Analysis of the desegregation and social inclusion policies in the Spanish context

In Spain, many experiences in slum clearance have been focused solely on eliminating the most visible aspects of the problem, creating social dissatisfaction. This article analyzes the extent to which new policies of relocation and dispersion recently applied in the slum areas of northwest Spain, foster favorable attitudes towards social inclusion of the displaced persons, who are mostly Roma. A questionnaire was applied in order to capture the perception of both those affected by relocations and those who receive the displaced persons in their neighborhoods. By quantitative, exploratory data analysis, we have detected a good social predisposition towards social inclusion of underprivileged people in the process of adaptation to live in houses with appropriate infrastructure, identifying those aspects needed to strengthen these kinds of interventions.


INTRODUCTION
The existence of shanty towns in cities is a major public issue due to the situation of poverty and abandonment of its inhabitants (United Nations, 2006:4).On one hand, these people cannot enjoy their internationally recognized right to an adequate standard of living and decent housing and on the other hand, they find themselves in a situation of "multiple deprivations" (Clarke, 2006:1).This is due to their social exclusion in different areas, either employment, education or health, or spatial and relational segregation.It should be added that most Spanish people living in shanty towns are members of the Roma ethnic group.This ethnic minority still suffers serious problems of discrimination and rejection by most of society.
Facing this severe problem, one of the most utilized public policies in Spain are those focusing on the relocation of people living in shanties, for subsequent demolition of the shanty towns.These people are provided with access to standardized housing, normally under subsidized rental status.At the same time, new houses are located favoring the spatial dispersion of the members of the group who lived in the shanty town, to improve their social integration.However, it is not clear that access to a decent house and de-segregation by themselves will improve the levels of social exclusion suffered by these people (Somerville, 1998:773;Bolt, Ozuekren and Phillips, 2010:170).For this reason, it becomes necessary to analyze to what extent these types of public policies lead to the social inclusion of inhabitants of shanty towns.
In addition, in many cases, public authorities address these policies as short-term goals to solve urban planning and urban development problems, instead of managing the long-term well-being of the shanty towns inhabitants.These types of "quick" actions carried out in recent times, have had dire consequences for this population.Not only have they avoided the improvement of their situation, but in fact have made it worse, despite the huge public economic effort.When the goal is simply a "clean" slum area of the city, due to the bad image they produce, then institutions will probably apply inadequate public policies.So, instead of solving the housing problems of slum dwellers, an increased social exclusion could arise.These actions solve the symptoms of the problem in the short term, but in the medium-and long-term, the problem could get worse or even become chronic, while the same policy is persistently implemented (Leal Maldonado, 2002:64).Because of this, the distrust of people affected towards any type of public intervention in the shanty towns has increased.This is now one of the greater obstacles to deal with.Positive attitudes, cooperation and consensus are essential for the success of these types of actions, in line with those recommended by Pressman and Wildavsky (1973:213).
Many authors address, from different points of view, the problem of the persistence of shanty towns in cities.Their work is aimed at analyzing the causes of this problem, fostering actions whose main objective is access to decent housing, while at the same time, boosting social cohesion.This allows people to have the same opportunities and the enough resources to take part completely in political, economic, social, educational, labor and cultural life, enjoying living conditions equal to others.These studies reach a special relevance due to the need to incorporate all interactions between race, ethnicity, religion and housing into urban development policies (Flint, 2009:428).
Over the last few years, a belief has been spread: ethnic concentration in cities, especially if coupled with a lack of resources, ends up becoming a focus of social disturbances and problems, which in turn hinder social cohesion.This argument focuses on the fact that residential segregation tends to limit the social integration of ethnic minorities, since it restricts the opportunities of its residents.People living in segregated neighborhoods are, for example, less likely to find a job, which favors a downward spiral through which the gradual increase in social isolation is reinforcing the marginalization of the labor market by reducing contacts that can provide information on employment opportunities (Gallie, Paugam and Jacobs, 2003:3).An erosion and attraction effect are also generated in these urban centers.The concentration of low-income inhabitants leads to an erosion of public facilities and services in the area, which is followed by residential abandonment and increased crime, which then causes an attraction effect of inhabitants in very precarious conditions to the area (Friedrichs, Galster and Musterd, 2003:798).Therefore, these poorly maintained urban environments are contributing to stigmatization of the area (Kearns and Parkinson, 2001Parkinson, :2105)).In addition, when there is racial discrimination and prejudice in the housing market, these vicious circles are reinforced (Massey, 1990:329).In this way, "a city of corners" is being developed.Marginal groups are being cornered increasingly away from the central space of coexistence and urban integration (Cartelle Fernández, Fernández de Sanmamed Sanpedro and Guillén Gestoso, 2003).
Given this situation, most of the public policies aimed at facilitating access to housing for people belonging to ethnic minorities, have residential dispersion as their goal, to favor their integration in the host society.This is the point of view officially promoted by the European institutions.The European regulation includes the need to implement integrated housing actions in support of marginalized groups, as well as all the necessary interventions to eliminate segregation (Parlamento Europeo y Consejo, 2010).Due to this, a series of requirements are established to obtain funds from the European Regional Development Fund (ERDF).Firstly, investment in housing should be accompanied by actions in the areas of education, health, social inclusion and employment.Secondly, the geographical situation of housing should ensure the spatial integration of marginalized groups, without contributing to segregation, isolation or exclusion (Comisión Europea, 2010).For its part, the European Council specifically recommends the promotion of spatial desegregation to ensure equality in access to housing for the Roma population (Consejo Europeo, 2013).
Yet there is no collective agreement on the relationship between spatial segregation and social inequality (Musterd, 2005:342-344).In countries with strong differences between the richest and the poorest urban population groups, as is the case of São Paulo in Brazil (Villaça, 2011:37), segregation is conceived as an important manifestation of social inequality.However, in other contexts such as Europe, the segregation levels tend to be more moderate and do not prevent contact with the rest of the urban population.In fact, urban segregated population groups in the city can show very different levels of integration in the labor market and in education (Musterd, 2005:342-344).The study conducted by Peach (1999:344) on the segregation models presented by the cities of London and New York show that, in both cases and despite their different ethnic geographies, socioeconomic class seems to provide little explanation for the degree of segregation.
