The ‘Integration’ of People of Dutch Descent in Superdiverse Neighbourhoods

In the past 40 years, researchers into migration and integration have focussed almost exclusively on migrants and their children. This one-sided focus has persisted, even though it is generally acknowledged that integration is a two-way process in which not only migrants, but also the established population play an important role (see, for example, Garces-Mascareňas and Penninx 2016; Martinovic 2013). Amsterdam and Rotterdam have both become majority-minority cities where now all ethnic population groups – including that of Dutch descent – form a numerical minority. This new reality makes it even more urgent to examine the group without a migration background. In cities like Amsterdam and Rotterdam, only one out of three young people under the age of 15 are of Dutch descent (Crul 2016). The transformation of the former majority group to a numerical minority group may well be one of the most significant urban transformations of our time. The current backlash against migrants and refugees has made it clear that the integration of people of Dutch descent into today’s superdiverse majority-minority neighbourhoods and cities is a topic that begs attention.


Introduction
In the past 40 years, researchers into migration and integration have focussed almost exclusively on migrants and their children. This one-sided focus has persisted, even though it is generally acknowledged that integration is a two-way process in which not only migrants, but also the established population play an important role (see, for example, Garces-Mascareňas and Penninx 2016;Martinovic 2013). Amsterdam and Rotterdam have both become majority-minority cities where now all ethnic population groups -including that of Dutch descent -form a numerical minority. This new reality makes it even more urgent to examine the group without a migration background. In cities like Amsterdam and Rotterdam, only one out of three young people under the age of 15 are of Dutch descent (Crul 2016). The transformation of the former majority group to a numerical minority group may well be one of the most significant urban transformations of our time. The current backlash against migrants and refugees has made it clear that the integration of people of Dutch descent into today's superdiverse majority-minority neighbourhoods and cities is a topic that begs attention.
All kinds of commonly used integration indicators show alarming outcomes for people of Dutch descent. In the political domain, there has been a sharp increase in support for anti-immigration parties, an indication of dissatisfaction with the increase in diversity. When it comes to social cohesion, we see that many people of Dutch descent are retreating into 'white' enclaves, for example, sending their children to schools outside their own neighbourhoods to avoid sending them to schools with a high concentration of migrant children ('white flight'). When we look at social contacts we see that people of Dutch descent form the group in a city with the least social contact with people from outside their own ethnic group M. Crul (*) · F. Lelie Free University Amsterdam, Amsterdam, The Netherlands e-mail: m.r.j.crul@vu.nl (Crul et al. 2012). If Dutch cities are compared, Amsterdam is often regarded as the example of a city where diversity is embraced by the majority of its population, while Rotterdam, on the other hand, is described as a more reluctant superdiverse city. Our research for this chapter confirms that people of Dutch descent in Rotterdam draw sharper boundaries between themselves and people of other ethnic backgrounds than their peers in Amsterdam do. The presented research, based on a local survey among people of Dutch descent living in superdiverse neighbourhoods, shows that these people in Rotterdam experience diversity more often as threatening. Why is it that in Rotterdam they perceive diversity more often as threatening than people of Dutch descent in Amsterdam? What are the underlying mechanism? It is striking that the contrast between Rotterdam and Amsterdam is particularly evident among people employed in the middle echelons of the labour market. What is causing this difference? In both cities, we see that people employed in the creative sector and those working in law enforcement occupations such as the police, army and security services are characterized by a stabile attitude towards ethnic diversity. The cities' general climate seems to exert a greater influence -both positively and negativelyon people working in administrative, technical, financial and social professions, as this is where we find less stable attitudes towards diversity.

