The paper was presented at the scientific-professional conference "Family – a Challenge for Social Policy and Helping Professions" held on 21 March 2014 at the premises of the St. Elizabeth University of Health and Social Work in Žilina. Introduction to the issue of residential segregation
Research into the phenomenon of residential segregation has received considerable attention abroad for several decades. It is a phenomenon occurring in various cultures, including our own. Segregation can be perceived on a value-neutral level, where it refers to determining the degree of distribution of diverse social groups in society within a defined geographic area. On a negative level, we speak of residential segregation when the actual degree of segregation of, for example, an ethnic group shows a significantly elevated value, which represents a societal problem often overlapping with other social or pathological problems. In industrially developed countries, segregation is defined in several basic categories, primarily based on ethnicity, income, social status, religion, or gender. (Blackburn, 2009)
Residential segregation describes the extent to which members of two or more (ethnic) population groups live separately in different urban territorial units. Ethnic groups may be distributed among the territorial units of cities and municipalities in such a way that in some areas they are represented to a significant degree compared to localities with minimal or no representation. (Massey, Denton In Saijeva, 2011) All forms of segregation have a direct impact on reducing the quality of life, well-being, and "life chances" of families, households, or individuals, particularly in the areas of housing, education, and employment. Areas characterized by a significant degree of residential segregation and isolation are associated with increased crime rates, deviant behavior, low educational attainment, unemployment, deteriorating housing stock, overcrowded dwellings, and low availability of basic civic amenities. However, this applies only in the case of involuntary, forced segregation arising rather as a consequence of discrimination.
Involuntary residential segregation represents a situation where housing conditions become unsatisfactory for certain groups of residents, but financial, institutional, or other barriers prevent a change of housing. (Schwab In Burjanek, 1997) In addition to the aforementioned segregation, we also distinguish voluntary separation, where selected groups of residents separate themselves from the rest of society based on their own decision, often motivated by their own interests and activities aimed at minimizing contact with their surroundings. (Toušek, 2007)
Settlement segregation is defined either by the economic status of households (vertical differentiation) or by ethnicity, household type, cultural characteristics, and household lifestyle (horizontal differentiation). Based on horizontal differentiation, we distinguish settlement types such as ethnic enclave (voluntary separation) and ghetto (segregation). The lower the economic status of the residents, the more segregated settlement moves into the category of slum, characterized by low economic status or poverty and segregation. At the opposite end of the imaginary axis lies the settlement type of citadel (gated communities), whose residents exhibit high economic status associated with voluntary separation. For these settlement types, the primary characteristic is the economic status of their members, while segregation/separation is a secondary characteristic conditioned by poverty or wealth and power of the respective group of residents. (Sýkora - Temelová, 2005)
In the conditions of the Slovak Republic, residential segregation of Roma households in rural settlements and slums (so-called "Roma settlements") currently appears to be the most "acute," while the issue of residential segregation in the urban environment remains in the background. Residential segregation in cities is often associated with discrimination, financial unaffordability of housing outside segregated areas, which in many cases is closely linked to households' reduced ability to maintain rental housing. State housing policy focuses primarily on creating legislative conditions for the development of social rental housing. The main actors in housing policy are municipalities and cities. Their primary areas of activity include creating spatial conditions for housing development in their territory, establishing a conceptual framework for the development of various areas of the municipality, cooperating in creating suitable housing conditions, securing land for housing construction, and improving the management of the housing stock. (MDVRR SR, 2012) Financial accessibility and sustainability of housing for vulnerable groups is often an overlooked problem of social and housing policy of local governments. Cities and municipalities often do not use available legislative tools aimed at eliminating residential segregation for low-income families and families living in poverty (often generationally reproduced). This mainly concerns support for securing the housing allowance, which, although its amount does not cover the actual costs associated with housing, at least creates a (minimal) possibility of maintaining permanent housing. In practice, this means that for certain groups of residents, social housing is currently inaccessible, even in the case of lower-standard dwellings. Residential segregation in cities and municipalities is a frequently overlooked problem and, moreover, in many cases is directly supported by non-conceptual (or intentional segregationist) housing and social policy. Many cities in the ŽA region, e.g.: Liptovský Mikuláš (2000, 2005, 2013, 2014), Martin (2007), Ružomberok (2012), Žilina (2006, 2009 to 2010) have in the past carried out/are carrying out the construction of lower-standard social housing precisely in residentially segregated localities, thereby supporting their continuous expansion and deepening of social problems of residents in the given area.
