Received: 23 December 2021 | Accepted: 19 May 2023 DOI: 10.1002/psp.2675 RESEARCH ARTICLE Age, gender, and cohort in residential segregation: The case of African immigrants in Spain, 2000–2020 Silvia Gastón‐Guiu1,2 | Jordi Bayona‐i‐Carrasco1,2,3 1 Centre for Demographic Studies (CED)/ CERCA, Barcelona, Spain Abstract 2 Department of Geography, Autonomous The composition of the immigrant population by gender and age, and University of Barcelona, Barcelona, Spain characteristics at the moment of arrival plays an influential role in the levels 3 Department of Geography, University of Barcelona, Barcelona, Spain of residential segregation reached, and in their evolution over time. In Spain, the African immigrant population is notable for a marked gender imbalance, where Correspondence men outnumber women, and for continuous growth in the number of Silvia Gastón‐Guiu, Centre for Demographic Studies (CED), CERCA, Autonomous immigrants over the past two decades. It is also one of the groups presenting University of Barcelona, Barcelona, Spain. the highest levels of segregation. Using data from the Continuous Population Email: sgaston@ced.uab.es Register, this study analyses the residential segregation of this group from the Funding information triple perspective of age, period, and cohort for the six main Spanish cities. The Ministry of Economy and Competitiveness of Spain, Grant/Award Number: PID2020‐ results reveal a significant differentiation in segregation levels by sex and age, 113730RB‐I00 with lower rates in the more recent cohorts. This perspective offers a dynamic view of segregation indicators and enables better understanding of its future evolution. KEYWORDS African immigrants, age, cohort, segregation, sex, Spain 1 | INTRODUCTION regarding ways of sharing urban space, ends up leading to the isolation of population groups (Fortuijin et al., 1998; Musterd, Analysis of the spatial distribution of the immigrant population 2003). Nevertheless, in the first stage of arrival, segregation can and of residential segregation understood as the different ways in have positive effects in easing the initial adaptation of the which two or more population groups share and are distributed in immigrant population (Cutler & Glaeser, 1998; Peach, 2005; Van the space in which they reside, has a long academic tradition of Kempen & şule Özüekren, 1998), although it can become tracing the levels and evolution of residential distance between negatively perceived if the situation is prolonged over time different population groups (Massey & Denton, 1988; Musterd (Sabater et al., 2013). et al., 1999). Its interest lies in the fact that this segregation is In origin, segregation is a multidimensional phenomenon, the deemed to be unfavourable inasmuch as it can be an obstacle in product of a combination of several factors, which can be economic, the process of integration, hindering social ascent and perpetu- social, demographic, or ethnic and, moreover, expressed differently in ating the existence of unequal zones, whether poor (forced each city depending on the characteristics of its urban structure and segregation) or rich (self‐segregation) (Galeano & Sabater, 2016; housing market. Furthermore, it has an evolutionary component since Massey et al., 1994) which, added to individual preferences its values vary over time. This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made. © 2023 The Authors. Population, Space and Place published by John Wiley & Sons Ltd. Popul. Space Place. 2023;29:e2675. wileyonlinelibrary.com/journal/psp | 1 of 15 https://doi.org/10.1002/psp.2675 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2 of 15 | GASTÓN‐GUIU and BAYONA‐I‐CARRASCO In the case of Spain, studies on the dynamics of segregation of more difficult to achieve for Africans since access to the residential the immigrant population are recent as they only began to gain and job markets is conditioned by social images and prejudice in the relevance over the last two decades, coinciding with the intense host society which do not affect all immigrants groups to the same migratory boom of the early years of the 20th century (Martori & degree. As a result, Africans face more discrimination than European Apparicio, 2011; Musterd & Fullaondo, 2008; Bayona‐i‐Carrasco, and Latin American immigrants (Cea d'Ancona, 2002). 2007), which meant heightened demographic and residential com- Hence, we are concerned, first, to find out whether the intensity plexity and dynamics, and the reshaping of Spanish society as a result of residential segregation depends on the demographic character- of the intense compositional diversification of the population. The istics of the group being analysed and whether there is a availability of an annual population register (Padrón Continuo) compositional effect based on gender and age in the levels of enables the study of segregation from an evolutionary perspective segregation. Second, we analyse the period effect on the segregation that shows declining indicators as the volume of settled immigrants indicators and whether the year of arrival, defined in terms of the increases, with clear differences among origins (Bayona‐i‐Carrasco & economic period, affects the levels of segregation that are reached. López‐Gay, 2011; Domínguez‐Mujica et al., 2010; Martori & Third and finally, we analyse the existence of a cohort effect, which is Madariaga, 2023). to say the impact of initial levels of segregation over time on a However, little or no attention has been given to the relationship fictitious cohort of immigrants defined by sex and age. The interest of between the demographic composition of groups and segregation this threefold approximation resides in the fact that it offers a better levels, or to the effect of age, even when this was already highlighted understanding of segregational processes while also providing as an interpretive element in the segregational processes of the interpretive elements regarding their future evolution. 1960s (Uhlenberg, 2000), or to the effect of the moment of arrival on these processes. These are precisely the main aims of this paper in which analysis of ethnic segregation is approached from the three‐ 2 | LITE RATURE REVIEW dimensional perspective that considers sex and age, the moment of observation, and cohort, which then makes it possible, by means of 2.1 | Residential segregation from a demographic these three classical standpoints of demographic analysis, to offer a standpoint more complete overview of the segregational process. The study group is the African population (distinguishing As noted above, owing to its multidimensional nature, residential between North Africa and sub‐Saharan Africa) which, as one of the segregation should be observed from several perspectives. Those largest migrant groups in Spain,1 shows one of the highest levels of with the most explanatory potential are the economic standpoint, segregation, as well as migratory strategies with a marked gender which relates structure and spatial positioning with the social sphere imbalance. It is also one of the groups that is most affected by the (Coulter & Clark, 2019; Massey et al., 2009), and the preferential or economic crisis (Gastón‐Guiu et al., 2021). Among the first to settle in discriminatory point of view where personal preferences end up Spain (Berriane, 2004; Cohen, 2009), this group faces situations of determining the location of settlements (Schelling, 1971). With a greater segregation in terms of both employment and housing, not theoretical model of spatial assimilation, the progressive disappear- only by comparison with the autochthonous population but also with ance of segregation is expected, while with the segmented assimila- the other immigrants. Most Africans work in the secondary labour tion model (Portes & Zhou, 1993) segregation is maintained due to market, with precarious conditions of employment, low wages, and the coexistence of different patterns of incorporation in accordance little chance of promotion (Cachón, 2009). As for housing and with the segment of society where the immigrants are inserted. residential and living conditions, Africans also differ from the native Although there are studies on the influence of demographic population (Checa et al., 2011). The Moroccan population is that dynamics on segregation and its evolution (Finney & Simpson, 2009; which has the most difficult living conditions (Colectivo IOE, 2005). Kauppinen & van Ham, 2019), little or no attention has been given to Since their homes are smaller, overcrowding is worse than it is for the relationship