This lack of evidence, together with the desire to obtain "quick" results by the public authorities, raises the concern that involuntary population mix policies are being used as a "shortcut" to control ethnic minorities (Harrison, Law and Phillips, 2006:3).In fact, segregation can be interpreted as the way that people from ethnic minorities get support, security and solidarity within their group, developing the so-called social capital of union (Bolt, Ozuekren and Phillips, 2010:170).Within this concept, dispersion policies would not lead to greater integration, but would lead to greater isolation of social networks that gave them access to different resources (Arin, 1991:208-210).
Given this disagreement, the causes of the segregation of the Roma minority in Spain should be analyzed.For this, the two-pronged process is used, in accordance with the Common Agenda for Integration (Commission of the European Communities, 2005:5).The relationship between integration and segregation should be a two-way relationship, involving both ethnic minority groups and the host society (Bolt, Ozuekren and Phillips, 2010:171).On the part of Roma people, it is necessary to consider their strong sense of belonging to the group.Roma people always put the welfare of the group before their individual welfare.Being a Roma means following a series of norms and values of their own, which are transmitted from generation to generation, and hence their concept of space is closely linked to the family.The process of sedentarization of this ethnic minority in the cities, initiated in the 50s, meant that their life began to develop surrounded by norms different from their own.To face this new situation, they chose to live together and segregated from the majority society, usually in neighborhoods or shanty towns without basic services or infrastructures (Rodríguez, 2011:238-239).Taking this aspect into account, when spatial desegregation policies imply an excessive distance between the different family units of the shanty town, these people may lose the solidarity they found within the group that sometimes is essential for their subsistence.This aspect also directly collides with one of the fundamental values of this community.For this reason, desegregation interventions must consider this aspect, ensuring that new locations allow them access to meeting points of the entire group.Closely related to this aspect is the problem of involuntary displacements.It is essential that the families that are coming into a public process of rehousing should take part in the decision making and that such a choice be taken jointly.Without positive attitudes, facing the slightest conflict in the new relocation, the return to the protection of the shanty town appears as a very probable scenario.Thus, the economic effort of the government would have been for nothing.In fact, previous interventions carried out in Spain have shown that in those cases in which the aspirations of those affected have been met, there have been very few abandonments of the new homes.In cases when abandonment occurred, it was caused by the impossibility of attending to the monthly payments, not by problems of coexistence (Montes Mieza, 1986:158-159).
As for the part of the majority group, it is necessary to point out the persistence of problems of discrimination against Roma people in the private housing market.Even when the members of this minority have enough economic means, very few landlords are willing to sell or rent their properties to them, so they are relegated to the substandard housing of their shanty towns.This process of withdrawal is not unique to Spain, but is found at European level.According to the data provided by the FRA 1 (2009:62-64), the European Union Minorities and Discrimination Survey shows that 11% of the members of the Roma community interviewed by the agency EU-Midis2 suffered discrimination in the previous twelve months for ethnic reasons when looking for rental or purchase housing in the seven member states in which they were interviewed.Seventy percent (70%) of them indicated that they suffered this discrimination more than once.In addition, about a quarter of the European citizens interviewed recognize that they would feel uncomfortable having a Roma as a neighbor.These data highlight the vulnerability of this group when they want to access standardized housing.
It is also necessary to consider that the type of policies that each government decides to implement will be one of the determining factors for explaining social inequity (Barreiro-Gen, Novo-Corti and Ramil-Díaz, 2013:239) and spatial inequity in cities (Jiménez, 2008:38).Since the 1960s, the housing policies carried out in Spain to facilitate access to groups with more difficulties were mainly based on the construction of public housing in the periphery of the cities, which were then sold at reduced prices.On one hand, these types of policies did not contribute to reducing the segregation of these groups, since a large part of these peripheral areas ended up becoming marginal neighborhoods.This phenomenon of displacement to other disadvantaged areas is called "horizontal displacement" (Bolt, Van Kempen and Van Weesep, 2009:512).Because of these failed experiences, local governments began to include transversal plans of social insertion within their relocation actions of groups with difficulties (Tormo Santonja et al., 2003).On the other hand, until the change of legislation in 2005, these houses stopped belonging to the social park as soon as they were sold.For this reason, local governments currently do not have enough public housing in which to accommodate the neediest (Rodríguez, 2010:131).Moreover, there is no housing policy that compensates or regulates the inaccessibility of the housing market (Subirats et al., 2004:14).
Given the social and economic relevance of this problem, it is necessary to analyze the actions aimed at the eradication of slums in terms of ethics and inclusive results and long-term well-being.In this work, the public policies carried out recently in the community of Galicia, located in the northwest of Spain, have been analyzed.In this, measures to promote inclusion and permanence in the new houses have been incorporated, considering the experience of previous failed actions.For this, a study of the context of the application of these measures has been conducted, analyzing specific examples and their consequences.Then, an analysis of the perceptions of the population affected by these measures towards the main areas of inclusion was carried out.This analysis has been accomplished by means of a questionnaire conducted in Galician municipalities in which dispersion policies have recently been applied for the relocation of the inhabitants of shanty towns.Then, following the two-pronged process for integration, the opinions that the two social groups have about housing, education, employment and health policies have been gathered, as well as other aspects such as social relationships, citizenship and gender differences.With the collected data, a quantitative comparison of means was performed.The results show that, on this occasion, the dispersion policies carried out do not seem to lead to greater isolation of the population, since there are favorable attitudes towards inclusion on the part of both social groups, gypsies and nongypsies.However, aspects in which it is necessary to reinforce public action have been detected.