Theoretical Framework: Integration Matrix for Superdiverse Majority-Minority Neighbourhoods
Steve Vertovec introduced the concept of superdiversity in his famous article Superdiversity and its implications (Vertovec 2007). Vertovec used this concept to try to capture a new reality emerging in large cities where a multiplicity of ethnic and religious groups, different age groups and socio-economic groups live alongside each other at neighbourhood level. He does not, however, provide a clear definition of what constitutes a superdiverse city or neighbourhood. It is important to us to provide a more precise definition of what we actually mean when we talk about a superdiverse neighbourhood and when it would be better to use a different label, such as a segregated neighbourhood or a neighbourhood with a certain degree of ethnic or socio-economic diversity (Crul et al. 2013;Crul 2016). We think that the label superdiversity is particularly valuable for describing neighbourhoods in which there is no longer a numerical ethnic majority group. In short, neighbourhoods in which all groups numerically are minority groups. In concrete terms, this means a neighbourhood where half or more of the inhabitants are not of Dutch descent. It is also useful to use this term if there is a substantial number of minority groups and not just two or three groups living in 'ethnic enclaves'. This is actually the case in most majority-minority neighbourhoods in Dutch cities. Even in Slotervaart, de Baarsjes and Bos en Lommer -three neighbourhoods in Amsterdam that are perceived as being dominated by people of Moroccan and Turkish descent respectively -these population groups do not form a majority by a long chalk. In Slotervaart only 18% of the population is of Moroccan descent, while in de Baarsjes and Bos en Lommer, 19% of the population is of Turkish descent (Wonderen and Broekhuizen 2012a, b). In superdiverse situations (in the sense of the word as we use it) the integration process is, understandably, different than in traditional assimilation or integration situations. Migrants and their children no longer integrate solely into the majority group, but into an amalgam of groups. People of Dutch descent also have to integrate into a superdiverse context in which they are no longer the dominant ethnic group: at least not in numerical terms. Traditional assimilation theories are based on the idea that the differences between groups will diminish over time, resulting in what Richard Alba (2009) described as the 'blurring of ethnic boundaries' and the 'opening of the mainstream to new groups'. Increasing support for anti-immigrant parties, however, shows that another scenario is also possible: one in which the former majority group draws increasingly sharp boundaries between their own group and other ethnic minority groups. The people doing so feel threatened by the increasing diversity in the cities (for comparable studies, see analyses in the United States: Craig and Richeson 2014;Danbold and Huo 2015). This could trigger a negative dynamic that would cause integration to stagnate. If the former majority group avoids contact, it may be more difficult for new groups to integrate. This could result in groups growing further and further apart instead of ethnic boundaries blurring over time as predicted by traditional assimilation theories.
Much research into ethnically diverse neighbourhoods shows that inhabitants without a migration history often have little or no contact with inhabitants in their neighbourhoods who do have a migration history (Beckhoven and Kempen 2003;Blokland and Van Dijk 2010;Butler 2003;Butler and Robson 2001;Jackson and Benson 2014;May 1996;Tach 2009). Proximity does not, therefore, automatically lead to more contact, more mixed networks, or -something even more ambitious -the transfer of social or cultural capital. These outcomes often lead researchers to the conclusion that mixed neighbourhoods do not contribute to social cohesion and integration. Some authors even stress the negative consequences of living together in mixed neighbourhoods. Competition arises when scarce neighbourhood facilities are being claimed and the dominant socio-economic group determines what is and is not the norm in the public space. Situations in which people have not made an active choice to live alongside people from diverse ethnic backgrounds and residents have to share important facilities such as a staircase, lift or courtyard garden may lead to conflicts and the sharpening of ethnic divides (Tersteeg and Pinkster 2016).
However, for our comparison, it is also important to look at explanations for why there is a group that does perceive living together in mixed neighbourhoods as an enriching experience. This does not necessarily have to fit within a scenario of increasingly blurred ethnic dividing lines, as it is also possible for people to live alongside each other contentedly without any or little personal contact. In her study of the London majority-minority district of Hackney, Wessendorf (2014) talks about commonplace diversity, whereby people consider diversity as a given fact or the norm in their everyday contacts (also see Schneider and Lang 2014). This does not, however, mean that they maintain personal contact across ethnic dividing lines. In his ethnographical study of his own neighbourhood in Antwerp, Blommaert (2013) uses the term conviviality for a kind of live-and-let-live attitude among residents of superdiverse neighbourhoods. Blokland and Nast (2014) describe this as occupying the space somewhere between the myth of the urban village (Gans 1962) on the one hand, and Simmel's (1903) anonymous city on the other hand.
Wise and Noble (2016) stress that the term conviviality should be used to both study potential positive as well as possible negative effects of living together, and, also, the simultaneity of both. They argue that in order to study the complexity of living together we also need to look at the ambivalent. Blokland and Nast (2014) introduced the idea of comfort zones in which people know what they can expect from interactions in neighbourhood situations without adopting other people's behaviour or even approving of it. If interactions fit within the routine (public familiarity) that characterizes a neighbourhood, they promote a feeling of familiarity, predictability, control and safety. Blokland and Nast (idem) emphasize the importance of fleeting encounters in the neighbourhood for making residents feel comfortable, safe and at home. They have described this as an undervalued aspect in the discussion on mixed neighbourhoods.
