect literature), I argue that they are the most
convenient micro-units for the purpose of this research. The reason is that the present re-
search's aim is to examine individuals' action to in�uence the public policy making process,
including the local political institutions, the LAs. In Britain, the local politics constitute a
signi�cant dimension of democracy as G. Parry et al. claim: �where a governmental agency
is involved in the matter, it is local rather national government which very often carries the
responsibility� (Parry, Moyser et al. 1992).
An additional advantage of the Citizenship Survey is that it contains a wide set of ques-
tions on political engagement which served to create the dependent variable: activities to
in�uence institutions collectively. Respondents reported if they developed the following po-
litical activities during the last 12 months: to attend a public meeting/rally, to take part in
a public demonstration or protest, to sign a petition, to attend a public meeting (about local
services) and, to get involved in a group to discuss local services. With this information a
created a dummy variable with the category 1 if the individual developed at least one of
the activities above listed and 0 coded as no participation. The grouping of these actions of
citizen cooperation was done following previous research using the same data (John, Field-
house et al. 2011). Unfortunately, the questionnaire does not provide information regarding
the nature, target or purpose of the political actions developed.
The key aggregate variable is the segregation of ethnic minority groups.[3] The segre-
gation measure was estimated using the dataset of population by ethnic groups (Census
2001) available at the Lower Layer Super Output Areas (LSOA). The formula I applied to
obtain the segregation scores was the traditional index of segregation (Duncan and Dun-
can, 1955). It measures the distribution of a particular population group in a geographical
space. It varies between zero and one, values corresponding respectively to an exactly equal
distribution and a maximum segregation distribution. The value of this index can also be
10
interpreted as the proportion of the minority group that would have to change residence to
obtain an equal distribution (Jakubs 1981, Massey and Denton 1988). The formula for the
index of dissimilarity, D, is given by the following,
D = 0.5∑[(
xij
Xl
)−(kijKl
)]where xij is the size of the ethnic group in the LSOA j within local authority l, Xl
represents the size of the ethnic group in the local authority l ; kij stands for the size of the
population from all other ethnic groups combined in LSOA j within local authority l, and Kl
corresponds to the size of the population from all other ethnic groups combined in the local
authority l. Literature from geographical studies explains that there is potential bias in the
values of the index of dissimilarity when the size of the ethnic group in the geographical unit
is small (Voas and Williamson 2000). Since the main purpose of this research is to explore
the segregation e�ects, I will focus the analysis in those localities where Asian and black
minorities represent a signi�cant proportion of the population. Therefore, I only retained
the scores of the localities which pass the threshold of 1,000 individuals from the ethnic
group considered (Asian and Black groups). Even though the 1,000 group population cuto�
may be considered arbitrary, it was also applied by some other research in geography studies
framed in English localities (Peach 1996, Peach 1998). Notwithstanding that the two ethnic
categories are rather crude (they are encompassing large number of diverse nations and
migration pattern), the grouping enable the calculation of segregation scores at local level.
Alongside the key independent variable, I placed a group of control variables at individual
and local level. Level-one controls include: gender (female as reference category), a dummy
variable for foreign birth, age and age squared, marital status (coded "1" if married, "O"
otherwise), a dummy for those whose highest quali�cation is a degree (coded �1�), a hous-
ing tenure dummy (home owner coded as �1�), length of time living in the neighborhood
(seven bands of years of residence), religion practice (coded as �1� when actively practic-
ing the religion), and �nally, an occupationally-based measure derived from the National
Statistics Socio-economic classi�cation (NS-SEC) displayed as a set of dummies including:
a. Higher and lower management; b. Intermediate occupations; c. Semi-routine and rou-
tine occupations (the omitted variable corresponds to those who never worked/long-term
unemployed/student). In the whole sample model, I included two ethnicity dummies for
Asian and Black respondents (whites are the reference group), measured as a self-reported
11
variable.
The covariates at aggregate level include the log-transformation of total population of
the locality (Mid-2005 estimated resident population released by the O�ce for National
Statistics), the 1991 to 2004 population change (in percentage) derived from ONS Mid Year
Population Estimates, the local authority average index of multiple deprivation 2004 (IMD)
and a measure of diversity of socio-economic conditions within the local authority which
consists on the standard deviation of the index of multiple deprivation score from each of
the sub-units which conforms the locality (LSOAs) (Greasley and John 2011).[4]
It will be fair to point out the limitation of the present investigation related to selection
e�ects. The residential self-selection of individuals might constitute a matter of concern in
this research because white people who tend to refuse to contact minority groups will prefer
to live in locations homogeneously white. Similarly, ethnic groups tend to move to localities
where they can keep customs and avoid threats and discrimination. While I will not deny
that this factor deserves attention, literature on its impacts indicates that it might not
necessarily imply a major concern. On the one hand, it can be argued that the economic
positions of households are the main driving factors when deciding residential locations.