SLUM CLEARANCE PUBLIC ACTIONS. THE NORTHWEST SPAIN CONTEXT
Shanty towns began to emerge in the autonomous community of Galicia because of the process of sedentarization of the Roma community.According to the Valedor do Pobo (2005:19-21), in the beginning of the 20th century, Galician Roma families were nomads and they made their living from the exchange of goods and services with the agrarian population, with trades such as grinders, basket makers or blacksmith, and helping in the harvesting.Their sedentarization took place in a short period of time from the fifties to the mid-sixties, and it was caused by the advance of industrialization, migratory field-city movements, the mechanization of agriculture and the development of new activities which required the establishment of a permanent home address.The change from rural to urban habitat meant a rupture of their social isolation and the beginning of a new type of social relations with the majority society.The Galician Roma community settled mainly in low-rent flats, abandoned houses or empty buildings, located in the center or periphery of cities, while people with fewer economic resources built shacks on empty land.
To solve the difficulties suffered by this ethnic minority in terms of housing access, the government published Housing Plans, which constituted the general legislative framework for support and financing measures for rehabilitation and access to housing.Each autonomous community must adapt these plans to the specific needs of each territory, by the exclusive competence conferred by the Spanish Constitution on housing.The State Plan to Promote Housing Rental, Building Rehabilitation, and Urban Regeneration and Renovation 2013-2016 (Ministerio de Fomento, 2013) is currently in force at state level.This plan devotes its chapter IX to the Program for the Promotion of Sustainable and Competitive Cities.One of its strategic lines is slum clearance accompanied by social programs.
In addition, in Galicia there are several regulations that include housing access measures for people in need.Specifically, Law 8/2012, of June 29th, Housing Galicia (Presidencia da Xunta de Galicia, 2012) devotes its article 49 to housing insertion or assistance, and establishes that the regional administration may implement specific lines of collaboration agreements with administrations and entities without spirit of profit to carry out housing access programs.There is also Law 10/2013, of November 27th, on the social inclusion of Galicia (Presidencia da Xunta de Galicia, 2013), which includes in article 48.1 the subsidy allocated for facilitating the transit to standardized housing for people who live in shacks or substandard housing.On the other hand, the Roma Development Plan finances the actions of Galician city councils, aimed at preventing and fighting the social exclusion of Roma communities.The calls for this plan establish the following as priority areas of action: the slum clearance and access to standardized housing, the normalization of Roma students, employability and socio-labor insertion, the integral promotion of health, access and proper use of standardized services and the improvement of social participation.Actions are subsidized in the areas of employment, housing, education and health, through personalized itineraries, following the guidelines of the European Union in terms of social inclusion.
As seen through the legislation, there is currently a general consensus when applying slum clearance policies in relation to the need to include measures that facilitate access to housing within a broader public program, encompassing actions aimed at accessing other areas such as education, employment or health.But these cross-cutting plans have not always been carried out when implementing these types of policies.As an example, we have compared the public interventions carried out in the Galician shanty towns of A Cubela and Penamoa (A Coruña), Freixeiro and San Mateo (Narón) and O Carqueixo (Lugo), highlighting the following evolution: • These shanty towns originated from the decisions of local governments, who looked for land outside the city for the construction of barracks-type homes to which they could move Roma people who resided in key areas of the cities.These types of measures were also taken at a general level in Spain in the 60s and 70s (San Román, 1997:217), but they have been repeated in the 80s in cases such as the eviction of the shanty town of A Cubela, located in a key area of the city of A Coruña.This eviction generated the emergence of the shanty town of Penamoa on the peripheries of this city in 1984, one of the most conflictive in northwest Spain, whose eradication was completed in 2012.
• Urban motives cause the eviction of families: the construction of a shopping center in the case of A Cubela, the construction of an important access road to the city in the case of Penamoa, the construction of the Macide Architect Hospital in Ferrol, which gave rise to the transfer of families to the shanty town of San Mateo in Narón in the mid-70s, or the urbanization of the Fingoi area in Lugo, which gave rise to the shanty town of O Carqueixo in the end of the decade of 1960.The growth of the cities themselves seems to expel the Roma families towards areas increasingly distant from the urban center.
• Once the displacements have been made, the provision of services and the maintenance work of barracks-type dwellings are insufficient, so the area suffers from gradual deterioration.
• When the new relocation plans begin, in the year 2006 in the case of Penamoa and in the year 2010 in the case of Lugo and San Mateo (Narón), measures are taken so that new families do not settle in the shanty town.In these cases, the procedure was the demolition of the shacks occupied by the rehoused.
• Along with new rehousing, to a greater or lesser extent, transversal actions are applied, which include various aspects such as education or employment.
• In these more recent relocation actions, the dispersal of resettled families has already been taken as a criterion of action, as we have observed in the eradication of Penamoa, in the last relocation plan of O Carqueixo and in the action carried out in San Mateo.This dispersion has taken place both within the municipalities and throughout the bordering municipalities.In the case of San Mateo, the new location of displaced families is in the municipality of Ferrol and, in the case of Penamoa, the dispersion was carried out, in addition to the municipality of A Coruña, in the municipalities of Sada (3.3%), Oleiros (6.6%), Cambre (10%), Culleredo (10%) and, mainly, Arteixo, where 23.3% of the new locations are located.