Measurement of residential segregation is carried out by measuring a particular dimension, with more than 20 different indices available. The first dimension is evenness, which captures the distribution of two or more ethnic groups within a certain geographic area. The second dimension is exposure, through which we capture the potential, the probability of contact between individual population groups. Concentration, as the third dimension, relates to capturing the size of the physical space within a geographic area occupied by individual population groups. The penultimate dimension is centralization, which indicates the extent to which specific population groups live in territorial units located near the city center. The last dimension is the measurement of clustering, which captures the extent to which members of a particular population group are dispersed across surrounding territorial units. (Massey - Denton In Saijeva, 2011)
The paper captures the degree of residential segregation in the urban environment in two dimensions: evenness and exposure. The degree of evenness is measured using the index of dissimilarity and the index of segregation. To determine the degree of exposure, we used the isolation index.
The most frequently used index for determining the degree of residential segregation is considered to be the index of dissimilarity (IDxy). The index compares the distribution, the proportional difference between two different categories of persons in certain geographic, territorial units, whereby the sum of these persons in individual groups is not the sum of the total city population. If we want to determine the degree of dissimilarity for all ethnic groups represented in the total population, we must repeat the calculation for every possible pair combination of the respective ethnic groups forming that population. If the total population consists of only two ethnic groups, the index will show the same values as the segregation index (ISx). The index of dissimilarity represents the proportion of the population of a given ethnic group that would have to "move" in order to achieve an even spatial distribution across the individual territorial units of the city. (Blackburn, 2009) The index of dissimilarity is calculated (Duncan & Duncan, 1955a):
Where: n - number of territorial units, xj - number of individuals x in territorial unit j, X - total number of individuals of type x across all territorial units, yj - number of individuals y in territorial unit j, Y - total number of individuals of type y across all territorial units.
An alternative to the index of dissimilarity is the segregation index (ISx), which compares the distribution or concentration of population groups relative to the rest of the total population as a whole (the remainder to 100% of the municipality's population). The index indicates what proportion of the population must be redistributed among the other territorial units so that the distribution of, for example, an ethnic group is spatially even across the entire city territory. The index was used in studying the segregation of Asians in Great Britain or African Americans in the USA. The segregation index is calculated (Gorard, 2000):
Where: n - number of territorial units, xj - number of individuals x in territorial unit j, X - total number of individuals of type x across all territorial units, ti - total number of individuals in territorial unit j, T - total number of individuals across all territorial units.
The last index used, capturing the average local concentration of an ethnic group, is the isolation index (P*). The index indicates the degree of possible exposure, contact between members of the same population group, or the extent to which members of a given ethnic group are reliant on contact and interaction exclusively among themselves. The higher the index value, the higher the probability of isolation of a population group in space, meaning that a member of an ethnic group will share a territorial unit with another member of the same ethnic group. (Echenique - Fryer, 2006) The isolation index is calculated (Lieberson, 1963):
Where: n - number of territorial units, xj - number of individuals x in territorial unit j, X - total number of individuals of type x across all territorial units, T - total number of individuals in territorial unit j. Index values range from 0 to 1, or from 0 to 100%, whereby an index value ≥ 0.6 (60%) is considered a very high index level; if the index value is < 0.6 (60%) and at the same time > 0.3 (30%), it is a moderate index level; an index value ≤ 0.3 (30%) is considered a low index level. For the isolation index, the higher the index value, the higher the probability of isolation for that ethnic group.