As van Ham and Manley (2012) explain, �a key determinant of choice in the residential
housing market is �nance� (pp. 4). This may entail that in-group ethnic preferences are less
important than economic at the time to select a residential location. Depending on their
�nancial resources, households are in better or worst position to exercise choice.[5] Simpson
(2007) showed with Census data that white population is not moving from highly non-white
areas. On the other hand, it was also suggested that surroundings where residents with
ethnic minority backgrounds live are associated with undesired features such as deprivation,
low-quality services, unemployment and crime. For that reason, the `racial proxy hypothesis'
proposes that the white majority prefers to avoid locations with higher amount of ethnic
minority groups not necessarily because they have racial animosity against them but because
these local areas are often deprived.
Housing segregation of ethnic groups is a complex process widely discussed in urban
studies. There is an extensive literature on both �choice� and �constrain� sorting. According
to Phillips (1998), segregation is not only derived by economic constraints but also �cultural
factors, which serve to anchor minority ethnic group members in the (often deprived) ethnic
territory, may intervene� (pp. 1691). The �choice� theories of segregation support the view
12
that people can select speci�c neighborhoods to settle their households not only based on
a range of a�ordable houses but also on the basis of other elements such as the ties with
co-ethnics or relatives established in the area previously (Dahya 1974, Winstanley, David et
al. 2002). On the other side, constrain theorists sustain that it cannot be disregarded the
fact that the roots of alleged `self-segregation' are found in the level of poverty rather on a
real choice (Hickman and Robinson 2006, Pawson and Watkins 2007, Hedman, van Ham et
al. 2011). Moreover, due to the still existing discriminatory estate practices, members of
ethnic minorities face restricted housing choices (Phillips 1998, Harrison and Phillips 2003).
The residential sorting of an ethnic group might not certainly entail positive or negative
racial attitudes towards whites or other ethnic groups but just simply historical patterns of
labour immigration, economic opportunities and housing conditions.
However, I must acknowledge that when studying the impact of the residential sorting I
cannot overcome the fact that �the overall measure makes no distinction between voluntary
and enforced segregation� (Simpson 2004). Since I developed a cross-sectional research
design I cannot derive de�nitive conclusion about the link between contextual measures and
participation rates. I will try to partially reduce the selection e�ects through the introduction
of variables to control for individual background features. Since this strategy may not fully
neutralize the selection e�ects, further longitudinal studies will bring more insight to the
unique contribution of segregation to explain individuals' political action.
IV. EMPIRICAL ANALYSIS
I start the empirical tests of the competing hypothesis on ethnic segregation e�ects by
analyzing some simple descriptive statistics. The joint frequency distribution of cases based
on ethnicity and the dependent variable, collective engagement, shows that the group of
the Asian respondents has the lowest percentage of participation (16.6%), followed by the
Black minority (20.4%) and White majority (30%). This ethnic gap in the level of political
engagement is in line with the �ndings of previous research in Britain (Saggar 2000, Anwar
2001, ElectoralCommission 2002, Fahmy 2005). In a further exploration of the data, I present
in �gure 1 the whole sample's rates of engagement in collective political activities taking into
account �ve bands of the index of Asian and Black segregation. The graph does not show a
striking variation in the degree of political engagement when we move from the lowest to the
13
Figure 1: Rates of collective political engagement by index of segregation of Asians and Blacks (All
sample)
10
15
20
25
30
< 0.2 0.2 - 0.3 0.3 - 0.4 0.4 - 0.5 > 0.5
Perc
enta
ge
Index of segregation
Asian segregation Black segregation
highest Asian and Black segregation scores. Along the �rst bands of the Black segregation
line, it can be noted a minimal variation in the degree of political action (around 21%)
and only in the most segregated places, the collective engagement of the sample population
increases up to 27%. On the other hand, the line of the Asian segregation presents a slightly
steeper downward trend in the collective participation rates of all respondents when the
degree of segregation increases (from 28% to approximately 20%).
If the data is disaggregated, we �nd that for the three ethnic groups considered, there is a
general downward trend as the degree of segregation increases. For blacks, as we move from
the least Black segregated places to the most segregated, the participation rates decrease
from 23% to 19%. Similarly, for Asians, whose rate goes from 23% to 17% as we move
along the Asian segregation scale, it is di�cult to outline a conclusion regarding the link
between ethnic minorities' segregation and collective political actions. Finally, in the case
of the whites' group, a mixed conclusion can be drawn. While the extent of their political
participation seems to be una�ected when the size of local Blacks segregation increases
(there is a variation of approximately 3%), a larger local Asian segregation is consistently
dropping the frequency of collective engagement among white individuals. As a group,
whites' participation levels decreases by roughly 13% as local Asians segregation gets higher.
For studying in detail the relationship between segregation e�ects, I will move from cross-
14
Figure 2: Rates of collective political engagement by index of segregation of Asians and Blacks
(three ethnic groups)
10
15
20
25
30
35
40
45
< 0.2 0.2 - 0.3 0.3 - 0.4 0.4 - 0.5 > 0.5
Per
cen
tag
e
Index of segregation
Asian segregation (Asians only) Black segregation (Blacks only)
Asian segregation (Whites only) Black segregation (Whites only)
tabulations to regression analysis. Due to the binary nature of the dependent variable, a
series of logistic models were performed taking the cluster structured of the data into account
when calculating the standard errors. It is relevant to recall that the coe�cients in the model
show the change in the log odds of participating in any political activity versus the non-
participation as a result of a one-unit increase in the independent variable. The simplest
models presented in table 1 display the coe�cients of the two key independent variables in
�rst column and the interaction e�ects in the second column.