METHODOLOGY
A quantitative analysis is employed in this study.A questionnaire has been prepared to collect the perceptions of the people affected by the public programs of slum clearance.The goal is to check whether the measures implemented have been effective in terms of the development and integration of those affected, in line with the European proposals for sustainable urban development.Two statistical procedures are conducted for the analysis of the data resulting from the questionnaire, allowing to contrast hypotheses over means: a t-test for two independent samples and a one-way analysis of variance.
The t-test for two independent samples allows the comparison of two different population groups, as in this case the Roma and non-Roma population.It is also used to compare, within each of these two groups, the female and male population.The t-test allows to test the null hypothesis that the population means are equal in both groups: If there is a difference between the means, the t-test will show a statistical significance or p value less than 0.05.It is necessary to specify that the program used, IBM SPSS Statistics, includes two versions of the t-test, depending on whether the prerequisite that the two population variances are equal can be assumed or not3 (Pardo Merino and Ruiz Díaz, 2005).
For the comparison of means of the quantitative variable, when the qualitative variable is divided into three or more categories, an analysis of the Anova variance is used.This test contrasts the null hypothesis that the means of the distributions of the quantitative variable in each of the independent groups are the same: If there is a different mean, the Anova test, which uses the F distribution, will show statistical significance or a p-value of less than 0.05.This test is used to contrast the means of the answers according to the municipality in which the respondents reside.To check the previous requirement of homogeneity of variances, the Levene test is used, whose p-value or significance must be higher than 0.05 to accept the null hypothesis of homoscedasticity.In the case that we cannot assume equal population variances, we have chosen the Brown-Forsythe statistic, as a robust alternative to the F statistic (Pardo Merino and Ruiz Díaz, 2005).
To differentiate between those groups in particular where the difference of means occurs, a series of multiple post hoc or a posteriori contrasts are conducted.The Tukey honestly significant difference procedure is used in the case of assuming equal variances and the Games-Howell method in the opposite case.The Dunnett Method is also used, since it has the unique possibility of establishing a control group with which to compare the other groups, controlling in turn the error rate for k-1 comparisons (Pardo Merino and Ruiz Díaz, 2005).
These post hoc tests allow us to test the null hypothesis that the means in each of the possible comparisons are the same: If the level of significance of these tests is less than 0.05, the null hypothesis is rejected due to the existence of significant differences between the two means of the i-th comparison.
In both methods of comparison of means, both in the t-test and in the analysis of variance, the previous requirement regarding the normality of the sample has been obviated due to its size (N = 209).
Regarding the developed questionnaire, it is necessary to highlight that it includes perceptions on social inclusion and inter-ethnic acceptance following the double integration path.Each item is designed to capture the perceptions of both the majority society and the ethnic minority.Through the different statements, the existence of barriers to accessing the different spheres that provide a decent standard of living is analyzed.We focus on the territorial dimension of exclusion and access to standardized housing, although the dimension of ethnic identification is also taken into account.For the elaboration of the questionnaire a strict methodological program has been followed.This was based on content validation through experts and on the performance of a pre-test prior to its application.In the content validation, the iterative method Delphi has been applied (Landeta, 1999).Through this method, consensus has been reached on the part of the 10 participating experts on the validity of the questionnaire to measure the attitudes towards social inclusion of the population affected by slum clearance programs.
The pre-test was administered to 30 people, who indicated their degree of agreement or disagreement with the items provided by a Likert scale of 5 points (1= strongly disagree, 5 = strongly agree).Special care was taken to ensure that this pre-test would be reproduced with the same degree of accuracy in the conditions in which it was finally put into practice (Cea D' Ancona, 2005:95).The respondents were recruited through non-random sampling, according to quotas of both social groups, seeking that the group of Roma people not fall below 25% of the total sample.Their collaboration was requested voluntarily and anonymously, and in all cases the questionnaire was self-reported, with the aim of reducing the bias of social desirability (Cea D' Ancona, 2009:22).This test was carried out during the months of March and April 2014.Sagrada Familia Municipal Civic Center, in A Coruña, Spain, was taken as a sampling point.Online questionnaires were distributed throughout the autonomous community.In these, a blank space for comments were left.The participants belong to the two social groups, Roma and non-Roma, in a proportion of 30 and 70 per cent respectively, and they are in an age range between 13 and 59 years.
Once the questionnaire had been refined, it was applied.This questionnaire was completed by 209 people, residents of cities where a dispersed relocation of people from shanty towns has recently been carried out (A Coruña, Arteixo, Cambre, Culleredo, Ferrol, Lugo, Narón, Oleiros and Sada).The participants indicated their degree of agreement or disagreement with the items provided through the same 5-point Likert scale that was used in the pre-test.Similarly, the respondents were recruited by non-random sampling, by quotas, so that the Roma group accounted for at least 25% of the sample.The participants have collaborated voluntarily and anonymously.The questionnaire was self-reported, with the exception of a very small number of participants with literacy problems, which were completed by a third person who read the items.The questionnaire was conducted during the months of June to September 2014.In the case of the Roma population, the questionnaire was delivered by specialized staff of the Fundación Secretariado Gitano (FSG).
The participants belong to the two social groups, Roma and non-Roma, in a proportion of 27 and 73 per cent respectively, and they are between 17 and 77 years old.