between segregation levels and the effects of the other population groups owing to subletting of rooms to reduce costs three classical approaches of demographic analysis. These models and the cohabitation of several family nuclei in the same dwelling, tend to be used to ascertain the effect of time on the occurrence of while their homes tend to be located in old buildings, which means an event (González et al., 2002) and the impact of each variable on that, in most cases, they are in a rundown state (Domingo & Bayona, the phenomenon which, in this case, is the residential segregation 2007). The usual tenancy regime is renting, while for Spaniards it is indicator. This analytical tool therefore enables consideration of ownership. Although it has been shown that, as the years of changes occurring over time, which may be the result of population residence in the host country increase, these situations of residential ageing, the fact of being born in a certain period, or sharing the disadvantage can change (Leal & Alguacil, 2012), improvement is experience of a particular cohort (Ramalle, 2009). Nevertheless, some authors consider that, as a result of the close links between variables, it is difficult to obtain the specific weight of 1 In 2022 Africans, numbering 1,371,755 residents, represented 2.9% of the Spanish each one because of the influence of their multicollinearity population. Approximately 77.8% are from North Africa, and 300,000 are from sub‐Saharan Africa. (Settersten, 1999). However, the importance of the effort to 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GASTÓN‐GUIU and BAYONA‐I‐CARRASCO | 3 of 15 distinguish between the three effects is due to the fact that, although experienced with high intensity in Spain in recent years have had a they allude to temporal effects, their influence on the effect being well‐known effect on international migrant flows (Prieto et al., 2018) analysed is different (Debiasi, 2018). Age effects are internal to but also on the real‐estate market. Hence, the economic situation is individuals, reflecting biological factors and social processes (Yang & reflected in the rental market, easing access to housing in periods of Land, 2013), to which considerable attention is devoted in keeping crisis and showing steep increases in times of growth. Finally, cohort with the different stages of the life cycle, especially changes effects derive from differences between groups that start from a occurring during youth as well as in the ageing process. In this way, common initial event, usually, the year of birth, and relate to the changes in the tendency of a phenomenon brought about by macroconditions that are experienced by the different cohorts in the chronological age are identified. course of life. In the case of the immigrant population, the aim is to In fact, traditionally, the age variable, although it has been seen ascertain whether the more recent cohorts follow the same as essential because of its influence in residential choices (McKenzie, residential path as earlier cohorts (Sabater & Catney, 2019). 1924; Schelling, 1969), has been given scant attention in segregation analysis (Giele & Elder, 1998; Halli & Rao, 1992; Rhein, 1988). Yet, as Cowgill (1978) states, structural changes are being stimulated by the 2.2 | International immigration, residential generalised population increase, causing differentiation and specia- segregation, and city in Spain lisation by age, and leading to residential patterns that transcend those produced by origins (Sabater & Catney, 2019). Another Spain conforms with the segregation patterns and dynamics observed determining factor is the calendar of different life‐cycle stages for southern Europe (Arbaci & Malheiros, 2009), with moderate or individuals are going through (Sullivan, 2020). All of this makes it even low levels of segregation, albeit with poor housing conditions. more likely that there will be geographic zones occupied by people This means that there is no direct relationship between segregation belonging to the same generation or cohort, and the dissociation or indicators and social inequality (Arbaci, 2019). In the Spanish case, distancing of others, thus obstructing intergenerational contact (Riley the settlements of immigrant populations present certain singularities & Riley, 2000). This separation can derive from natural, voluntary, or by comparison with the rest of Europe since they were late in involuntary processes due to social and residential preferences, occurring, with a relatively short and high‐intensity temporal which are largely conditioned by the different stages of life (Deng & development.2 Moreover, the volume, origins, and characteristics of Mao, 2018). However, in the case of the immigrant population, it flows have been changing and, unlike what is happening in other could also be due to the existence of a range of migratory strategies countries, are notable for their high degree of heterogeneity3 which differ in accordance with the characteristics of flows (Domínguez‐Mujica et al., 2010). depending on origin. For example, in the African case with which In the national scientific literature on residential segregation we are concerned here, there is a predominance of young men, and linked to the immigrant population the conclusions drawn coincide this could have an effect on overrepresentation in certain urban with trends observed for other European countries, with relatively areas. For Canada, Okraku (2008), introduced gender as well as age moderate levels tending to gradual reduction at the same time as the and found the existence of distinctive patterns generated by both numbers of immigrants rise, to fall with the economic crisis, and to variables, with higher segregation rates among older women by stabilise with recovery. Nonetheless, the considerable variability comparison with the rest of the population. Recently, Wong and Das observed in segregation levels by origin should not be overlooked Gupta (2023) have shown that non‐Whites experienced higher age since, while Eastern Europeans and Latin Americans generally present segregation than Whites, while older adults experienced higher racial low or very low segregation levels (Vono & Bayona‐i‐Carrasco, 2010), segregation than the other age groups. On the European scale, among the best contributions to the 2 study of segregation by origin and age are those produced by Sabater The first and not very significant inflows of immigrants date back to the 1970s. A second phase (1985–1995) coincides with a phase of consolidation of Spain as a country of (2010) and Sabater and Catney (2019) which, from an evolutionary settlement. This period was followed by the well‐known period of acceleration (1995–2007) standpoint, analyse the determinant role of age and origin in in which immigrant populations grew rapidly, as did the diversity of origins. From one million and 2.7% of the population in 1996, the numbers jumped to six million and 13.1% in 2008. segregational processes in England and Wales. They find evolutionary Later, after the onset of the economic crisis, with a smaller volume of arrivals and a greater variability in the indicators and also the existence of fluctuating number of departures, the overall number of immigrants remained stable. After 2014, however, there has been a new upsurge in arrivals, with inflows in 2019 similar to those of results according to age (higher at initial and more mature ages). The the period of greatest growth. rates rose or fell in keeping with the influence of life stages, with the 3 Until the 1990s, most immigrants came from European Economic Community (EEC) lowest segregation values appearing at the intermediate adult ages member countries. They were joined by the first political exiles from Latin America and, later, because of economic questions, by Ecuadorians, Peruvians, Dominicans, and Colombians. (20–44) by comparison with the rest, and also changes in incidence After the 1990s, inflows from Africa also increased, although their numbers had been depending on origin. notable since the 1970s. At present, together with Latin Americans, they constitute the most numerous non‐EU contingent. Notable in the case of Africa are inflows from the north Then again, period effects are the result of historical events or (Morocco followed by Algeria) and from sub‐Saharan Africa (e.g., Senegal). Eventually, in the changes occurring over time and at particular points in time. Any early 2000s, there was a steep rise in arrivals from Eastern Europe (mainly Romania and Bulgaria) to the extent of becoming another of the more numerous population groups in contextual or circumstantial change can affect demographic phe- certain areas of the country. Likewise, and also more recently, immigrants of other Asian nomena (Portrait et al., 2002). The various economic stages origins (China and Pakistan), have joined the traditional inflows from the Philippines. 