The null model suggests that the e�ects of segregation of Asians and Blacks on political
engagement are both signi�cant but they have opposite signs. While the concentration
of blacks in the locality raises the likelihood of individuals' collective involvement, Asians
segregation depresses it. This initial exploration gives some support for both arguments of
the positive and negative e�ects of ethnic segregation. However, this model does not permit
to uncover the relationship between segregation and each speci�c ethnic group. Therefore,
15
Table I: Logistic regressions of collective political action for all respondents (without controls)
Collective political action Null Null with
interactions
Asian -1.08***
(0.22)
Black -0.27*
(0.13)
Index of segregation of blacks 1.45*** 0.71*
(0.35) (0.32)
Index of segregation of Asians -0.98*** -1.00***
(0.26) (0.20)
Index of segregation of black*black -0.10*
(0.04)
Index of segregation of Asian*Asian 0.10
(0.06)
Constant -1.21*** -0.86***
(0.03) (0.03)
Observations 13,062 13,053
Robust standard errors in parentheses, with clustering by local authority
* p<0.05 ** p<0.01 *** p<0.001
the second column displays coe�cient of the model with interactions between ethnicity of
the respondent and the segregation of the two ethnic minority groups considered.
The coe�cients indicate that being member of any of the two minority ethnic groups
has statistically signi�cant negative e�ects on political engagement. The model without
covariates also points out that the main e�ects for Black and Asian segregation on the
reference category (i.e. for whites) are positive and negative respectively. Interestingly, the
interactions terms seem to be suggesting that only residential segregation of Blacks has some
e�ects on the same ethnic group: Black residents living in local areas with elevated Black
segregation have less chances of collective engagement than white residents. Living in places
where co-ethnics are more segregated does not necessarily have implications to all ethnic
groups' probabilities of participation.
As previously stated, this research is mainly interested in examining the e�ects of seg-
regation on political participation but the relationship might be conditioned by some other
factors, especially by the degree of a�uence or deprivation of the locality. For that reason,
table 2 displays the coe�cients of the key independent variables controlling for additional
16
Table II: Logistic regressions of collective political action for all respondents (with controls)
Collective political action Main Interaction
effects effects
Asian -0.87***
(0.22)
Black -0.02
(0.15)
Index of segregation of blacks 0.508 0.71
(0.392) (0.37)
Index of segregation of Asians -0.587* -0.86***
(0.246) (0.26)
Index of segregation of black*black -0.11*
(0.05)
Index of segregation of Asian*Asian 0.11
(0.05)
Constant -2.730*** -2.80***
(0.435) (0.44)
Observations 12,356 12,356
Robust standard errors in parentheses, with clustering by local authority
* p<0.05 ** p<0.01 *** p<0.001
covariates at individual and local levels.
For collective participation, the measure of Asian segregation is still signi�cant and keeps
the negative sign. In this case, therefore, the relationship between the dependent variable
and the index of segregation of Asians cannot be attributed to other individual or contextual
factors traditionally considered in the literature of political behavior. The overall model also
indicates that the variable corresponding to the segregation of Blacks maintains the positive
sign but fails to reach conventional standards of statistical signi�cance.
Once interactions are included, the main e�ects for Asian segregation (related to the refer-
ence category, `white') show a statistically signi�cant negative coe�cient; Black segregation
main e�ects preserve the positive sign but show no statistical signi�cance. In addition, es-
timates in the second column demonstrate that even after controlling for a bunch of other
variables those who have Asian backgrounds have signi�cantly lower probabilities on en-
gaging in some collective political activities than whites. This is not the case for Black
respondents. Finally, the estimates of the interactions term are displaying little variation in
terms of signs and statistical signi�cance from the null models. Member of the Black minor-
17
Table III: Logistic regressions of collective action for three ethnic groups (with controls)
Collective political action Asian Black White
respondents respondents respondents
Index of Black segregation in LA -0.60 0.31
(0.46) (0.45)
Index of Asian segregation in LA 0.18 -0.93**
(0.46) (0.29)
Constant -3.86*** -2.95** -2.71***
(0.85) (1.04) (0.55)
Observations 4,369 2,910 5,077
Robust standard errors in parentheses, with clustering by local authority
* p<0.05 ** p<0.01 *** p<0.001
ity living in places with higher levels of co-ethnic segregation are less engaged in collective
political action than whites.
In the last step, I performed the analysis dividing the sample into three ethnic groups
(white, Black and Asian). When modeling separately Whites', Asians' and Blacks' politi-
cal behavior, I intent to take a distance from the traditional strategy which assumes that
majority and minorities share exactly the same determinants of participation. Moreover, I
attempt to avoid the supposition that the behavior of Asians generalizes to that of other
British minorities such as the Black population. For that purpose, I will also allow parame-
ter estimates to vary across the ethnic groups considered in order to assess the varying rates
of collective political engagement.