The most important theoretical notion that we take from the literature referred to in this article is the idea of ambivalence and the acknowledgment to both positive and negative effects of living together. One important element in this ambivalence that we will look at is the observable gap between people's attitudes towards diversity on the one hand and their practises on the other. People may, for instance, regard living together in an ethnically mixed neighborhood as enriching, but in practice not have an interethnic friendship group. Based on these observations we have drawn up a super-diversity matrix in which four outcomes are possible: (A) little interethnic contact and sees diversity as a threat; (B) much interethnic contact but sees diversity as a threat. (C) much interethnic contact and sees diversity as enriching; (D) little interethnic contact but sees diversity as enriching.
On the basis of some of the literature cited above (Beckhoven and Kempen 2003;Blokland and Van Dijk 2010;Butler 2003;Jackson and Benson 2014;May 1996;Tach 2009), we may expect to find a lot of people of Dutch descent who have little contact outside their own ethnic group in the typology. Some of them will nevertheless see diversity as enriching, while others will perceive it as a threat.

Starting Point: Differences Between Rotterdam and Amsterdam
The starting point for the empirical part of this book chapter are people of Dutch descent living in Rotterdam. In the most recent local elections, the anti-immigrant party Leefbaar Rotterdam received 28% of the votes, while in the general election of 2017, the PVV received 16% of the votes. More than one in three of eligible voters of Dutch descent voted for Leefbaar Rotterdam. At the very least, this seems to express a degree of uneasiness among people of Dutch descent living in Rotterdam concerning the theme of diversity in the city. In contrast to Rotterdam, Leefbaar Amsterdam just 7% of the voters cast in the 2017 general election voted for Geert Wilders' PVV. The stark contrast between these cities demonstrates the different political reactions to increasing diversity. This fact is our starting point for the comparison in this book chapter. How can the differences between these cities be explained and what do they mean for the atmosphere and political climate in the city and contact between different ethnic groups? There are major differences regarding the way in which people of Dutch descent either embrace or reject increasing diversity in their living and work environment in each neighbourhood and city. In this chapter, we will examine why some people of Dutch descent feel threatened in superdiverse majority-minority neighbourhoods, while other people from the same group perceive diversity as enriching.

Methodological Approach: TIES Data
For the empirical part of this chapter we used data from the TIES study. TIES stands for 'The Integration of the European Second generation'. During the TIES survey in 2008, research was conducted in 15 cities in 8 countries, involving almost 10,000 respondents between the ages of 18 and 35. In each city, 250 respondents from each target group were interviewed. In the Netherlands, the interviews were conducted in Amsterdam and Rotterdam. The Municipal Records Database (for the sample frame, see Groenewold 2008) was used to obtain the samples. As indicated by the project's acronym, second-generation respondents were the main focus of the research, with the respondents of Dutch descent functioning as a control group. Using the data from the Municipal Records Database, we approached secondgeneration respondents in proportion to their distribution across neighbourhoods in Amsterdam and Rotterdam. The respondents of Dutch descent were approached in the same neighbourhoods in the same proportion. This means that the neighbourhoods in which the respondents of Dutch descent were interviewed are predominantly majority-minority neighbourhoods because these are the neighbourhoods where the young people of the second generation live. Both Rotterdam and Amsterdam are majority-minority cities at the level of the entire city, so the fact that the young people of the second generation live in majority-minority neighbourhoods is not remarkable. Out of the total TIES sample, 79% of the people of Dutch descent who were interviewed lived in such a majority-minority neighbourhood. Only these respondents (381) were selected for the analysis for this chapter.
The majority-minority neighbourhoods where we selected respondents are, without exception, neighbourhoods in which despite their over-representation, the Turkish-Dutch and Moroccan-Dutch inhabitants are just one of the many minority groups. Nowhere did the individual share of Turkish-Dutch and Moroccan-Dutch residents exceed 20% of the inhabitants. These neighbourhoods can be characterized as neighbourhoods with a great diversity of different ethnic groups.
The majority-minority neighbourhoods from which we sampled our respondents for the TIES survey therefore fit within the definition of superdiverse neighbourhoods that we gave earlier.