The indices used also have their specificities concerning a certain distortion of their results. While the index of dissimilarity (IDxy) and the segregation index (ISx) are independent of the relative size of individual population groups, the isolation index (P*) depends on the size of the compared group. The values of the indices used can also be affected by the internal division of territorial units in the studied cities. The division of territorial units in a city/municipality can significantly influence the measurement of residential segregation to a segregation value of 0 (example A) or the index may show a value of 1, i.e., complete segregation (example B). (Fryer In Saijeva, 2011)
Selection of data used in the analysis of residential segregationThe availability of data concerning national minorities living in the territory of the Slovak Republic is considerably limited. The existence of data on the Roma national minority can essentially be divided into two basic sources.
The first source is data from the Statistical Office of the Slovak Republic, which collects data related to national minorities within the SR, primarily through the Population and Housing Census (SODB), which is part of the worldwide program of population and housing censuses. The census is repeated regularly with a 10-year periodicity (measurements are comparable over time), and it is the only comprehensive count of all inhabitants and households in the SR, except for foreigners enjoying diplomatic privileges and immunity. The advantage of the census is that it captures several indicators concerning families, individuals, as well as dwellings and houses simultaneously. For the purposes of our analysis, data relating to the declared nationality of respondents as well as to their place of residence – the territorial units of the city/municipality – are essential. Data for SODB 1991, 2001 and 2011 were requested from the databases of the Statistical Office of the SR from the Bratislava and Žilina workplaces (Tab. 1).
For the purposes of the census, nationality is understood as affiliation with a nation or ethnic group, based on one's own decision, whereby neither the mother tongue nor the language used or spoken by the resident was decisive, but rather the resident's own decision about belonging to a particular nation, national minority, or ethnic group. The nationality of children under 15 is identified according to the nationality of the parents; in the case of different nationalities of the parents, the nationality is determined according to one of them. (Juhaščíková – Škápik – Štukovská, 2012)
Tab. 1: Roma nationality SODB 1991, 2001 and 2011 in selected cities of the ŽA region
Source: ŠÚ SR, 2014
In the Žilina Region, according to data from the Statistical Office of the SR, 2,264 persons declared Roma nationality in 2011, representing a decrease compared to previous censuses (SODB 1991 – 2,590 persons, SODB 2001 – 2,795 persons), while from the total population they constitute just over 0.3% (SODB 1991 – 0.4%, SODB 2001 – 0.4%). Overall, among all regions of the SR, the Žilina Region (2011) is the third with the lowest percentage share of persons declaring Roma nationality.
A different number of inhabitants of Roma nationality is identified by the Atlas of Roma Communities 2004 and 2014 (Tab. 2). Roma settlement or increased concentration of Roma is identified on the basis of anthropological features, cultural affiliation, way of life – lifestyle, living space subjectively perceived by the majority as features attributed to Roma regardless of their positive or negative content. Data from the Atlas of Roma Communities are not suitable for determining the degree of residential segregation. There is a lack of spatial distribution of members of the Roma minority within the city territory (it does not capture households or individuals at the level of settlement units), including both segregated and integrated Roma populations. The very attribution of "Roma ethnicity" by the majority part of the population to persons based on external or visible features and behavioral manifestations, which in many cases may only lead to confirmation of stereotypes towards Roma, also appears problematic. In determining the number of members of the Roma nationality, data collection relied heavily on qualified estimates by local government employees, which could have led to an artificial and purposeful overestimation of the number of Roma in the city in order to gain the possibility of applying for financial resources from funds designated for inclusive policy and related measures.
Tab. 2: Roma nationality Atlas of Roma Communities in selected cities of the ŽA region
Source: Hurle – Mušinka – Škobla, 2014 * Data collection was not carried out in the city of Žilina in 2004
For the purpose of determining the degree of residential segregation using selected indicators, we used data from the census of the Statistical Office of the SR 2001 and 2011. From the census data conducted in 1991, it is not possible to determine the distribution of persons declaring Roma nationality in individual settlement units because the numbers of persons for each nationality are given for the city as a whole. The censuses of 2001 and 2011 capture the nationality of residents of cities and municipalities in individual settlement units.