As it can be seen from the �rst and second column of table 3, the coe�cients for the
Asian and Black groups show that local co-ethnic segregation does not play any role; this
entails that any segregation e�ects appears when separate models are �tted. In few words,
minorities' involvement in collective types of actions is not determined by the degree of local
clustering neither for Asian nor for Black individuals. However, for the white sub-sample,
once again I found that elevated levels of local Asian clustering seem to be depressing the
likelihood of political collective action. On the contrary, Black residential segregation within
the boundaries of the local authority has not statistically signi�cant consequences on whites'
collective action. To illustrate the e�ect, I calculated the whites' predicted probabilities of
18
engaging in some collective political activity. The predicted e�ects are taken by holding all
independent variables at their mean and allowing the index of Asian segregation to vary
over its full range found in the data. Moving from the lowest local Asian segregation, where
white respondents have an almost 0.24 probability of participation, to localities with the
highest Asian segregation implies a reduction of approximately 10 percent, holding all other
factors constant.
The covariates of the last conditional models deserve some analysis. Looking at the
individual level only, I found that there is certain consistency of the coe�cients of three
independent variables across all models and ethnic groups. For all three groups, those who
are foreign born show a signi�cant and negative likelihood of becoming engaged in some
political action. These �ndings related to the political behavior of non-native population
are in line with some previous research (Shaw, de la Garza et al. 2000, Blais, Gidengil et
al. 2004, Just and Anderson 2012). The community attachment or residential stability,
captured in this research as the respondent's length of residence in the neighborhood, also
shows a statistically signi�cant and positive coe�cient. Therefore, it con�rms previous in-
vestigation suggesting that the longer individuals reside in a locality, the more they are aware
and informed about locality's political process and the higher the chances of engagement
(Budge and Urwin 1966, Larcinese 2007). Similarly, as an extensive research on political
participation has been demonstrating, coe�cient show that more educated individuals have
greater propensities to participate (Parry, Moyser et al. 1992, Nie, Junn et al. 1996).
Data con�rms partially the traditional socio-economic model of participation which indi-
cates that the higher the socio-economic position of the individual, the more likely he or she
is to engage in some political activity (Wol�nger and Rosenstone 1980, Verba and Nie 1987,
Leighley and Nagler 1992). For whites, having semi-routine and routine occupations means
lower levels of engagement and, for ethnic minorities, occupying upper categories of the
structure of socio-economic positions represent larger probabilities of political involvement.
Finally, the data supports only in part the �ndings of the literature on religion mobilization
(Sobolewska, Heath et al. 2012, Brady, Verba et al. 1995, Beyerlein and Chaves 2003) since
the frequency of religion practice is relevant for the activism of white respondents only.
A wide proportion of the literature which studies the impact of socio-economic contexts
has been demonstrating that civic and political involvement is generally more intense in
wealthier environments. In general, these settings are composed by individuals with high-
19
incomes and prolonged educational backgrounds (Huckfeldt 1979). Data in this research
does not con�rm this prediction for none of the models. In fact, none of the aggregate level
controls presents any statistical signi�cance.
V. DISCUSSION
In this research I initially assumed that, as some previous studies have indicated, there are
contexts where the organization of collective activities are easier than some others (Agnew
1989, Cho, Gimpel et al. 2006). I tried to assess the e�ect of segregation as one the most
relevant factor for con�ict and identity-forming processes. However, I could not �nd any
strong evidence to claim that ethnic segregation is a key driving force for collective political
action in England. In short, for none of the ethnic groups considered, the ethnic segregation
leads to an increase in con�ictual or cooperative political actions. It is interesting to note that
segregation in any case engender more political action, even more if we take into account, for
instance that �British local authorities manage a signi�cant share of local housing, providing
concrete stakes that could fuel local inter-group disputes.�(Hopkins 2011, pp.503)
Much of the public discourse tends to reproduce the idea that segregation of Asians
merely isolates them from the main institutions of British society, generating alienation or
indi�erence to what is happening in their immediate environment. According to these argu-
ments, the strengthening of their ethnic or religious identities in contexts of concentration
constrains their chances to carry out collective political activities with the rest of society.
The results obtained in this research do not go in that direction. Living in a context of segre-
gation does not lead Asian to participate less in political activities that require cooperation
from other citizens.
At the same time, media often links the political mobilization of the Asian group (espe-
cially in areas with high concentration of Muslims) with activities which exclusively aimed
at achieving greater recognition of religious rights or protection of their ethnic identity (for
instance, demands to provision of religious instruction in local schools, petitions to follow
religion and dietary practices in local public spaces, or petition to broadcast calls to prayer
from a mosque). The argument would indicate that the collective mobilization of Asians
is intense in places with high concentration of co-ethnics because local leaders or cultural
and religious communities are strong enough to in�uence local political institutions. This
20
expectation of greater collective involvement in localities with greater segregation of Asians
is not supported by the data used here.