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 of 15 | GASTÓN‐GUIU and BAYONA‐I‐CARRASCO the opposite is true for Asians and Africans who show sharp population despite the fact that some councils have made it difficult divergence from other population groups, especially in the case of for some groups to register. The data has been available since 1998, Asians. The effect of the economic crisis, which has an impact on the and this study analyses figures from the period 2000–2020. In the reduction of segregation, should not be ignored either. A separate representation of results, four moments are highlighted, coinciding case is that of rural areas, especially in the south of Spain, where the with the economic phases that Spain has gone through in recent demand for labour is high owing to the productive model based on years. In 2002, this would correspond with the period of economic intensive agriculture. In these situations, the segregation levels are growth and the migratory boom; 2008 and 2014 would refer to the much higher than those observed in big cities. And added to these are economic crisis (the first phase and final stage, respectively); while exclusion from urban areas and peripheral locations in a situation 2020 would account for the recent years of incipient economic predominantly affecting African immigrants (Checa et al., 2011). recovery. Similarly, there are higher levels of segregation in the southern cities The evolutionary analysis is applied to a selection of specific of the country, where immigration is quantitatively less substantial groups of immigrant population resident in Spain through the and more recent, especially at the metropolitan level (Benassi variable ‘country of birth’. Immigrants of African origin have been et al., 2020). selected as the population under study, disaggregated by geo- Despite the large increases in the immigrant population and the graphical area into those born in the countries of the north effects of the economic crisis, segregation levels have remained (Morocco, Algeria, Egypt, Tunisia, and Libya) and those in sub‐ moderate for most origins. However, hidden behind this fact is a Saharan countries (a group comprising the rest of the African growing concentration in the poorest parts of cities (Benassi & countries), as this is one of the groups with the oldest presence in Iglesias‐Pascual, 2022; Iglesias‐Pascual et al., 2023), and increasing Spain, and also that presenting the greatest divergence by gender, vertical segregation, with a notable concentration of immigrants from with a preponderance of young men in the early stages of arrival low‐income countries on lower floors (Leal & Sorando, 2022). Crucial and subsequent family reunification in which women would be the in this distribution is the housing market with its discriminatory outstanding group. Furthermore, and in a complementary sense, practices, especially in renting, which have the effect of perpetuating other groups of immigrants—Latin Americans and Europeans from segregation (Bosch et al., 2015). the former EU‐15—are included in the analysis for comparative purposes and to offer an overall view of the segregational phenomenon in Spain. Hence, the comparison is made between 3 | D A T A A N D ME T H O D O L O GY the two largest groups of African immigrants with the two groups that are most established in Spain. The indicators are applied to the In this study, we use microdata from the Padrón Continuo de country's six largest cities, in the following order: Madrid, Población (Continuous Population Register), which is produced by the Barcelona, Valencia, Seville, Zaragoza, and Malaga, with the aim National Statistics Institute (INE) and based on information collected of conducting a double analysis: microspatial and comparative. The in registers and variables of municipal censuses managed and cities were chosen because of the size of their populations (from maintained by each local council. This resource is subject to constant 578,460 inhabitants in Malaga to 3,334,730 in Madrid), as well as revisions and updates because of changes occasioned by registra- the fact that they are the places that have traditionally received tions and cancellations arising from births, deaths, and migratory the largest volumes of migrant flows, which have had a direct movements. It provides information about the official populations of impact on their growth and population structure that is not visible towns and cities in Spain and includes variables pertaining to the in other, smaller cities. location of residence, gender, age, nationality, and place of birth. The To measure segregation, an equality indicator is used, namely the census tract is the smallest unit in which data are collected in Spain Duncan Dissimilarity Index (DI) (Duncan & Duncan, 1955), which and ranges in population size between 1000 and 2500 people compares the distribution of a population group in the geographic (between 72% of the tracts of Madrid and 94.5% of those for space being studied with another population group, usually the Barcelona). The divisions are of similar population weight, so we majority group, in this case, the autochthonous population born in consider that the comparison of cities is not affected by zoning Spain. The results obtained oscillate between 0 (absence of problems4 (known as MAUP effects, Wong, 1997). In our case, they segregation) and 1 (maximum segregation, when no territory is vary from 433 in Malaga to 1068 in Barcelona, and 2433 in Madrid. shared). It is formulated as follows: The periodicity of the Continuous Register is annual, dating from 1 n xi yi January 1. One of the main characteristics of this register is that it D= ∑ 2 i =1 X − Y 0 ≤ D ≤ 1, considers the entire population, independently of the legal situation in the country, which means a good count of the immigrant where xi is the population of the minority group in census tract i; X is the total population of the minority group in the municipality; yi is the population of the majority group (autochthonous population) in tract 4 This problem would only by eliminated by using a Population GRID. However, we have i; Y is the total autochthonous population of the municipality; and n is discarded this possibility since those existing in Spain do not have the disaggregation by sex and age required by this study. the number of tracts in the municipality. 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GASTÓN‐GUIU and BAYONA‐I‐CARRASCO | 5 of 15 The DI is highly sensitive to group size. Hence, changes in biggest Spanish cities. When the Location Quotient6 is used as a unit segregation patterns should be interpreted with caution since they of measurement, the maps show the existence of areas of evident could be the result of natural population growth or decline instead of concentration, as is the case of the city of Barcelona, or a marked strictly segregational causes (Simpson, 2004). This can be appreciated absence, as occurs in the wealthy neighbourhoods of Barcelona, in the varying results according to population size in the evolving Madrid, and Zaragoza, where entry is difficult for immigrant study by Napierala and Denton (2017) measuring residential populations (Rubiales Pérez, 2020). A shared pattern is found in the segregation in the United States from 2000 to 2010 through census six cities where the biggest concentrations of the African population data. Others like Allen et al. (2015), corroborating the sensitivity of are to be found in areas of lower average income, lower housing the indicator, apply a series of statistical models and adjustments in costs, and where infrastructure is meagre or in a state of disrepair. their calculations to overcome these limitations. In our case, the These zones are usually inner‐city areas and old working‐class correlation between the values obtained for age, sex, and group size neighbourhoods which were originally conceived for and inhabited by has been observed. The results are unequal. While this correlation families of the rural exodus that occurred in Spain in the mid‐20th occurs and is high in Barcelona and Valencia (r2 = 0.68 and 0.69), it century, or for specific urban family relocation plans. In Madrid, the exists but is less intense in Zaragoza