The percentage representation of Roma in selected cities of the Žilina Region in 2013, according to published data from the Atlas of Roma Communities, ranged from 0.1% to 10.0%. Out of the total number of municipalities (315), Roma – that is, those to whom this ethnicity was attributed – live in 27 municipalities and cities. This is the lowest proportion of municipalities (2%) with a Roma national minority within the SR. (First Results of the ATLAS, 2013)
The selection of cities for the purpose of the residential segregation analysis was conditioned by the number of persons declaring Roma nationality. Given that the available data sources concerning the ethnic composition of cities in the Žilina Region have shortcomings in fully capturing relevant data for individual ethnic groups, we conditioned the selection of cities on the size of the group of persons declaring Roma nationality. We included in the analysis only those cities where in 2001 the Roma national minority was the largest (Liptovský Mikuláš, Ružomberok), or the second largest ethnic group in the city (Čadca, Dolný Kubín, Kysucké Nové Mesto, Martin, Žilina). In the remaining cities, Roma nationality was represented minimally or not represented by a single person according to the SODB methodology.
Degree of residential segregation in selected cities of the Žilina Region
In selected cities of the Žilina Region in 2011, the indices determining the degree of residential segregation for persons declaring Roma nationality ranged in the band of elevated moderate segregation and very high segregation.
Index values in selected cities of the ŽA region in 2011
The average value of the index of dissimilarity in selected cities of the Žilina Region for 2001 was 63.12% (median 62.08%), while in 2011 it reached a value of 59.9% (median 56.93%). In three cities of the ŽA region in 2011, IDxy was in the very high segregation band. In the other cities, it ranged in the upper part of the moderate segregation band. The highest index value was in the city of Žilina (2011), where more than 79% of Roma would have to move among inhabitants declaring Slovak nationality in order for these nationality groups to be evenly distributed among the settlement units in the city. The index of dissimilarity shows an elevated value by 15.09% compared to 2001 (64.32%). For comparison, the index of dissimilarity in 2011 showed the degree of residential segregation in the city of Žilina for the most represented national minority in the city (Czech nationality) at a value of 18.65%. The second highest index of dissimilarity for the Roma minority is shown by the city of Ružomberok (66.47%), which increased by 4.39% compared to 2001. Conversely, the lowest values are in the cities of Kysucké Nové Mesto (48.65%) and Dolný Kubín (51.13%). In both cities, the most significant reduction in the index rate compared to 2001 was recorded – by 35.11% (Kysucké Nové Mesto) and by 19.67% (Dolný Kubín); however, in both cities the index is significantly above the threshold of moderate segregation (≤ 30%).
Tab. 3: Index of dissimilarity (IDxy) in selected cities of the ŽA region
The average value of the segregation index in the Žilina Region represented a value of 62.13% (median 61.85%) in 2001 and 58.58% (median 56.78%) in 2011, representing values close to or exceeding the level of very high segregation. The segregation index shows similar values to the index of dissimilarity, but its values represent the degree of segregation, i.e., what percentage of citizens declaring Roma nationality must be redistributed among the total city population so that their distribution is even across the entire city territory within all settlement units. The highest index values were recorded in 2011 in the cities of Žilina (79.4%), Ružomberok (65.29%).
Tab. 4: Segregation index (ISx) in selected cities of the ŽA region
The isolation index in selected cities of the Žilina Region reached on average low values – 6.33% (median 2.64%) in 2001 and 9.94% (median 1.04%) in 2011. An exception is the city of Žilina, where the index reached a value of 33.88%, indicating almost a 34% probability that inhabitants declaring Roma nationality in 2011 were reliant on contact and interaction exclusively with citizens declaring the same nationality. The index increased significantly compared to 2001, when it reached a negligible value (2.64%). The second highest index rate was found among Roma living in the city of Liptovský Mikuláš, where in 2011 there was a 27.42% probability that Roma would be reliant only on ties and contact with members of the same ethnic group (in 2001 it was 27.39%).