CHARACTERISTICS OF THE SAMPLE
The main characteristics of the people who participated in this study are summarized below.They have been analyzed by separating the two population groups that are in the sample, Roma and non-Roma people.
Age.The Roma population is concentrated in the age range from 17 to 30 years, while the majority population is more spread out among the different age ranges, the majority from 31 to 45 years.
Sex.There is a parity proportion in the Roma group, while women predominate in the majority population (67% of the sample).
Municipality.In the territorial classification, 44.7% of the respondents belong to the municipality of A Coruña.There is a similar proportion of people within the two groups that reside in this area.Then, there is the city of Lugo, with 19.1% of the sample and Culleredo with 15.8%.The remaining municipalities have the following proportions: Arteixo 6.6%, Cambre 5.3%, Ferrol 3.9%, Oleiros 2.6%, Sada 1.3% and Narón 0.7%.
Housing.Regarding the regime of housing tenure, 62.5% of the respondents of the majority group own property, while a similar proportion of the Roma respondents (61.4%) reside in rented housing.
Education.Almost half of the majority population surveyed has university studies or equivalent, while 52.6% of the participants of Roma ethnicity have primary education, although it is noteworthy that 22.8% are in the next level of secondary mandatory education.
Internet access.Regarding the handling of new information and communication technologies, we have considered Internet access as an indicator in this study.Although the percentage of people in the majority group with access is much higher, the percentage of Roma people surveyed who also manage this resource is high, at 75.4%.
Income.Regarding the source of income, most of the participants in the majority group are salaried workers, compared to 56.1% of Roma respondents who declare that they do not have any source of income.
Health.It is also worth noting that all the Roma who participated in this study have access to the public health system through the health card.

RESULTS
We will now analyze the main results obtained with the questionnaire, separated according to the study areas.