and Madrid (r2 = 0.5 and 0.36) greatest concentrations of the African population appear in the city's but we cannot say the same about Seville and Malaga where the southern working‐class districts7 (Sorando & Leal, 2019). In Barcelona, 2 correlation is very weak (r = 0.19 and 0.14). At the aggregate level, besides working‐class areas,8 the highest representation is to be found in there is no correlation between group size and segregation. To avoid the historic centre, in the Ciutat Vella district, a first‐arrival area with a the effect of size on the calculation of the indicator, the DIs have high presence of immigrants (58% of the population of the district). In been estimated using a 95% confidence level.5 Valencia, their absence in the southern part of the city is perceived as coinciding with districts with the highest per capita incomes, while their presence is high in the working‐class neighbourhoods of the old city 4 | I M M I G R A T I O N I N SP A N I S H C I T I E S centre. Seville shows the greatest concentrations in the central districts, once again coinciding with historic or working‐class areas.9 In Zaragoza, The presence of immigrants varies greatly, both in volume and by contrast, the largest concentrations are in the old city and the composition, among the country's six largest cities. Hence, they periphery,10 where living conditions are worse, rents are lower, and range from maximum values of Barcelona (461,000 immigrants and infrastructure is more rundown than in other neighbourhoods, while they 27.8% of the population in 2020), Madrid (787,000 and 23.6%), are not represented in more recently established areas like La Romareda. and Valencia (152,000 and 19.1%), all with a percentage of Finally, in Malaga, the presence of Africans is again detectable in areas of immigrants that exceeds that for Spain as a whole (15.2%), through lower family incomes, although they also have a significant incidence in to 15.1% for Zaragoza (102,000 immigrants), to significantly lower central areas of higher purchasing power.11 values like 12.7% in Malaga (73,000) and barely 8.5% in Seville This distribution presents some differences according to the (58,000), the latter two being cities in the south of the country. characteristics of migrants by age. To give an example, in Barcelona, Again, these cities present different patterns in the reception of 36% of North Africans aged from 15 to 19 years reside in Ciutat Vella, immigrant flows in terms of their composition by origin. While in which is the district of first entry for many immigrants. Yet, this district Malaga, Zaragoza, and Seville, Africans have a significant presence shows lower values for the remaining age groups, for instance, 11.2% for in their settlements (28.5%, 19.9%, and 17.5% of their immigrant 10–14 years, and 18.3% for 35–39 years. A similar pattern with a populations respectively in 2020), in Madrid, Barcelona, and concentration of young people in Ciutat Vella is identified among sub‐ Valencia they have a much smaller presence (6.5%, 6.6%, and Saharan Africans. The existence of first‐stop places and others of later 9.5% respectively), which is well below their overall presence in settlement explains these differences, which justify the hypothesis that Spain (18%). Barcelona, however, is notable for the proportion of Asian immigrants (18% compared with just 7.2% in Spain), and Madrid for that of Latin American immigrants (66.6%, which is well 6 To map the presence of African immigrants in the six Spanish cities the Location Quotient above the 42.3% for the country as a whole) (Table 1). Meanwhile, (LQ) has been used. This relates the percentage of a population group in a statistical subunit with the percentage of the same group in the population as a whole (Brown & Chung, 2006). Zaragoza, Valencia, and Barcelona would be the cities with the It is formulated as follows: largest presence of European Union citizens (22.1%, 18%, and Xi /Ti LQ = X /T 16.4%), while, in Malaga, the presence of immigrants from elsewhere in Europe (11.6%) is notable. where Xi is the population of a group in spatial unit i; Ti the total population of the same unit; and X and T the population of the group analysed in the total of the area being studied and From a territorial standpoint (Figure 1), immigrants of African the total population. Values above 1 represent the existence of territorial concentration and origin show clear patterns in 2020 of spatial concentration in the those below 1 express a smaller representation than that for the whole. 7 Vallecas, Vicálvaro, Villaverde, and San Blas. 8 The districts of Nou Barris, Sant Martí, and Sants‐Montjuïc. 9 This is the case with the Macarena, Sur, Cerro, and Amate neighbourhoods. 5 The DI and its confidence intervals have been calculated using the R package ‘Segregation’ 10 Delicias, Oliver‐Valdefierro, and Miralbueno. 11 (Elbers, 2021). The confidence intervals can be consulted in the section, Supporting The areas with the greatest presence are in or near Bailen‐Miraflores, Cruz de Humilladero, Information, in the online version. Palma‐Palmilla, and Carretera de Cádiz. TABLE 1 Immigrant population by continental group and gender in the largest cities of Spain (2020). 6 of 15 Weight of Weight of | % Born immigrant % Born immigrant Total abroad population Men Women % Women Total abroad population Men Women % Women Barcelona Africa 30,372 6.6 1.8 18,350 12,022 39.6 Seville Africa 10,248 17.5 1.5 6004 4244 41.4 EU 75,716 16.4 4.5 37,602 38,114 50.3 EU 8,066 13.8 1.2 3705 4361 54.1 Rest of Europe 36,240 7.8 2.2 16,497 19,743 54.5 Rest of Europe 4,411 7.5 0.6 1708 2703 61.3 America 235,517 51.0 14.2 102,361 133,156 56.5 America 30,476 52.2 4.4 11,890 18,586 61.0 Asia 83,210 18.0 5.0 48,741 34,469 41.4 Asia 5,137 8.8 0.7 2572 2565 49.9 Total 461,960 100 27.8 223,551 237,504 51.5 Total 58,432 100 8.5 25,879 32,459 55.6 Total population 1,664,182 Total population 691,395 Weight of Weight of % Born immigrant % Born immigrant Total abroad population Men Women % Women Total abroad population Men Women % Women Madrid Africa 51,221 6.5 1.5 28,710 22,511 43.9 Zaragoza Africa 20,418 19.9 3.0 12,665 7753 38.0 EU 100,425 12.7 3.0 47,107 53,318 53.1 EU 22,717 22.1 3.3 10,681 12,036 53.0 Rest of europe 31,527 4.0 0.9 13,696 17,831 56.6 Rest of Europe 3,795 3.7 0.6 1623 2172 57.2 America 525,378 66.7 15.8 223,158 302,220 57.5 America 48,920 47.7 7.2 19,922 28,998 59.3 Asia 78,212 9.9 2.3 37,414 40,798 52.2 Asia 6,741 6.6 1.0 3631 3110 46.1 Total 787,765 100 23.6 350,085 436,678 55.5 Total 102,659 100 15.1 48,522 54,069 52.7 Total population 3,334,730 Total population 681,877 Weight of Weight of % Born immigrant % Born immigrant Total abroad population Men Women % Women Total abroad population Men Women % Women Valencia Africa 14,503 9.5 1.8 9041 5462 37.7 Malaga Africa 20,952 28.5 3.6 11,780 9172 43.8 EU 27,528 18.0 3.4 13,401 14,255 51.5 EU 11,185 15.2 1.9 5444 5741 51.3 Rest of Europe 11,801 7.7 1.5 5275 6526 55.3 Rest of Europe 8,527 11.6 1.5 3488 5039 59.1 America 80,176 52.5 10.0 34,589 45,587 56.9 America 27,911 37.9 4.8 11,595 16,316 58.5 Asia 18,443 12.1 2.3 10,858 7585 41.1 Asia 4,927 6.7 0.9 2580 2347 47.6 Total 152,627 100.0 19.1 73,164 79,415 52.0 Total 73,591 100 12.7 33,017 36,355 52.4 Total population 800,215 Total population 578,460 Source: Continuous Population Register (INE). GASTÓN‐GUIU and BAYONA‐I‐CARRASCO 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 7 of 15 Spatial distribution of African immigrants in the main Spanish cities, 2020. Source: Continuous Population Register (INE). | GASTÓN‐GUIU and BAYONA‐I‐CARRASCO FIGURE 1 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 of 15 | GASTÓN‐GUIU and BAYONA‐I‐CARRASCO there are different segregation patterns among ages and also between the 5.2 | The effect of demographic characteristics sexes. To analyse the effects of structure, the DI has been calculated disaggregated by sex and age and only for the last year, with data from 5 | RESULTS 2020. The data for each age and sex group are compared with data from the autochthonous population for the same age and sex. The DI shows a 5.1 | The temporal evolution of Dissimilarity Index common pattern among North Africans and sub‐Saharan Africans, (2000–2020) although it is more intense and higher for the latter group and shows more variations in the former group (Figure 3). The trend shown by the Evolutionary analysis of the DIs by population group (Figure 2) confirms line representing the total responds to high values among younger ages that immigrants of African origin show the highest segregation