Regarding the blacks, segregation of co-ethnics is not nurturing more collective political
actions. As Statham (1999) explains, there may be inequalities in terms of �class� within
the Black communities, whose political claims and demands are vanished when the potential
leaders are co-opted. He claims that �the relative absence of grass-roots and community-type
'black' organisations, suggests that the incorporation of upwardly mobile African-Caribbeans
may be to the disadvantage of the majority of African-Caribbeans.� (pp. 621). Moreover,
the integration policies of British institutions aim to grant rights to address experiences of
racial discrimination through individual rather than collective mechanism such as courts or
industrial tribunals. In relation to the political behavior of the ethnic majority, I found that
the e�ects of segregation of Black and Asian people in the local area are not the same. The
level of commitment in collective political activities among individuals belonging to the white
majority remains una�ected by contexts of greater segregation of blacks. The expectation
of �nding more collective participation to protect the values, norms and interests of the
white majority in contexts of a large concentration of Black population is not backed up
by this data. In brief, the segregation of Black residents does not raise a perceived threat,
which in turn generates more political participation. The results can be explained partly by
the perception among white individuals of greater cultural integration of Afro-Caribbean:
they mostly speak English, have a high rate of intermarriage and are mainly Catholics.
Neither seems to be con�rmed the hypothesis that the concentration of blacks depressed
white individuals' chances to get involved in collective political activities as consequence of
a reduction of the interpersonal trust.
On the other side, the concentration of Asian population in the locality does not imply a
political reaction by white residents of the locality. Thus, we cannot con�rm the hypothesis of
political mobilization motivated by con�icting preferences or threats to the majority's values.
However, data seems to suggest quite the opposite. The geographical concentration of Asians
within the boundaries of the local authority tends to depress the collective participation of
individuals identi�ed as white. In the models, I consistently found that whites are less likely
to get involve in actions which require cooperation among residents. Thus, the hypothesis
that segregation intensi�es the perception of social distance between residents of a locality,
thereby reducing the likelihood of collective activities appears to �nd support in the results
21
obtained in the empirical analysis.
An explanation of these results may be that whites notice that the encapsulation of
those who conform the Asian group is a spatial representation of the socio-cultural di�er-
ences which inhibit a joint successful enterprise. As past research indicated, �cultural di�er-
ences prevent whites from engaging in the sorts of social interactions with Pakistanis and
Bangladeshis� (Bowyer 2009). Therefore, white individuals may be interpreting the Asian
bubbles as a disconnection from the broad society or as a rejection to get fully integrated
politically; this consequently may be dropping majority's perception of the existence of a
common ground for political action. Thus, living among culturally and linguistically distinct
South Asians not only can discourages contact with them but also the perception that the
local area has a share fate which may deserve a collective project. Current public discussion
on the existence of ghettos stimulate even more the �general perceptions not only of spatial
distance being correlated with social distance, but also of enclaves of di�erence�(Johnston,
Forrest et al. 2002).
VI. CONCLUSION
Since the lack of data was a shortcoming for a long time, scholars had fewer chances to
explore the relationship between local contexts and individual political non-electoral partici-
pation. To date, no other study has tested the e�ects of re�ne dimensions of the local ethnic
pro�le on collective forms of political participation. Using Citizenship Survey data matched
with aggregate data from English localities, I found that the ethnic minorities' residential
distribution within the local area does not constitute a signi�cant determinant of collective
political action among the ethnic minorities. In few words, there are neither positive nor
negative segregation e�ects.
Whites' living in districts with large ethnic minority segregation may perceive that they
pose a risk to their status and the traditional English norms and ways of life, but this
hostility may not be enough to engender more political collective actions. On the contrary,
there appear to be some ethnic minority segregation negative e�ects within the group of
whites. When examining the evenness with which population from minority and majority
groups are distributed across the geographic units that make up the local authority, I found
that only the segregation of Asian population is a signi�cant predictor of collective political
22
participation. Therefore, it can be rejected at least two initial expectations: on the one
hand, majority's political involvement does not increase as consequence of a perception of
threat; on the other hand, Asian and Black segregation a�ects does not a�ect equally. As
previous research found, whites react very di�erently to the concentration of each of the
largest English ethnic minority groups. (Bowyer 2009)
The segregation of Black origin individuals does not entail any consequences at the level
of political action among white individuals. The high social distance, or even the growing
Islamophobia among the whites, may constitute the roots of the reduced inter-personal social
trust which inhibit collective actions.
I conclude by outlining a couple of future avenues of research. First, it is required an
alternative research design to address the possible self-selection e�ects which are commonly
found in contextual research. In my study, the cross-sectional design may hinder more
complex dynamics which can be uniquely isolated if we use panel data. Second, further
research should delve on the possibilities of assessing the impact of other measures of spa-
tial patterning of ethnic distribution (i.e. index of isolation), placing individuals in smaller
geographical scale (for instance at Middle Layer Super Output Areas) and using more de-
tailed ethnic categories (Pakistani, Caribbean, etc.). The Candle report, prepared in the
aftermath of the summer's race riots in 2001, indicated that �separate educational arrange-
ments, community and voluntary bodies, employment, places of worship, language, social
and cultural networks, means that many communities operate on the basis of a series of
parallel lives� (Cantle 2001). Underlying the overall discussion on the current social issues
in Britain it usually appears the emergence of ethnic ghettos. It is generally assumed that
segregation have a pernicious consequences on the degree on ethnic minorities' political and
social engagement. The paper intended to give some more arguments to the discussion on
the consequences of segregation on political participation proposing that, as previous re-
search on electoral studies have found, �segregation is clearly not the problem it is perceived
to be�(Fieldhouse and Cutts 2008). If any, it concerns exclusively to the white majority's
political action.
23
References
Agnew, J. A. (1989). The devaluation of place in social science. The power of place: Bringing
together geographical and sociological imaginations. J. A. Agnew and J. S. Duncan. Boston,
Unwin Hyman: 9-29.
Alex-Assensoh, Y. (1997). "Race, concentrated poverty, social isolation, and political be-
havior." Urban A�airs Review 33(2): 209-227.
Anwar, M. (2001). "The participation of ethnic minorities in British politics." Journal of
Ethnic and Migration Studies 27(3): 533-549.
Anwar, M. (2010). The participation of Asians in the British political system. South
Asians overseas: migration and ethnicity. C. Clarke, C. Peach and S. Vertovec. New York,
Cambridge University Press.
Beyerlein, K. and M. Chaves (2003). "The political activities of religious congregations in
the United States." Journal for the Scienti�c Study of Religion 42(2): 229-246.
Blais, A., et al. (2004). "Where does turnout decline come from?" European Journal of
Political Research 43(2): 221-236.
Blalock, H. M. (1967). Toward a theory of minority-group relations. New York, Wiley.
Blumer, H. (1958). "Race prejudice as a sense of group position." The Paci�c Sociological
Review 1(1): 3-7.
Bowyer, B. T. (2009). "The contextual determinants of whites' racial attitudes in England."
British Journal of Political Science 39(03): 559-586.
Brady, H. E., et al. (1995). "Beyond SES: A resource model of political participation."
American Political Science Review: 271-294.
Budge, I. and D. W. Urwin (1966). Scottish political behavior: a case study in British
homogeneity, Longmans Green.
Cantle, T. (2001). "Community cohesion in Britain: a report of
the Independent Review team." Retrieved 3 September, 2012, from
http://dera.ioe.ac.uk/14146/1/communitycohesionreport.pdf.
24
Carmines, E. G. and R. Huckfeldt (1996). "Political behavior: An overview." A new
handbook of political science: 223-254.
Cho, W. K. T. (1999). "Naturalization, socialization, participation: Immigrants and (non-)
voting." Journal of Politics 61(4): 1140-1155.
Cho, W. K. T., et al. (2006). "Residential concentration, political socialization, and voter
turnout." Journal of Politics 68(1): 156-167.
Chong, D. (1991). Collective action and the civil rights movement, University of Chicago
Press.
Clarke, C. G., et al. (1984). Geography & ethnic pluralism, G. Allen & Unwin.
Cohen, C. J. and M. C. Dawson (1993). "Neighborhood poverty and African American
politics." American Political Science Review: 286-302.
Coleman, J. S. (1987). Norms as social capital. Economic imperialism: The economic
method applied outside the �eld of economics. G. Radnitzky and P. Bernholz. New York,
Paragon House: 133-155.
CRE, C. f. R. E. (1998). Stereotyping and Racism: Findings from Two Attitude Surveys.
London, Commission for Racial Equality.
Dahya, B. (1974). "The nature of Pakistani ethnicity in industrial cities in Britain." Urban
ethnicity 77: 118.
ElectoralCommision (2005). Understanding Electoral Registration: The Extent and Nature
of Non-Registration in Britain, Electoral Commission.
ElectoralCommission (2002). Voter Engagement Among Black and Minority Ethnic
Communities: Research Report. London, Electoral Commission.
Enos, R. D. (2010). The persistence of Racial Threat: evidence from the 2008 election.
American Political Science Association Annual Meeting, Washington, D.C.
Enos, R. D. (2011). "Racial Threat: Field-experimental evidence that a proximate racial
outgroup activates political participation." Working Paper, The Harris School, The Univer-
sity of Chicago, January 7: 1-50.
Fahmy, E. (2005). �Ethnicity, Citizenship and Political Participation in Britain,
ESRC/ODPM Postgraduate Research Programme�, Working paper, Swindon, 1 January.
25
Fieldhouse, E. and D. Cutts (2008). "Diversity, density and turnout: The e�ect of neigh-
bourhood ethno-religious composition on voter turnout in Britain." Political Geography
27(5): 530-548.
Fieldhouse, E. and D. Cutts (2008). "Mobilisation or marginalisation? Neighbourhood
e�ects on Muslim electoral registration in Britain in 2001." Political Studies 56(2): 333-354.
Giddens, A. (1984). The constitution of society: Outline of the theory of structuration,
Univ of California Press.
Giles, M. W. and M. K. Dantico (1982). "Political participation and neighborhood social
context revisited." American Journal of Political Science: 144-150.
Greasley, S. and P. John (2011). "Does Stronger Political Leadership Have a Performance
Payo�? Citizen Satisfaction in the Reform of Subcentral Governments in England." Journal
of Public Administration Research and Theory 21(2): 239-256.
Gui, B. (1987). "Eléments pour une dé�nition d'�économie communautaire��." Notes et
documents 19(20): 32-42.
Harrison, M. and D. Phillips (2003). Housing and Black and minority ethnic communities,
Review of the evidence base. London, O�ce of the Deputy Prime Minister.
Hedman, L., et al. (2011). "Neighbourhood choice and neighbourhood reproduction."
Environment and Planning-Part A 43(6): 1381.