Tab. 5: Isolation index (P*) in selected cities of the ŽA region
Conclusion
The issue of residential segregation of Roma in the conditions of Slovak cities is among the greatest challenges of the present. A considerable part of the effort necessary to change the current situation requires a change in approach both at the level of state housing policy and at the level of local government. Increased attention needs to be paid to supporting inclusive tools such as: microloans, regulated rent, building acceptable social infrastructure, and the application of transitional and inclusive housing. Financial resources must be directed not only towards the desegregation of rural slums (Roma settlements), but also towards urban concentrations showing elevated or high levels of residential segregation.
Unfortunately, the currently ongoing construction of lower-standard social housing does not solve the issue of residential segregation; rather the opposite. Essentially, it is only a change of dwelling without a fundamental change of "neighbor." Residents of these still residentially segregated areas have been left on their own to address key issues such as employment and accessibility of services. Residentially segregated households suffer from long-term unemployment and do not have incomes through which they could secure permanent housing (they are particularly threatened by non-payment) or make significant investments in housing quality. The level of their income in many cases reduces the possibility of obtaining housing outside the segregated area to nearly zero.
Municipalities and cities, through adopted strategic documents such as Urban Plans of Cities and Municipalities, Community Plans of Social Services, or Programs of Economic and Social Development, have the opportunity to comprehensively plan, coordinate, and eliminate the expansion of residentially segregated areas and thereby negate their impacts on residents.
Author: PhDr. Mgr. Martin Brňak, PhD.* *List of references
[1] BLACKBURN, M. R. 2009. Measuring Occupational Segregation and its Dimensions of Inequality and Difference. [online]. [2014-01-15]. Available at: http://www.sociology.cam.ac.uk/srg/workingpapers/CS12MeasuringSegregationpdf [2] BURJANEK, A. Segregace. In Sociologický časopis. Praha: Sociologický ústav AV ČR. ISSN 0038-0288. 1997, vol. 33, no. 4, pp. 423-443 [3] ECHENIQUE, F. - FRYER, G. R. 2006. A measure of segregation based on social interactions In Journal of Economics. New York: Springer Science. ISSN 1617-7134, 2007, pp. 441-485 [4] HURLE, J. – MUŠINKA, A. – ŠKOBLA, D. 2014. Atlas rómskych komunít 2014. Bratislava: UNDP, 2014 [5] JUHAŠČÍKOVÁ, I. – ŠKÁPIK, P. – ŠTUKOVSKÁ, Z. 2012. Základné údaje zo sčítania obyvateľov, domov a bytov 2011. Bratislava: ŠÚ SR, 2011. 92 p. ISBN 978-80-8121-206-2 [6] MINISTERSTVO DOPRAVY, VÝSTAVBY A REGIONÁLNEHO ROZVOJA SR. 2012. Správa o plnení zámerov Koncepcie štátnej bytovej politiky do roku 2015. Bratislava: MDVRR, 2012. 39 p. Without ISBN [7] PRVÉ VÝSLEDKY ATLASU RÓMSKYCH KOMUNÍT 2013. [online]. [2014-01-13]. Available at: http://www.minv.sk/?atlas_2013 [8] SAIJEVA, H. 2011. Segregation and Employment in Swedish Regions. Jönköping: Jönköping University, 2011. 42 p. Without ISBN [9] SÝKORA, L. - TEMELOVÁ, J. 2005. Prevence prostorové segregace. Praha: Univerzita Karlova v Prahe, 2005. 119 p. Without ISBN [10] TOUŠEK, L. 2007. Sociální vyloučení a prostorová segregace. [online]. [2014-01-12]. Available at: http://www.caat.cz/publikace/44-prehledove-studie/143-socialni-vyloueni-a-prostorova-segregace.