values by (15–24) and falls with the central ages coinciding with the working stage comparison with those obtained with Latin American and European (EU‐ of life but rises once again for mature ages. However, depending on the 15) immigrants, thus corroborating what has been noted in the scientific origin, oscillations in the indicator are more or less pronounced. Among literature on immigration in Spain. Nonetheless, among Africans and sub‐Saharan Africans, the increases and decreases in levels of segregation differentiating in accordance with geographical area of origin in the are considerably more noticeable than they are with other Africans, with continent, there are significant differences. Hence, immigrants of sub‐ rising levels appearing at earlier adult ages and always with indicators Saharan origin show much higher levels and the greatest gaps, these being moving in the higher figures. Among North Africans, however, growth is understood as distance with regard to other origins, a situation that is more progressive, except in the city of Malaga where the trend with rising more notable in the cities of the south of Spain (Seville and Malaga) and age tends to decrease. This particular behaviour can be explained on the less so in Barcelona and Zaragoza. In general, between 2000 and 2020, basis of two hypotheses: either there is a marked socioeconomic the levels of residential segregation for sub‐Saharan immigrants have differentiation between older immigrants and present ones (which we fallen, and even markedly in some cases. This occurred in Seville and do not know as we do not have the census data), or they are Spanish Malaga in the early years of this century while, in Madrid and Barcelona, citizens who were born in the old colonies that are now part of Morocco. the reduction is more continuous. Accordingly, from showing high or very At some ages, the low number of immigrants observed produces random high segregation values (even above 0.45), sub‐Saharan immigrants variations in the dissimilarity indicator. Evolution is marked by an evident presently have, in the two biggest cities, average segregation values size effect: the smaller the group analysed, the greater increase of (0.36 and 0.42, respectively, in 2020). segregation. But the levels obtained differ according to origin. This In the case of immigrants from North Africa, however, a tendency explains why segregation for the group as a whole is lower than presented toward increased segregation is detected in most of the cities observed. In by genders. Valencia, Seville, and Malaga, the evolution is similar to that for Latin As for gender, women generally show higher segregation Americans and Europeans, although always with slightly higher values. indexes, with the largest gaps among sub‐Saharan women of all However, in Barcelona and Zaragoza, they do not fit this trend shared with age groups and in all cities, although the gap widens with middle the other origins as their results are comparable with those obtained for age and ends up with an approximation to male values after the age sub‐Saharan Africans. As a result, and with the exception of Barcelona— of 60. However, among North Africans, there are no such distances where, however, the recent tendency is once again rising—the segregation as the lines, which are slightly higher for women, draw similar paths levels in 2020 are higher in the cities studied than they were in the early by gender, showing the greatest variations in the cities of 2000s. The negative economic situation partly explains this evolution Barcelona, Valencia, and Zaragoza after the age of 40, or even (Achebak et al., 2017), since, in general, the indicators rose midway earlier. In this case, segregation is comparable with that for men in through the second decade of the 21st century. the initial stages of the migratory process, although the level is Throughout the evolution one sees maintenance, more or less higher (about a 10th) in the central ages. sustained over time, of most of the values reported, although there are Another distinguishing feature with these two population groups is some exceptions. For example, in Barcelona, the falling values of indicators that, at mature ages, women from the north manage to reduce their are constant, except those for immigrants from the EU, and the increased indices to below those for men on some occasions while sub‐Saharan rates of segregation for North Africans since 2015. In Madrid, the trend is immigrants only manage to reduce theirs. Malaga is the main exception, very steady for all groups with little differentiation in values (hence, at with similar and even lower values for women of all ages. Consequently, present, they are all between 0.2 and 0.5 points, showing the smallest the results, which coincide with findings in the international literature, reported distances). The Madrid situation could be extrapolated to that of suggest that age has a cyclical effect, reducing segregation in the central Malaga (except for sub‐Saharan immigrants, who are highly segregated) ages of the life cycle, and affecting the young and the old more severely. and Valencia (with values between 0.2 and 0.4) but with figures for sub‐ Saharan immigrants that are well above those for other groups. Finally, in Zaragoza, since the economic crisis, all groups have subtly and 5.3 | Dissimilarity indicators by gender and cohort progressively shown rising results that have kept the same distances by origin from 2000 until 2020, while the lines in other cities have even Three cohorts have been selected for this final analysis, those born merged at certain times. from 1966 to 1970, from 1971 to 1975, and from 1976 to 1980. At 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GASTÓN‐GUIU and BAYONA‐I‐CARRASCO | 9 of 15 FIGURE 2 Evolution of the Dissimilarity Index (DI) of the main immigrant groups, 2000–2020. Source: Continuous Population Register (INE). the first observation point in 2000, these cohorts were aged 30–34, Figure 4 shows the trend in the DI from a dual perspective, 25–29, and 20–24, respectively, thus avoiding representation of the namely intra‐cohort and inter‐cohort, which allows a complete group aged from 15 to 19 years where the numbers are smaller and comparison. When each generation is taken individually, higher the indices more random. The selected cohorts have been observed segregation values have been obtained for the younger ages (in for two decades, until 2020 when, now 20 years older, their ages the 20–24‐year‐old group of the 1976–1980 generation) in all were situated, respectively, at 50–54, 45–49, and 40–44, this giving cities and for both origins, except for North Africans in Barcelona. 12 five moments of observation in the graph. The analysis has been It is true that the sub‐Saharan members of this age group start specifically for the two groups of Africans being studied. out with higher levels (around or above 0.8 points) than those for immigrants from North Africa, and also the distance with regard to the younger members of the other cohorts (25–29 and 30–34) is less significant than it is for North Africans among 12 A fictitious cohort model has been constructed, in which it is assumed that follow‐up is not whom the distances in terms of a declining indicator are more carried out with the same individuals since we do not have the information that would allow us to perform such an analysis. pronounced. 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10 of 15 | GASTÓN‐GUIU and BAYONA‐I‐CARRASCO F I G U R E 3 Dissimilarity Index (DI) by age, gender, and origin—North and sub‐Saharan Africa—in Spain, 2020. Source: Continuous Population Register (INE). F I G U R E 4 Dissimilarity Index (DI) by population group—North Africa and sub‐Saharan Africa—and Generation in Spain. Source: Continuous Population Register (INE). 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GASTÓN‐GUIU and BAYONA‐I‐CARRASCO | 11 of 15 One shared feature is decreasing indicators as the generations Hence, the new generations start with lower segregation values get older, this being most evident in the case of the youngest than those of older cohorts, a fact that could be due either to cohort since the rest are in the central ages of the life cycle which, greater heterogeneity in demographic profiles among the new as noted above, are associated with lessening segregation. cohorts, which would lead to greater dispersion or settling in a city Moreover, in the older generation, there appears a shift, though where the volume of compatriots is greater than it was in previous slight, towards an increase in the values obtained. The two groups generations. analysed follow this pattern, although the sub‐Saharan group while Finally, this cohort perspective is expanded by adding the showing these falling rates, are always situated above the North gender variable, which is represented in the last two graphs Africans. Hence, after beginning with high values, segregation (Figures 5 and 6). Besides reproducing the trends discussed for all stabilises and even increases with age within the same generation, immigrants, among the North Africans there is a smoothing of but the increase is less than what was envisaged from the differences between men and women in the youngest generation, standpoint of age. The trend of increasing segregation with age while in the more mature generations, the women are more is therefore confirmed. segregated than the men. This could be explained by the particular However, if a reading is made of the inter‐cohort figures and masculinisation of African migratory flows and a possible opening they are compared with the same age (starting from 30 to up of women's migratory projects apart from family reunification 34 years), more recent generations report lower segregation processes. values than those of the other cohorts at the same age. This is This situation does not occur in the sub‐Saharan case, except especially evident among the sub‐Saharan Africans where the in the cities of Valencia and Zaragoza, where the results for the largest gaps are to be found between the 1976–1980 and younger generation of women is on a par with the male results. 1966–1970 cohorts at 30–34 years, but it is also the case for However, among the women of older generations, the segregation North Africans. For both North and sub‐Saharan Africans, this values are always much higher than those for men. From this point falling trend by generation is nuanced by city. For the former of view, we find that segregation values would therefore be higher group, this very clearly occurs in Barcelona while, in the rest of the among the cohorts that are pioneers of the migratory process, cities, instead of a decrease, stabilisation according to age is while the new cohorts arriving in a context with a larger presence observed, or even an increase in Madrid and Malaga which, for this of immigrants show greater dispersion and, accordingly, less group, functions differently from the other cities, as noted above. segregation. F I G U R E 5 Dissimilarity Index (DI) by population group—North Africa—gender and Generation in Spain. Source: Continuous Population Register (INE). 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 12 of 15 | GASTÓN‐GUIU and BAYONA‐I‐CARRASCO F I G U R E 6 Dissimilarity Index (DI) by population group—sub‐Saharan Africa—gender and Generation in Spain. Source: Continuous Population Register (INE). 6 | DISC US SION A ND C ONCLUSIONS groups. From the results obtained in the present study, evolu- tionary analysis of the segregation index indicates that, indeed, The main conclusions obtained after calculating the segregation this population group is more segregated than other immigrants, indices and reading the results are presented below. This section is and the sub‐Saharan Africans are the most affected. Unlike Latin organised into three sections that refer to the structure in which the American immigrants, whose patterns are very similar in the six results have been presented. cities, both North and sub‐Saharan Africans show significant particularities in the intensity and evolution of segregation, depending on the city in question. Consequently, the character- 6.1 | Segregation of Africans (in relation with other istics of each city represent an important element in the origins) and the period effect segregation of Africans, which complicates comparative analysis. Nevertheless, there is a generalised tendency to show higher Despite the fact that the reception of the immigrant population in levels of segregation than those of the other main origins over the Spain occurred late, its peculiarities in terms of high levels of 20 years studied. These levels, moreover, have increased in all six intensity and heterogeneity mean that its study is a far‐reaching cities for North Africans since, in the only case that has shown exercise in terms of both discovering its effects on the Spanish falling values, Barcelona, segregation has increased again in social and population structures and in representing a particular recent years (coinciding with economic recovery and greater case of considerable interest in the international context. Analysis pressure on the real‐estate market). The significance of this of residential segregation plays a decisive role here since it makes evolution lies in the fact that this group, one of the oldest and it possible to study the ways in which individuals coexist and are most predominant in Spain, is also subject to more severe distributed over geographic space and has taken on great stigmatisation of its presence. importance following the growth and dynamisation of the Spanish As for temporal evolution, the economic crisis represents a population. change in the evolution of some values of the indicator, as The existing literature on Spain highlights the impact of origin does the incipient recovery. We, therefore, find period effects on in the greater or lesser incidence of segregation, concluding that the trend of the indicator, although not always in the same Africans constitute one of the most segregated population direction. 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GASTÓN‐GUIU and BAYONA‐I‐CARRASCO | 13 of 15 6.2 | Existence of clear patterns of segregation by significant. The generational perspective suggests that the high age and gender differences differences observed by age could disappear in future, bearing in mind that this is a relatively novel phenomenon with an age structure With analysis of the impact of demographic characteristics, it is that is still relatively young. observed that age presents a clear shared pattern with higher At the same time, when comparing cohorts, only in some segregation at younger and older ages, and lower segregation in the cases, for example, the city of Barcelona, is there a significant middle years of life. In this case, origin also determines incidence with decrease in segregation between generations which contrasts with the highest values for sub‐Saharan Africans and evident differences the pattern of stability in other cities. Meanwhile, increased between the cities. The results for the oldest and youngest ages segregation in some cities like Seville is not visible from the could be magnified by the effect of the group size on the indicator generational perspective. For sub‐Saharan Africans, the decrease since among the youngest and oldest the numbers of immigrants are in segregation is much clearer when viewed from the generational lower. However, the persistence of the age patterns confirms this perspective. The younger members of this group start with lower distribution. segregation values, which will lead to a continued decline of In the case of gender, women show higher segregation indices, segregation for this group in future. with the greatest gaps among sub‐Saharan Africans. However, at In light of the results, it can be concluded that the three classical more mature ages, gender differences diminish or even reverse, perspectives of demographic analysis offer a richer interpretation of which is what occurs among North Africans. These differences are the evolution of segregation. Hence, analysis of segregation should not due to an effect of group size and can be explained by household include not only traditional variables like origin but it should also be strategies in family reunification. combined with other demographic variables like age and gender, the To sum up, gender and age play a determining role in the greater moment of observation, and the cohort of belonging, which, as or lesser intensity of the values associated with segregation. Being a shown above, influence the results. This gives a more complex view woman implies being more segregated. Age, however, has a cyclical and enables a complete understanding of the phenomenon of effect that reduces the segregational impact in middle ages, and segregation and, in particular, of the mechanisms that explain the increases it in the early and later years of life. evolution in one direction or another of the values of segregation. The results achieved can be investigated in greater depth through new lines of research, addressing some of