HOUSING
Table 1 shows the results of the items corresponding to the housing area.There is a large proportion of respondents who agreed with the statements that propose an adaptation and improvement of the living conditions of people who move to reside in standardized housing, located in an integrated environment.However, in items 2, 4 and 8 such tendency is not observed.In the case of items 2 and 8, which refer to relocating in the same neighborhood and renting houses without distinction of ethnicity, the percentages in the five point scale are very similar.There are also significant differences in both items according to whether the respondents belong to the minority group.In the case of item 2, we obtain the following values of the test:4 t 87,642= -2.341 (p=0.022).People of Roma ethnicity are more in disagreement with the relocation in the same neighborhood. 5his data is in line with what previous research has demonstrated, about the preference of the Roma people surveyed to move to areas with non-Roma neighbors due to the fear of disagreements between Roma people of different families (Gago-Cortés and Novo-Corti, 2015:55).Also, the results of item 8 are: t 207 =7.036 (p=0.000).People of the minority group are more in favor of renting housing without distinction of ethnicity.The average of their answers is 4.088, while in the majority group is much lower, 2.684.Finally, in the case of item 4, referring to the opinion on municipal action policies, 23.9% of respondents say they have little or no agreement with the measures carried out in the shanty towns.In this case, there are also significant differences between the opinions of both groups.The results of the test are: t 207 =5.099 (p=0.000).People of Roma ethnicity are more in agreement with the municipal actions. 6n Anova test was also conducted for the items in this section, according to the municipality in which the respondent resides.The results show that there are no statistically significant differences between the evaluations of the participants according to their municipality.

EDUCATION
Table 2 shows the results for education.In this area there is a common agreement on the importance of schooling, non-segregation in the classrooms, the improvement of education and intercultural learning.In the Anova test conducted for these items according to the municipality of residence of the respondents, the existence of significant differences is only shown in item 11, referring to the segregation of Roma students in specific classrooms.The results obtained for this item are the following: Levene=8.89(p=0), .Therefore, we proceed to perform the post-hoc test of Games-Howell, to check in which municipalities there are significant differences.The results show that there are significant differences between the municipalities of Culleredo and Arteixo (GH=1.13 with p=0.006).The respondents of Culleredo are more in agreement with this statement, with an average of answers of 2.20, than the respondents from Arteixo, with an average of 1.07.Yet, it is necessary to point out that the average of both municipalities is within the range of disagreement of the proposed Likert scale (1 strongly disagree, 2 somewhat disagree).

SUBSIDIES
Table 3 shows the results of item 15, referring to the dependence on economic subsidies for carrying out relocation processes in a standardized housing.A large majority of respondents agree with this aspect.The Anova test performed for this item shows statistically significant differences between the municipalities of Narón and Oleiros.The results are the following: Levene = 1.85 (p = 0.07); F 8,200 =2.11 (p=0.036).The post hoc test shows significant differences between the municipalities of Narón and Oleiros (Tukey=2.27with p=0.02).The neighbors of Narón are much more in agreement with this statement, with an average of answers of 4.9.However, on average, the respondents from Oleiros somewhat disagree, with a value of 2.6.

SOCIAL RELATIONSHIPS
The results for the relational area are collected in table 4.These results point towards a division of opinions in item 17, and in items 21 to 24, while in the remaining items there is a high proportion of agreement to inter-ethnic coexistence (almost 70% somewhat agree or strongly agree), with the weight of crime hindering inclusion, with the existence of mutual trust, with the maintenance of culture in the integration processes and with the children of both ethnicities sharing play areas.Item 17 shows the existence of a good relationship between both groups, and although most of the participants show some degree of agreement, almost 24% say they strongly disagree or somewhat disagree with the statement.There are also significant differences in the answers given by each group to this item, according to the results of the test: 7 t 83,244 =4.240 (p=0.000).The minority group is who agrees more that the members of both groups get along.On the other hand, items 21 and 22, referring to marriage between members of both groups, share very similar proportions in all response options, with a very slight difference when it is the son who marries (64.6% neither agree nor disagree, somewhat agree or strongly agree), or if it is the daughter (62.3% have marked one of those options).The results of the tests for these items do not show significant differences in the responses of each social group.In the case of item 23, which proposes carrying out support activities for Roma families, there is a considerable percentage that strongly disagrees or somewhat disagrees (22.4%).Finally, item 24 suggests, in the view of the participants, problems of adaptation to the new neighborhood environment of the relocated (70.3% strongly disagree, somewhat disagree or neither agree nor disagree with there are no important complications with the new neighbors).In this item there are statistically significant differences in the test performed: t 207 =4.545 (p=0.000), with the minority group indicating more agreement with the statement. 8In the Anova test conducted according to municipality, significant differences were detected for items 16, 17 and 20.In relation to item 16, referring to the inter-ethnic coexistence capacity, the results are as follows: Levene=1.96(p=0.05);F 8,200 =2.35 (p=0.02).The post hoc test carried out for this item shows significant differences between the municipalities of Narón and Oleiros (Tukey=2.28with p=0.02).The neighbors of Narón are again more in agreement with this statement (average of 4.9), in contrast to the opinions of the inhabitants of Oleiros, who have an average of only 2.6.
Regarding item 17, the Anova test shows the following results: Levene=1.01(p=0.43);F 8,200 =2.942 (p=0.004).In Tukey's post hoc test, no comparison has turned out to be significant.Observing the means categorized by city council, Narón and Culleredo are the two cities in which the population is most in agreement with the fact that the two social groups have a good relationship.Sada obtains the worst average of answers (table 5).Therefore, we decided to carry out Dunnett's post hoc test, taking as a control group the municipality of Sada and comparing it with the other municipalities.As expected, the test shows significant differences between Sada and Culleredo (Dunnett=2.3 with p=0.02) and between Sada and Narón (Dunnet=2.38with p=0.03).
Finally, in item 20, referring to the need to separate from one's own culture to achieve integration, the following results have been obtained: Levene=1.8(p=0.08);F 8,200 =1.993 (p=0.049).The post hoc test shows significant differences between Narón and A Coruña (Tukey=1.603with p=0.03) and between Narón and Arteixo (Tukey=2.179with p=0.01).The neighbors of Narón are more in agreement with this statement, with an average of 3.75, while those of A Coruña and Arteixo show on average a lower degree of agreement (2.15 and 1.57 respectively).