We use the TIES survey, even when it is already 10 years old, because it is the only available survey in which people of native Dutch descent were asked about integration and diversity in a majority-minority neighbourhood context. The TIES survey contained a great number of questions about their experience of and opinions on integration and diversity, as well as how they actually behave in practice. like the school choice for their children or where they spend leisure time.

The Super-Diversity Matrix: Practice and Opinions
On the basis of the answers from the TIES survey, we first analysed how the inhabitants of Dutch descent in majority-minority neighbourhoods are distributed across the four quadrants of the super-diversity matrix in Fig. 10.1. In Table 10.1, we did this by combining the respondents in Amsterdam and Rotterdam. We are of course restricted by the questions that were asked in the survey. We used one question about the number of friends with a different ethnic origin and one opinion question about how they perceive living in a culturally diverse city. More precisely, we can combine the degree of contact outside their own ethnic group and the degree to which they perceive cultural diversity in their city as either enriching or threatening. The actual questions asked by the survey are: "Do you experience living in the city with people from different cultures as threatening or enriching?" and "How many of your friends are of another ethnic origin than yourself?" For both questions, we used a five-point scale that we reduced to three answer options by combining the two most extreme answers on each side. The outcomes confirmed what we had already cited from the literature: most people of Dutch descent living in superdiverse majority-minority neighbourhoods have little or no interethnic friendships. The quadrant on the bottom right contains the most respondents: people who have little or no interethnic friendships but who perceive diversity as enriching. This was followed by people who fall in-between the two bottom quadrants. They have little or no interethnic friendships and say that the diversity in their city does not make any difference to them. There we see some of the ambivalence on this issue.
The degree of interethnic friendships is not statistically correlated to opinions on diversity in the city: in other words, having more interethnic friends does not lead to different opinions on diversity. Below we show the superdiversity matrix for both cities separately. Although there are considerable differences between the cities, almost all of the respondents in both cities are on the axis of little or no interethnic friendships. Amsterdam, however, has almost twice as many people in the 'diversity is enriching' quadrant on the bottom right as Rotterdam. Conversely, Rotterdam has twice as many people in the 'diversity is a threat' quadrant. There are relatively more people in Amsterdam who have many interethnic friends and say that diversity makes no difference to them (Tables 10.2a and 10.2b).
In Amsterdam there is a moderately significant correlation (p < 0.1) between the degree of having interethnic friends and the extent to which respondents see diversity as being enriching. There is no significant correlation in Rotterdam. In the next two paragraphs we will examine contacts and opinions separately.

Diversity: Practice
In this paragraph we will take a closer look at practices in terms of the number of interethnic friendships. We will first show the share of interethnic friends for people of Dutch descent in superdiverse majority-minority neighbourhoods in both cities. In both cities we see approximately the same pattern. Almost three-quarters of the respondents of Dutch descent in these neighbourhoods only has friends who are also of Dutch descent. The respondents in Amsterdam have a slightly more mixed circle of friends (p < 0.05) than those in Rotterdam (Table 10.3a).
If we compare these outcomes to those of the Turkish and Moroccan second generation in superdiverse majority-minority neighbourhoods, we see that they have a mixed circle of friends much more often. Just one in five has no friends from a different ethnic background. A quarter of them has a predominantly mixed circle of friends (Table 10.3b).
Although all three ethnic groups live in the same neighbourhoods, the young people of Turkish and Moroccan descent have friends from a different ethnic background much more often. The ethnic group to which they belong may be two to three times smaller than the group of Dutch descent in their neighbourhood because there are so many different ethnic groups, but we still see that in proportion the second generation has a mixed group of friends much more often.
These questions referred to the respondents' circle of friend. The TIES survey also posed a question about the respondents' three best friends and their ethnic origin. When we look at the ethnic background of the three best friends, we see, not surprisingly, the same pattern as with the question about their wider circle of friends. Out of the 699 three best friends of people of Dutch descent in Rotterdam, only 113 (16%) were of non-Dutch descent, including 22 people (3%) of Turkish or Moroccan descent. The outcomes were very similar in Amsterdam. Out of 738 best friends, 125 (17%) were of non-Dutch descent, including 28 people (4%) of Turkish or Moroccan descent. The respondents of Dutch descent we are referring to live in  neighbourhoods where at least 50% of their neighbours are of non-Dutch descent and where a significant part of the inhabitants in their own age group are of Turkish or Moroccan descent. When asked whether they would send their children to a school where half the pupils had a migrant background (actually a reflection of the neighbourhood they live in), 62% of the respondents of Dutch descent in both Rotterdam and Amsterdam said that they would not do so. The figures for the Turkish and Moroccan secondgeneration respondents were the exact opposite of those of the respondents of Dutch descent. Approximately two-thirds said that they would send their children to a ethnically mixed school. School choice, of course, entails a multi-layered choice process, that includes issues as diverse as choices regarding pedagogies, distance to the school and social class. In the survey question the only reference made was to ethnic composition. Since these respondents lived in ethnically diverse neighbourhoods, not sending their children to a school with a majority of pupils of migrant descent probably meant enrolling them in a school outside the neighbourhood, which is generally inconvenient. When asked whether they would socialize in places that are popular with people of non-Dutch descent, 69% of the respondents in Rotterdam and 62% of the respondents in Amsterdam said that they would not do so.

Diversity: Opinions
In contrast to the degree of interethnic contact, which was very similar in both Rotterdam and Amsterdam, we observed major differences between the cities when we turned our attention to opinions. The difference was extremely significant (p < 0.001) (Table 10.4). In Amsterdam twice as many respondents in majority-minority neighbourhoods perceived the cultural diversity in their city as enriching as in Rotterdam. More than twice as many respondents in Rotterdam saw diversity in the city as threatening or quite threatening. The respondents who experienced diversity as a threat also had significantly more negative scores to other questions on their opinions regarding diversity. Not unsurprisingly, they said significantly more often that they have 'colder' feelings towards Muslims, Turks and Moroccans. They also thought significantly more often that people of non-Dutch descent should adapt to Dutch norms and values in public spaces and that more attention should be paid to the Dutch language. They were significantly less likely to think that special measures should be taken to combat discrimination. These opinions were also expressed when asked about practices. They said significantly more often that they did not want their children to go to a school with a majority of migrant children and indicated significantly more often that they preferred not to go out to places frequented by a majority of people with a migrant background. The question about cultural diversity in the city is therefore representative of a much more comprehensive body of negative ideas regarding diversity and interactions with people of non-Dutch descent.
In summary: although there are few differences between the cities when it comes to the interethnic contact though friends, there are considerable differences between Rotterdam and Amsterdam regarding the opinions on diversity held by the respondents of Dutch descent living in superdiverse majority-minority neighbourhoods. It is therefore more about a difference in their 'taste for diversity', than striking differences in behaviour. The pattern in Amsterdam corresponds to the patterns that Wessendorf (2014) found in her research on the London district of Hackney. She describes how there was a positive appreciation of diversity in everyday situations, without this being expressed in people's choice of close friends and acquaintances.

Explanation of the Differences Between People of Dutch Descent in Superdiverse Majority-Minority Neighbourhoods in Amsterdam and Rotterdam
One of the reasons that have often been put forward to explain the large differences between Rotterdam and Amsterdam regarding opinions about diversity is the difference in the socio-economic composition of the two cities. One is a harbour city with a large low-educated population of Dutch descent, while the other has a major financial sector and large number of highly-educated professionals of Dutch descent. The argument goes that as people living in Amsterdam have a stronger economic position, they have less to fear from the consequences of migration and globalization (Entzinger and Scheffer 2012). It is true that in Amsterdam there are more people of Dutch descent working in jobs that require a higher level of education in comparison with Rotterdam and that there is a larger group of people of Dutch descent with lower levels of education in Rotterdam (idem). This may indeed contribute to the different outcomes at city level. However, for this chapter, we are looking at outcomes at the neighbourhood level. The TIES survey was conducted in the same type of neighbourhood in both cities and in both cities we only selected people of Dutch descent living in superdiverse majority-minority neighbourhoods. For this reason, we did not find any significant differences in professional status between the respondents of Dutch descent in these neighbourhoods in Rotterdam and Amsterdam. Another explanation that is often put forward is the precarious position that people of Dutch descent may occupy in the labour market and accompanying fear of competition and job loss. This also cannot explain the difference: in Rotterdam 28% of this group had a temporary contract, while this figure in Amsterdam was 30%.
As these oft-mentioned socio-economic explanations for the differences in reactions between respondents in Rotterdam and Amsterdam did not show any statistical effect, we then set out to see what other possible differences in background characteristics we could find to explain the contrast between the two cities. One of the other explanations put forward in the literature is Florida's (2005) theory about the creative class, which is supposedly more tolerant of ethnic diversity. This assumption prompted us to look at occupational groups in the TIES survey and their opinions on diversity. People who did skilled or unskilled manual labour in both cities had the same opinions on cultural diversity. The group performing unskilled or skilled manual labour is the most negative about cultural diversity out of all occupational groups in the TIES survey. The group that perceived diversity as a threat was the same size in both cities. We found no significant differences. As this group is larger in the entire city of Rotterdam, this goes part of the way towards explaining the difference in voting behaviour at city level between Amsterdam and Rotterdam. However, there is no difference between the size of this group in both cities in the TIES survey, as the samples were taken in neighbourhoods where many people have a migration background (18% in Amsterdam versus 20% in Rotterdam). We also found no differences between Rotterdam and Amsterdam when we looked at the opinions held by the group of self-employed persons in the TIES survey.
We then went on to look at the respondents from the various occupations included in the 'creative class'. We used the broad definition of the creative class as used by Reijndorp (2004): artists, designers, researchers at universities, consultants and advisors. The creative class was indeed the most distinctly positive about cultural diversity. The idea that the 'creative class' is more open was confirmed by our analyses. In his concept of the 'creative class ' Florida (2005) only looks at the group that promotes openness and toleration, but he does not examine the group that may have a negative influence on this climate. These are the people who propagate a climate of intolerance and exclusion and who may also have a negative effect on contact between groups in daily practice due to their attitude and opinions. In our research it appears that these respondents often work for the army, police or security firms: enforcement or supervisory occupations. In addition to people in these occupations, we also found that a relatively high number of people in white-collar jobs, such as administrative occupations, health care and the social sector, also perceive diversity in the city as a threat. The only large and significant differences in opinion that we could find between the two cities were those held by people in white-collar jobs. This group in Rotterdam expressed negative opinions on diversity much more often than the same group in Amsterdam. We first checked a number of control variables to see whether we could find any differences between these groups. However, we found no significant differences between the two groups in the different age cohorts (18-24; 25-35) and there were no differences between men and women that could have explained the difference between the two cities.
In Table 10.5 we show the outcomes for the different occupational groups. Once more we have combined the two most extreme categories as there were not many people in some occupation categories.
Approximately half of the people working in education, health care or the social sector, or in mid-management jobs in the technical or financial sector think that cultural diversity is enriching. But in contrast to the people in the creative sector or higher management positions, there is also a substantial group (approximately 1 in 5) that perceives cultural diversity as a threat. The group with negative opinions regarding cultural diversity was slightly larger among those occupying mid-management administrative positions.
When we now look at the differences between the two cities, we see a strong similarity between the two groups at the opposite ends of the spectrum. In other words, people in creative occupations in Rotterdam did not say that they saw cul- tural diversity as a threat more often because they live in Rotterdam. They are not influenced by the more negative atmosphere surrounding diversity in their city. Conversely, people working in enforcement occupations in Amsterdam were not influenced by the more positive atmosphere surrounding cultural diversity in that city. These groups appear to be extremely stable in their opinion on cultural diversity, independent of the context in which they live. The groups occupying mid-management positions in the other sectors were less stable. There are substantial differences between Amsterdam and Rotterdam in all three groups in the mid-management segment. In Rotterdam we found four times as many people in the social, health care or education sectors who saw diversity as a threat as in Amsterdam. In the other two sectors, this was three times as often. On the opposite side of the spectrum -people who perceive diversity as enrichingthere are also considerable differences between the two cities (Table 10.6).
The survey did not contain a very large number of respondents in these three occupational categories, so some caution about the results is in place. We therefore conducted further analyses, shown below, by looking at people in white-collar jobs in these three occupational groups together. There is a substantial difference between the opinions of people in these white-collar jobs in Rotterdam and Amsterdam. In Amsterdam almost six times as many people saw cultural diversity in their city as enriching as in Rotterdam. This is a significant difference (p < 0.01) (Table 10.7).  This raises the question of what makes these white-collar groups in Rotterdam so different to their peers in Amsterdam. Once more, the explanation does not lie in socio-economics. The respondents in Rotterdam, for example, had a permanent employment contract even more often than those in Amsterdam. We also found no difference in the type or size of company they were working for. There was also no difference in family background characteristics. In both cities, this group of respondents had attained a higher socio-economic position than their parents. There were no differences with regard to the number of people who had been given a religious upbringing (one third) and the number that still identify themselves as being religious (one in five). The neighbourhoods where they live also do not differ in terms of the proportion of people with and without a migration background. The people in the middle groups in Rotterdam stated significantly more often (p < 0.1) that they would like to live in a less diverse neighbourhood. This indicates that they reject the degree of diversity in their neighbourhood.
As there are no differences with regard to socio-economic position or neighbourhood composition, it appears that we must look to the difference in climate between the two cities to explain the contrast. The more negative climate with regard to diversity in Rotterdam seems to be pulling the less stable middle level group towards a negative opinion while the more positive climate in Amsterdam seems to be pulling the middle group in a more positive direction.

Conclusion and Debate
Very little research has been conducted on people of Dutch descent living in superdiverse majority-minority neighbourhoods. How do they integrate into superdiverse neighbourhoods in which they form a numerical minority? On the basis of our analysis of the TIES survey data, occupational characteristics go some way towards explaining the extent to which people of Dutch descent in superdiverse majorityminority neighbourhoods perceive living alongside people from a different ethnic and cultural background as enriching. People who work in occupations characterized by the creation of new ideas, adaptation and innovation seem to perceive cultural diversity in the city as enriching. People whose job involves supervising or controlling people appear inclined to see cultural diversity more as a threat. As the group of people in creative occupations is larger in Amsterdam than in Rotterdam, this has an important effect on the climate in this city.
The occupational groups found at the extremes of this spectrum are the most context-independent. Their opinions regarding cultural diversity remain intact, almost independently of whether the general climate is one of tolerance or intolerance. Some occupational groups, however, are influenced by the context. In groups where ideas on the added value of diversity are less strongly entrenched, the urban climate may influence the direction in which their opinions shift. In a more open and tolerant climate, this group may move towards greater openness, while in a more negative climate, they will express themselves more negatively. This suggests that the opinions regarding diversity held by an important middle group can be influenced by the environment in which they live. It is possible that this negative process is being reinforced in Rotterdam as an increasing number of people express negative opinions about diversity, both in their daily interactions and the local public debate. This may make people feel less inhibited about holding and expressing such opinions. It is possible to use qualitative research to examine this hypothesis in more depth. The most important provisional conclusion of this chapter is that it is possible to specify occupational groups whose opinions on diversity are more dependent upon the urban context. This is a crucial insight for our study of how opinions on cultural diversity are formed and how they may also change. This also provides an opportunity for policymakers to aim policies at people from these occupational groups or to develop interventions aimed at groups who are receptive to changing their opinions.
There are, however, two other types of explanations that can be studied in more depth by subsequent research. One important factor influencing the differences we found between Amsterdam and Rotterdam may be that it is more difficult for people living in Rotterdam to move to suburbs where there is less ethnic diversity than in the city. The conflicts in neighbourhoods on the periphery of Amsterdam (IJburg) or the high number of votes for the PVV in suburbs (Almere), show that some people of Dutch descent want to escape the diversity of the city by moving. Many people living in the old neighbourhoods near the centre of Amsterdam have chosen to live in or move to a superdiverse neighbourhood. They either have no problem with cultural diversity or see it as enriching. The other explanation that requires more research is the influence of the immediate living environment. A living environment that requires relatively little contact between neighbours and has a low risk of nuisance due to its architecture may make it easier for people with different lifestyles and household composition to live alongside each other (see Crul and Steinmetz forth coming). There may also be differences between the two cities in this regard.
The analysis of the TIES data reveals that even people of Dutch descent who perceive diversity as enriching, hardly ever have a mixed circle of friends. A great deal of research focuses on interethnic contact and concludes that integration has failed if this is not present. Blokland and Nast (2014) show in their research in Berlin that it may be necessary to re-evaluate this point of view. Superficial contacts in the public space can also lay a foundation for feeling comfortable in a neighbourhood. This may be enough to prevent any minor irritations from escalating or even developing into conflicts between ethnic groups. If a part of the population in a neighbourhood is open to superficial contacts, this may have a positive influence on the neighbourhood's atmosphere as a whole. If contact is made primarily, or even exclusively, on an involuntary basis, as the result of nuisance or conflict, there is a higher risk of tensions arising, especially if the basic attitude is one of rejecting diversity. Further research must show if there are also differences between the two cities in this regard.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.