the limitations 6.3 | The generational perspective of the present work. On the one hand, by homogenising the units of analysis through the construction of a Grid, which will facilitate In the analysis by cohorts, sub‐Saharan Africans are, once again, more comparison between cities. On the other hand, by using multigroup segregated than North Africans. This is particularly true among the segregation indicators that take into account the interaction of the older generations while the gap narrows slightly among the younger group analysed with other immigrant groups, especially in contexts ones. In terms of age profile, North Africans show higher levels of such as Spain's, which is marked by high diversity. Third, by segregation at the youngest and oldest ages in all generations. Here, redefining city limits using a metropolitan approach, especially for gender also presents peculiarities where the figures for women are the cases of Barcelona and Madrid, and finally, with the incorporation higher, especially at the older ages of each generation. However, this of data from the new 2021 Population Census, which will help to effect could result from the classic masculinisation of African differentiate residential patterns based on the characteristics of migratory flows and the subsequent processes of family reunification migrants and territories. in which women would be predominant. Thus, in a single cohort the values would tend to rise with age, although the increase would be ACKNOWL EDGEM ENTS much less than that observed from the age perspective. In spite of This paper is part of the Project ‘Demographic Metabolism, Migration this, the trend is confirmed, a fact that can be explained by two and Social Change in Spain’ (PID 2020‐113730RB‐I00), funded by the hypotheses: either there is a selection effect, where those with more Ministry of Economy and Competitiveness. Jordi Bayona‐i‐Carrasco segregated locations continue to be found in the city, or there is a is a Serra‐Húnter Fellow at the Geography Department of the tendency toward concentration in internal mobility, aspects which University of Barcelona. The authors are grateful for the comments are beyond the scope of the present research. received, all of which contributed to improvements in the text. Hence, and by way of synthesis, it could be concluded that, at the ‘inter‐cohort’ level, as age increases and birth cohorts grow older, CONFLIC T OF INTEREST STATEM ENT the gender differences become more pronounced, while age There is no conflict of interest. attenuates them as they increase within each generation (intra‐ cohort). DATA AVAILABILITY STATEMENT If, however, the graphs are viewed from a comparative The data that support the findings of this study are available from perspective, the high degree of stability observed within a single Instituto Nacional de Estadística (INE). Restrictions apply to the generation, contrasting with divergences found according to age, is availability of these data, which were used under license for this 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 14 of 15 | GASTÓN‐GUIU and BAYONA‐I‐CARRASCO study. Data are available from the author(s) with the permission of Coulter, R., & Clark, W. A. V. (2019). Ethnic disparities in neighbourhood Instituto Nacional de Estadística (INE). selection: Understanding the role of income. International Journal of Urban and Regional Research, 43, 947–962. https://doi.org/10.1111/ 1468-2427.12697 ORCID Cowgill, D. O. (1978). Residential segregation by age in American Silvia Gastón‐Guiu http://orcid.org/0000-0001-8497-3579 metropolitan areas. Journal of Gerontology, 33(3), 446–453. Jordi Bayona‐i‐Carrasco http://orcid.org/0000-0003-2819-9085 https://doi.org/10.1093/geronj/33.3.446 Cutler, D. M., & Glaeser, E. L. (1997). Are ghettos good or bad? The Quarterly Journal of Economics, 112(3), 827–872. https://doi.org/10. REFERENCES 1162/003355397555361 Achebak, H., Bayona‐I‐Carrasco, J., & Domingo i Valls, A. (2017). Debiasi, E. (2018). Age‐period‐cohort analysis: A summary of analytical Evolución y pautas geográficas de la segregación residencial de los approaches and results (No. 676060). Lund University. marroquíes en España. Estudios Geográficos, 78(283), 417–443. Deng, G., & Mao, L. (2018). Spatially explicit age segregation index and https://doi.org/10.3989/estgeogr.201714 self‐rated health of older adults in US cities. International Journal of Allen, R., Burgess, S., Davidson, R., & Windmeijer, F. (2015). More reliable Geo‐Information, 7(9), 351. https://doi.org/10.3390/ijgi7090351 inference for the dissimilarity index of segregation. The Econometrics Domingo, A., & Bayona, J. (2007). Movilidad, vivienda y distribución Journal, 18(1), 40–66. https://doi.org/10.1111/ectj.12039 territorial de la población marroquí en Cataluña. Estudios Geograficos, Arbaci, S. (2019). Paradoxes of segregation: Housing systems, welfare 68(263), 465–496. https://doi.org/10.3989/egeogr.2007.i263.65 regimes and ethnic residential change in southern European cities. Domínguez‐Mujica, J., Parreño, M., & Díaz, R. (2010). Inmigración y ciudad en Wiley Blackwell. España: Integración versus segregación socio‐territoriales. Scripta Nova, Arbaci, S., & Malheiros, J. (2009). De‐segregation, peripheralisation and XIV(331). http://www.ub.es/geocrit/sn/sn-331/sn-331-50.htm the social exclusion of immigrants: Southern European cities in the Droogleever Fortuijn, J., Musterd, S., & Ostendorf, W. (1998). Interna- 1990s. Journal of Ethnic and Migration Studies, 36(2), 227–255. tional migration and ethnic segregation: Impacts on urban areas. https://doi.org/10.1080/13691830903387378 Urban Studies, 35(3), 367–370. Bayona‐i‐Carrasco, J. (2007). La segregación residencial de la población Duncan, O. D., & Duncan, B. (1955). A methodological analysis of extranjera en Barcelona: ¿una segregación fragmentada? Scripta segregation indexes. American Sociological Review, 20, 210–217. Nova, XI(235). http://www.ub.es/geocrit/sn/sn-235.htm Elbers, B. (2021). A method for studying differences in segregation across Bayona‐i‐Carrasco, J., & López‐Gay, A. (2011). Concentración, segrega- time and space. Sociological Methods & Research, 52(1), 5–42. ción y movilidad residencial de los extranjeros en Barcelona. https://doi.org/10.1177/0049124121986204 Documents d'Anàlisi Geogràfica, 57(3), 381–412. https://doi.org/10. Finney, N., & Simpson, L. (2009). Population dynamics: The roles of natural 5565/rev/dag.234 change and migration in producing the ethnic mosaic. Journal of Benassi, F., & Iglesias‐Pascual, R. (2022). Local‐scale residential concen- Ethnic and Migration Studies, 35(9), 1479–1496. https://doi.org/10. tration and income inequalities of the main foreign‐born population 1080/13691830903125935 groups in the Spanish urban space: Reaffirming the model of a Galeano, J., & Sabater, A. (2016). Segregación residencial y formación de divided city. Journal of Ethnic and Migration Studies, 49, 673–696. enclaves residenciales en España 2000‐2014. In A. Domingo (Ed.), https://doi.org/10.1080/1369183X.2022.2067137 Inmigración y diversidad en España. Crisis económica y gestión Benassi, F., Iglesias‐Pascual, R., & Salvati, L. (2020). Residential segrega- municipal (pp. 13–48). Icaria. tion and social diversification: Exploring spatial settlement patterns Gastón‐Guiu, S., Treviño, R., & Domingo, A. (2021). La brecha africana: of foreign population in Southern European cities. Habitat Desigualdad laboral de la inmigración marroquí y subsahariana en International, 101, 10220. https://doi.org/10.1016/j.habitatint. España, 2000–2018. Migraciones, 52, 177–220. https://doi.org/10. 2020.102200 14422/mig.i52.y2021.007 Berriane, M. (2004). Regiones institucionales y focos migratorios. In B. Giele, J., & Elder, G. (1998). Life course research: Development of a field. López‐García & M. Berriane (Eds.), Atlas de la inmigración marroquí en In J. Giele & G. Elder (Eds.), Methods of life course research: España (pp. 125–128). UAM ediciones. Qualitative and quantitative approaches. Sage Publications. Bosch, M., Carnero, M. A., & Farré, L. (2015). Rental housing discrimina- González, J. R., Llorca, F. J., & Moreno, V. (2002). Some methodological tion and the persistence of ethnic enclaves. Series, 6, 129–152. aspects of age‐period‐cohort models. Application to cancer mortal- https://doi.org/10.1007/s13209-015-0122-5 ity trends. Gaceta Sanitaria, 16(3), 267–273. https://doi.org/10. Brown, L. A., & Chung, S.