CITIZENSHIP
Regarding the results for citizenship, listed in table 6, it should be noted that a large proportion of respondents agree with the access of the minority group to public services and political positions, and that the normalization of citizen behavior can be favored through the access to standardized housing.However, there is a discrepancy in the responses to item 28, which includes compliance with social norms by Roma people.Almost 70% of the participants are some, little or not at all in agreement with this statement.There are statistically significant differences in the responses to this item according to the social group:9 t 80,012 =6.037 (p=0.000), with the minority group showing the strongest degree10 of compliance with the social norms.In the Anova test performed for this group of items, no statistically significant differences were found in the answers according to the municipality of origin.

EMPLOYMENT
Table 7 shows the results for the employment area.We can observe a large majority of participants in agreement about the fact that the minority group faces greater difficulties when they look for a job, with the need for public support, along with the acceptance of the members of this group as co-workers.The Anova test does not show differences in the answers for these items according to the municipality in which the respondent resides.2.9 1.0 6.7 13.9 75.6 100.0 Source: Elaborated by the authors based on SPSS frequency chart.

HEALTH AND HYGIENE
The results for the health and hygiene area also show a high proportion of respondents in agreement with the difficulty of maintaining proper cleanliness in the slums, the importance of healthcare for coexistence, the need for public support in this scope and the importance of image and hygiene for acceptance in society (table 8).An Anova test was performed for this group of items, which did not show significant differences in the answers according to the municipality of belonging.

GENDER DIFFERENCES
The proportions of respondents who completely agree with the items in this section are very high (table 9).This reflects a clear tendency towards equality between men and women in basic aspects such as education, employment, distribution of domestic obligations and decision-making.However, it should be noted that there are statistically significant differences between the answers provided by each group in these items, with a higher average of responses from the majority group (table 10).Within each social group, the responses of women and men have been compared for each of these items.The results of the tests show that there are no statistically significant differences between both sexes in the majority group.And in the minority group, the only significant differences have been found in item 40:11 t 37,379 =2.640 (p=0.012).Roma women agree with working outside the home more than men. 12he Anova test conducted for these items does not show statistically significant differences in the answers according to the municipality of residence of the respondents.Source: Elaborated by the authors based on the SPSS Test of independent samples and statistical of group.In these tests, no equal variances have been assumed.