Hickman, P. and D. Robinson (2006). "Transforming Social Housing: Taking Stock of New
Complexities." Housing Studies 21(2): 157-170.
Hill, K. Q. and J. E. Leighley (1999). "Racial diversity, voter turnout, and mobilizing
institutions in the United States." American Politics Research 27(3): 275-295.
Hopkins, D. (2011). �National Debates, Local Responses: The Origins of Local Concern
about Immigration in Britain and the United States, British Journal of Political Science,
British Journal of Political Science 41 (03): 499-524.
Huckfeldt, R. (2001). "The Social Communication of Political Expertise." American Journal
of Political Science 45(2): 425-438.
Huckfeldt, R. R. (1979). "Political participation and the neighborhood social context."
American Journal of Political Science: 579-592.
26
Huckfeldt, R. R. (1983). "Social contexts, social networks, and urban neighborhoods: En-
vironmental constraints on friendship choice." American Journal of Sociology: 651-669.
Huckfeldt, R. R. (1986). Politics in context: Assimilation and con�ict in urban
neighborhoods, Agathon Pr.
Huckfeldt, R. R. and J. D. Sprague (1995). Citizens, politics, and social communication:
Information and in�uence in an election campaign, Cambridge Univ Pr.
Ireland, P. (2008). "Comparing responses to ethnic segregation in urban Europe." Urban
Studies 45(7): 1333-1358.
Jakubs, J. F. (1981). "A distance-based segregation index." Socio-economic planning sci-
ences 15(3): 129-136.
John, P., et al. (2011). "How Civic is the Civic Culture? Explaining Community Partici-
pation Using the 2005 English Citizenship Survey." Political Studies 59(2): 230-252.
Johnston, R., et al. (2002). "Are there ethnic enclaves/ghettos in English cities?" Urban
Studies 39(4): 591-618.
Just, A. and C. Anderson (2012). "Immigrants, Citizenship and Political Action in Europe."
British Journal of Political Science 1(1): 1-29.
Key, V. (1949). Southern politics in state and nation. New York, Vintage Books
Kirby, A. (1989). "A sense of place." Critical Studies in Mass Communication 6(3): 322-326.
Larcinese, V. (2007). "Does political knowledge increase turnout? Evidence from the 1997
British general election." Public Choice 131(3): 387-411.
Leighley, J. E. (2001). Strength in numbers?: The political mobilization of racial and ethnic
minorities, Princeton Univ Pr.
Leighley, J. E. and J. Nagler (1992). "Socioeconomic class bias in turnout, 1964-1988: The
voters remain the same." The American Political Science Review: 725-736.
Lowndes, V., et al. (2002). Social capital and political participation: how do local insti-
tutions constrain or enable the mobilisation of social capital? Paper presented at Social
Capital Seminar. St John's College, University of Cambridge.
Mason, D. (2000). Race and ethnicity in modern Britain, Oxford University Press Oxford.
27
Massey, D. S. and N. A. Denton (1988). "The dimensions of residential segregation." Social
forces 67(2): 281-315.
McClurg, S. D. (2003). "Social networks and political participation: The role of social
interaction in explaining political participation." Political Research Quarterly 56(4): 449-
464.
Miles, R. and A. Phizacklea (1984). White man's country: Racism in British politics, Pluto
Press London.
Miller, A. H., et al. (1981). "Group consciousness and political participation." American
Journal of Political Science: 494-511.
Miller, B. A. (2000). Geography and social movements: Comparing antinuclear activism in
the Boston area, Univ Of Minnesota Press.
NeighbourhoodRenewalUnit (2002). "The National Strategy for Neighbour-
hood Renewal. Fact Sheet No 2." Retrieved 3 September, 2012, from
http://www.creatingexcellence.org.uk/uploads/factsheet2_neighbourstrategy.pdf.
Nie, N. H., et al. (1996). Education and democratic citizenship in America, University of
Chicago Press.
Parry, G., et al. (1992). Political participation and democracy in Britain, Cambridge Univ
Pr.
Pattie, C. J., et al. (2004). Citizenship in Britain: values, participation and democracy,
Cambridge Univ Pr.
Pawson, H. and D. Watkins (2007). "Quasi-marketising access to social housing in Britain:
assessing the distributional impacts." Journal of Housing and the Built Environment 22(2):
149-175.
Peach, C. (1996). "Does Britain have ghettos?" Transactions of the Institute of British
Geographers: 216-235.
Peach, C. (1998). "South Asian and Caribbean ethnic minority housing choice in Britain."
Urban studies 35(10): 1657.
Peach, C. (2009). "Slippery segregation: discovering or manufacturing ghettos?" Journal of
Ethnic and Migration Studies 35(9): 1381-1395.
28
Phillips, D. (1998). "Black minority ethnic concentration, segregation and dispersal in
Britain." Urban studies 35(10): 1681-1702.
Phillips, T. (22 September 2005). "After 7/7: Sleepwalking to segregation."
Website of Commission of Racial Equality. Retrieved 23 August 2012, from
http://www.cre.gov.uk/Default.aspx.LocID-0hgnew07s.RefLocID-0hg00900c002.Lang-EN.htm.
Phillips, T. (2005). "After 7/7: Sleepwalking to segregation. Speech given at the Manch-
ester Council for Community Relations, 22nd September 2005." Retrieved June, 2012, from
http://www.humanities.manchester.ac.uk/socialchange/research/social-change/summer-workshops/documents/sleepwalking.pdf.
Ramakrishnan, S. K. and T. J. Espenshade (2001). "Immigrant Incorporation and Political
Participation in the United States1." International Migration Review 35(3): 870-909.
Rex, J. (1981). Urban segregation and inner city policy in Great Britain. Ethnic segregation
in cities. C Peach, C. Robinson and S. Smith. London, Croom Helm: 25-42.
Rex, J., et al. (1979). Colonial immigrants in a British city: a class analysis, Routledge &
Kegan Paul London.
Saggar, S. (2000). Race and representation: electoral politics and ethnic pluralism in Britain,
Manchester Univ Pr.
Schlichting, K., et al. (1998). "Racial segregation and voter turnout in urban America."
American Politics Research 26(2): 218-236.
Shaw, D., et al. (2000). "Examining Latino turnout in 1996: A three-state, validated survey
approach." American Journal of Political Science: 338-346.
Simpson, L. (2004). "Statistics of racial segregation: measures, evidence and policy." Urban
studies 41(3): 661.
Simpson, L. (2007). "Ghettos of the mind: the empirical behavior of indices of segregation
and diversity." Journal of the Royal Statistical Society: Series A (Statistics in Society)
170(2): 405-424.
Sobolewska, M., et al. Religion and minority political participa-
tion: evidence from Great Britain, Retrieved 23 August 2012, from
http://www.ecprnet.eu/MyECPR/proposals/reykjavik/uploads/papers/711.pdf.
29
Sta�ord, M., et al. (2010). Racial discrimination and health: exploring the possible protec-
tive e�ects of ethnic density. Ethnicity and Integration. Understanding Population Trends
and Processes. J. Stillwell and M. van Ham. Dordrecht, Springer. 3.
Statham, P. (1999). "Political mobilisation by minorities in Britain: Negative feedback of
`race relations'?" Journal of Ethnic and Migration Studies 25(4): 597-626.
Tinsley, J. and M. Jacobs (2006). "Deprivation and ethnicity in England: a regional per-
spective." Regional trends 39(6).
Uhlaner, C. J. (1989). ""Relational Goods" and Participation: Incorporating Sociability
into a Theory of Rational Action." Public Choice 62(3): 253-285.
Uhlaner, C. J., et al. (1989). "Political participation of ethnic minorities in the 1980s."
Political Behavior 11(3): 195-231.
Uslaner, E. M. (2002). The moral foundations of trust. Cambridge, Cambridge Univ Pr.
Uslaner, E. M. (2008). Trust as a moral value. The handbook of social capital. D. Cas-
tiglione, J. W. Van Deth and G. Wolleb. New York, Oxford University Press, USA: 101-121.
Uslaner, E. M. (2011). "Trust, Diversity, and Segregation in the United States and the
United Kingdom1." Comparative Sociology 10(2): 221-247.
van Ham, M. and D. Manley (2012). "Segregation, Choice Based Letting and Social Housing:
How Housing Policy Can A�ect the Segregation Process." Institute for the Study of Labor,
Discussion Paper Series (IZA DP No. 6372).
Verba, S. and N. H. Nie (1987). Participation in America: Political democracy and social
equality, University of Chicago Press.
Voas, D. and P. Williamson (2000). "The Scale of Dissimilarity: Concepts, Measurement
and an Application to Socio-Economic Variation Across England and Wales." Transactions
of the Institute of British Geographers 25(4): 465-481.
Winstanley, A., et al. (2002). "Moving house, creating home: Exploring residential mobil-
ity." Housing Studies 17(6): 813-832.
30
Wol�nger, R. E. and S. J. Rosenstone (1980). Who votes? New Haven, Yale Univ Pr.
[1] The dispute emerged as consequence of a speech delivered by T. Phillips
at the Annual Conference of the Royal Geographical Society. See Phillips,
T. (22 September 2005). "After 7/7: Sleepwalking to segregation." Web-
site of Commission of Racial Equality. Retrieved 23 August 2012, from
http://www.cre.gov.uk/Default.aspx.LocID-0hgnew07s.RefLocID-0hg00900c002.Lang-EN.htm.
[2] Due to reasons of con�dentiality many surveys avoid giving openly information of the location
of the respondents. However, under a special license, the National Centre for Social Research
(NatCen) provided me the LAs' codes identi�ers current as at April 2009. On 1 April 2009, there
was a reorganisation of local government that created nine new unitary authorities (UAs) which
are aggregates of existing local authorities. Since the geographical identi�ers given by NatCen
do not match to the data at aggregate level collected before that year, it was not possible to
include in the �nal dataset the respondents living in old local authorities.
[3] In this paper, I refer to concentration and segregation interchangeably because by de�nition,
segregation implies spatial concentration.
[4] The Index of Multiple Deprivation is an o�cial statistical measure of deprivation for small areas
in England which combines a number of indicators (income, employment, health and disability,
education, skills and training, barriers to housing and services, living environment and crime).
[5] In fact, Simpson (2007) showed with Census data that white population is not moving from
highly non-white areas.
31