‐Y. (2006). Spatial segregation, segregation 1016/s0213-9111(02)71672-5 indices and the geographical perspective. Population, Space and Halli, S., & Rao, K. (1992). Advanced techniques of population analysis. Place, 12, 125–143. https://doi.org/10.1002/psp.403 Plenum Press. Cachón, L. (2009). La España inmigrante: Marco discriminatorio, mercado de Iglesias‐Pascual, R., Benassi, F., & Hurtado‐Rodríguez, C. (2023). Social trabajo y política de integración. Anthropos. infrastructures and socio‐economic vulnerability: A socio‐territorial Cea d′Ancona, Ma. Á. (2002). La medición de las actitudes ante la integration study in Spanish urban contexts. Cities, 132, 104109. inmigración: evaluación de los indicadores tradicionales de “racis- https://doi.org/10.1016/j.cities.2022.104109 mo”. Revista Española de Investigaciones Sociológicas, 99, 87–111. Kauppinen, T. M., & van Ham, M. (2019). Unravelling the demographic https://doi.org/10.2307/40184401 dynamics of ethnic residential segregation. Population, Space and Checa, J. C., Arjona, Á., & Checa, F. (2011). Segregación residencial de la Place, 25, e2193. https://doi.org/10.1002/psp.2193 población extranjera en Andalucía. España. Papeles de Población, Leal, J., & Alguacil, A. (2012). Vivienda e inmigración: las condiciones y el 17(70), 219–246. comportamiento residencial de los inmigrantes en España. Anuario Cohen, A. (2009). España en la encrucijada migratoria (trans)‐ CIDOB de la Inmigración, 2012, 126–156. https://raco.cat/index. mediterránea. Una revisión sociogeográfica. Cahiers de civilisation php/AnuarioCIDOBInmigracion/article/view/354080 espagnole contemporaine, 4. https://doi.org/10.4000/ccec.2718 Leal, J., & Sorando, D. (2022). Social mix and vertical segregation in Madrid. In Colectivo IOE. (2005). Inmigración y vivienda en España. Ministerio de T. Maloutas (Ed.), Vertical cities. Micro‐segregation, social mix and urban Trabajo y Asuntos Sociales. housing markets (pp. 116–128). Edward Elgar Publishing. 15448452, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/psp.2675 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GASTÓN‐GUIU and BAYONA‐I‐CARRASCO | 15 of 15 Martori, J. C., & Apparicio, P. (2011). Changes in spatial patterns of the d'Anàlisi Geogràfica, 66(1), 83–105. https://doi.org/10.5565/ immigrant population of a southern European metropolis: The case rev/dag.581 of the Barcelona metropolitan area (2001–2008): Changes in spatial Sabater, A. (2010). Ethnic residential segregation over time and cohorts in patterns of immigrant populations. Tijdschriftvoor Economische en England and Wales, 1991–2001. In J. van Stillwell & M. van Ham Sociale Geografie, 102(5), 562–581. https://doi.org/10.1111/j.1467- (Eds.), Ethnicity and integration: Understanding population trends and 9663.2011.00658.x processes (pp. 47–62). Martori, J. C., & Madariaga, R. (2023). Residential segregation by Sabater, A., & Catney, G. (2019). Unpacking summary measures of ethnic nationalities: A global and multilevel approach to Barcelona and residential segregation using an age group and age cohort perspec- Madrid (2008–2018). Population, Space and Place, e2655. https:// tive. European Journal of Population, 35(1), 161–189. https://doi.org/ doi.org/10.1002/psp.2655 10.1007/s10680-018-9475-3 Massey, D. S., & Denton, N. A. (1988). The dimensions of residential Sabater, A., Domingo, A., & Galeano, J. (2013). La transformación de las segregation. Social Forces, 67, 281–315. https://doi.org/10.1093/sf/ comunidades mayoritarias y la formación y evolución de enclaves 67.2.281 étnicos en España. Migraciones, 34, 11–44. Massey, D. S., Gross, A. B., & Shibuya, K. (1994). Migration, segregation Schelling, T. C. (1969). Models of segregation. American Economic Review, and the geographic concentration of poverty. American Sociological 59(2), 488–493. Review, 59, 425–445. Schelling, T. C. (1971). Dynamic models of segregation. The Journal of Massey, D. S., Rothwell, J., & Domina, T. (2009). The changing bases of Mathematical Sociology, 1(2), 143–186. https://doi.org/10.1080/ segregation in the United States. The Annals of the American 0022250X.1971.9989794 Academy of Political and Social Science, 626(1), 74–90. https://doi. Settersten, R. (1999). Lives in time and place: The problems and promises of org/10.1177/0002716209343558 developmental science. Baywood Publishing Co. McKenzie, R. D. (1924). The ecological approach to the study of the Simpson, L. (2004). Statistics of racial segregation: Measures, evidence human community. American Journal of Sociology, 30(3), 287–301. and policy. Urban Studies, 41(3), 661–681. https://doi.org/10.1080/ https://doi.org/10.1086/213698 0042098042000178735 Musterd, S. (2003). Segregation and integration: A contested relationship. Sorando, D., & Leal, J. (2019). Distantes y desiguales: El declive de la Journal of Ethnic and Migration Studies, 29(4), 623–641. https://doi. mezcla social en Barcelona y Madrid. Revista Española de org/10.1080/1369183032000123422 Investigaciones Sociológicas, 167, 125–148. http://doi.org/10.5477/ Musterd, S., & Fullaondo, A. (2008). Ethnic segregation and the housing cis/reis.167.125 market in two cities in Northern and Southern Europe: The cases of Sullivan, J. W. (2020). Structure and segregation: The importance of age Amsterdam and Barcelona. Arquitectura, Ciudad y Entorno, 3(8), structure in households, neighborhoods and the population for 93–115. residential patterns in the United States, 1940–2015. UCLA. Musterd, S., Kesteloot, C., Murie, A., & Ostendorf, W. (1999). Urban Uhlenberg, P. (2000). Introduction. The Gerontologist, 40(3), 261–266. social exclusion and modes of integration: Literature review. https://doi.org/10.1093/geront/40.3.261 Urbex Series, Vol. 1. Fourth RTD Framework Programme, TSER, Van Kempen, R., & Sule Özüekren, A. (1998). Ethnic segregation in cities: AME, Amsterdam. New forms and explanations in a dynamic world. Urban Studies, Napierala, J., & Denton, N. (2017). Measuring residential segregation with 35(10), 1631–1656. https://doi.org/10.1080/0042098984088 the ACS: How the margin of error affects the dissimilarity index. Vono, D., & Bayona‐i‐Carrasco, J. (2010). Características territoriales del Demography, 54(1), 285–309. https://doi.org/10.1007/s13524- asentamiento de los latinoamericanos en las principales ciudades 016-0545-z españolas entre 2001 y 2009. Notas de Población, 91, 129–159. Okraku, I. O. (2008). Age residential segregation in Canadian cities. Wong, D. W. S. (1997). Spatial dependency of segregation indices. Canadian Review of Sociology/Revue Canadienne de Sociologie, Canadian Geographer/Le Géographe Canadien, 41, 128–136. https:// 24(3), 431–452. https://doi.org/10.1111/j.1755-618X.1987. doi.org/10.1111/j.1541-0064.1997.tb01153.x tb01105.x Wong, D. W. S., & Das Gupta, D. (2023). Age‐race‐ethnicity segregation in Peach, C. (2005). The ghetto and the ethnic enclave. In D. P. Varady (Ed.), the United States: Where do minority older adults stand? Population, Desegregating the city: Ghettos, enclaves and inequality (pp. 31–48). Space and Place, 29, e2642. https://onlinelibrary.wiley.com/doi/10. State University of New York Press. 1002/psp.2642 Portes, A., & Zhou, M. (1993). The new second generation: Segmented Yang, Y., & Land, K. C. (2013). Age‐period‐cohort analysis: New models, assimilation and its variants. The Annals of the American Academy of methods, and empirical applications. CRC Press. Political and Social Science, 530(1), 74–96. https://doi.org/10.1177/ 0002716293530001006 Portrait, F., Rob, A., & Deeg, D. (2002). Disentangling the age, period, and SUPP ORTING INFO RM ATION cohort effects using a modeling approach (120/3). Tinbergen Institute. Prieto, V., Recaño, J., & Quintero‐Lesmes, D. (2018). Migration responses Additional supporting information can be found online in the of immigrants in Spain during the Great Recession. Demographic Supporting Information section at the end of this article. Research, 38, 1885–1932. https://doi.org/10.4054/DemRes.2018. 38.61 Ramalle, E. (2009). Modelos de edad, periodo y cohorte en epidemiología social Su aplicación a un caso riojano. Berceo, 156, 7–27. How to cite this article: Gastón‐Guiu, S., & Bayona‐i‐ Rhein, C. (1988). La ségrégation démographique. In C. Bonvalet & P. Carrasco, J. (2023). Age, gender, and cohort in residential Merlin (Eds.), Transformation de la famille et habitat. INED. segregation: The case of African immigrants in Spain, Riley, M. W., & Riley, J. W. (2000). Age integration. The Gerontologist, 40(3), 266–270. https://doi.org/10.1093/geront/40.3.266 2000–2020. Population, Space and Place, 29, e2675. Rubiales Pérez, M. (2020). Segregación en las metrópolis españolas https://doi.org/10.1002/psp.2675 2001–2011: Un análisis con detalle territorial. Documents