CONCLUSIONS
Throughout this work we have analyzed the evolution in public actions for the slum clearance in northwestern Spain over the last years.In this regard, it is important to differentiate between those applied in the 60s, 70s and 80s, and the most recent ones carried out in 2006 and 2010.The first ones consisted of the spatial displacement and the concentration of families in the peripheries of the city.These interventions, far from solving the problem, have contributed to its persistence.The most recent ones have opted for a certain spatial dispersion of families, including plans to combat exclusion in different areas such as health, education, employment and housing, encompassing the European guidelines of the National Roma Integration Strategies (European Commission, 2012:6).Faced with the academic discussion generated about whether the desegregation of the most disadvantaged groups is favoring their social inclusion, a questionnaire has been designed.This reflects the perceptions of both the minority group and the majority group in the main areas of inclusion.It has been applied in municipalities where a policy of dispersion of families from shanty towns has been implemented.The results at a general level show favorable attitudes towards inclusion in both groups, which seems to indicate that these transversal desegregation policies are favoring a correct integration in the new neighborhood, in line with the guidelines set by European institutions.These also seem to be helping to alleviate this "city of corners" (Cartelle Fernández, Fernández de Sanmamed Sanpedro and Guillén Gestoso, 2003), giving a better position to those affected by the difficulty in accessing different facilities and public services.
In contrast to what might be expected due to their habits and strong family ties, and against research on social union capital that warned about the possibility of further isolation of the affected population after dispersion measures (Bolt, Ozuekren and Phillips, 2010;Arin, 1991), the Roma people surveyed are those least in favor of concentrating their relocation in the same neighborhood or area of the city.This shows a positive attitude towards coexistence with the majority population.The explanation can be found, on one hand, in the concentration policies carried out in the 60s, 70s and 80s.Through these measures, the local public authorities put together Roma families of diverse lineage, in a new space located on the periphery of the city, originating in different towns located in key areas of the city.This concentration led to the disappearance of the Roma person of reference followed by each of the families, as happened in the case of the shanty town of Penamoa, one of the most problematic in northwestern Spain.This caused numerous conflicts among the Roma who, faced with the risk that this could happen again, prefer to live with non-Roma people (Gago-Cortés and Novo-Corti, 2015).On the other hand, the dispersion carried out in these more recent actions, has not meant an excessive distance between the members of each family.When the relocation has not been possible within the municipality, this has been done in the bordering municipalities.In this way, the affected people can continue maintaining their strong family bonds, while also continuing to benefit from their networks of solidarity and protection.
The results also show that the Roma people strongly agree with the municipal actions carried out in the shanty towns.This agreement seems to show that these actions have been implemented taking into account the importance of involving those affected by the planning of intervention measures, so that decisions have been made in a consensual manner, in accordance with co-production (Hill and Hupe, 2011:155).
Regarding the need for intervention and public subsidies to carry out these types of actions, although a large majority of respondents indicate that they agree with this aspect, it is necessary to point out that there are significant differences depending on respondents' municipality of origin.The neighbors of Narón are much more in agreement with the need of this public support than the neighbors of Oleiros.Therefore, there are different perceptions regarding public intervention that must be taken into account and evaluated before implementing these types of actions.As mentioned, the cooperation of all participants involved is essential for the success of these public policies.
In the area of relationships between the two social groups, although there is a high proportion of respondents in favor of an inter-ethnic coexistence, significant differences have been detected between the municipalities of Narón and Oleiros.Oleiros has a greater degree of disagreement with this coexistence, an aspect that must be taken into account when choosing new locations for families moved.
Discrepancies of opinions have also been observed in regard to social relationships.Roma evaluates the existence of good relationships to a greater degree.The study of this aspect by municipalities seems to indicate significant differences between the municipality of Sada and the municipalities of Culleredo and Narón.In Sada the respondents agree less about the existence of good relationships.Therefore, the launch of campaigns aimed primarily at the majority population is recommended, especially in this municipality of Sada, in order to promote awareness and mutual understanding of both groups.
In addition, on the part of the minority group, the data collected have shown the persistence of problems in adapting to the new neighborhood.This result emphasizes the importance of family education and monitoring once the relocation has been carried out, to correct and avoid these types of problems as much as possible.With this monitoring, the minority group's compliance with social norms must also be addressed, since a large percentage of those interviewed think that this aspect should improve.
In the cultural sphere, there are significant differences between the municipalities of Narón and A Coruña and Arteixo.The inhabitants of Narón have a greater degree of agreement with the need to separate from the traditional culture to promote integration.Therefore, campaigns to reconcile both cultures and promote the Roma culture are recommended in this municipality, to make their identity known and to help in preserving it.
As for the gender section, discrepancies have been detected among Roma men and women, regarding the possibility of Roma women working outside the home, where women is more in favor of it than men.For this reason, it would be advisable to take advantage of this positive predisposition among women, to promote the employment of women of this ethnic group to alleviate, as far as possible, its traditional relegation to the home care.
In summary, although there are aspects to be improved, favorable attitudes and a good disposition have been detected in order to achieve a complete social inclusion of the people affected by public programs of desegregation and slum clearance.This seems to show that transversal actions in which access to housing has been combined with social development programs, can achieve the goal of attaining long-term welfare of this people.However, it is necessary to point out that this study is only a first approximation, and that it would be necessary to continue comparing the results with other areas where slum clearance policies have been applied.Likewise, temporary monitoring is also necessary to verify the continuity and solidity of results, since there are programs which started in 2010 that have not yet been completed, as in the case of Lugo and Narón.So, in these cases, it would be necessary to monitor their progress, and others such as the case of Penamoa, which started in 2006 and concluded in 2012, it would be very interesting to observe the sustainability of those public actions carried out over time.These will be the main questions of future investigations in this field.

TABLE 1
1.In my opinion, there should not be shacks in the cities.2.I think that people who leave the shanty towns should be relocated to the same neighborhood.Source: Elaborated by the authors based on SPSS frequency chart.

TABLE 5 RESPONSE
MEANS TO ITEM 17 BY MUNICIPALITY OF RESIDENCE OF THE PARTICIPANT 17.I believe that in my city, Roma and non-Roma people get along.Source: Elaborated by the authors based on SPSS descriptions by municipality.
Elaborated by the authors based on SPSS frequency chart. Source:

TABLE 8
Elaborated by the authors based on SPSS frequency chart. Source: