UNIVERSITY OF AMSTERDAM
GRADUATE SCHOOL OF SOCIAL SCIENCES
Who Is Against Immigration?
The effect of Threat Perception, Immigrants
Ethnicity, and Contextual Indicators
A Comparative Analysis between the Netherlands and Israel
Kerem van Mil 11657308
Master's Thesis Sociology: Migration and Ethnic Studies 2018 Supervisor: Dr. Agnieszka Kanas
Second Reader: Dr. Bram Lancee Keremvm@gmail.com
04.07.2018 Amsterdam
Table of contents
Introduction ... 2
The setting ... 5
Migration to the Netherlands ... 5
Migration to Israel ... 7
Theoretical Framework ... 9
Individual-Level Explanation to Attitudes Formation ... 9
Ethnic Hierarchies Theory: Explaining the Role of Different Ethnicities in Attitudes Towards Immigrants ... 12
The Mediation Role of Realistic Group conflict and Symbolic Threat ... 14
Data and Methods ... 16
The Dataset ... 16
Methods ... 17
Measurements ... 18
Results... 201
Descriptive Statistics ... 21
Multivariable OLS Regression Results- Hypothesis Testing ... 21
Conclusions and discussion ... 24
Individual- Level Analysis ... 25
Contextual Analysis ... 26
General Discussion ... 28
Reference ... 29
Introduction
An informal conversation, in a casual place, with native-born Dutch person (who refers to himself as ‘Autochtoon’1) prompted me to ask who is against immigrants and why? Does the economic position of a person determine his or her attitudes toward immigrants? Do native-born feel an economic threat? Or are attitudes driven by cultural issues? Is there also fear for the preservation of majority culture? Finally, is it universal or do the country’s immigration history and the present migration situation influence attitudes to immigrants as well?
This research focuses on individual attitudes towards immigrants in two countries, Israel and the Netherlands. After the Second World War in Europe, the emergence of transnational and multinational forces and regional organizations was crucial to migration movements; globalization processes made the flow of goods, money, and people, easy and safe. The Netherlands and Israel2 have become increasingly open to the movement of products, capital, services and money, and yet increasingly closed to the movement of people. The countries are increasing their desire and capability to control immigration and making stringent efforts to distinguish between the wanted and unwanted (Morris, 1997). In recent years, migration has become a hotly-debated subject, especially in the countries of North western Europe. Even though migration is not a new phenomenon, increasing numbers of undocumented migrants and refugees and rising illegal border crossings have moved immigration to the top of the political debates (Menz, 2009). International migration challenges the notion of borders, citizenship, and nationality, and affects the receiving countries and their population in many ways, both economically and culturally. Anti-immigration statements, negative attitudes, racism, and discrimination, become more and more common (Holmes & Castañeda, 2016).
Both in Israel and the Netherlands, immigration is a loaded subject. Recent debates in Israel and the Netherlands focus on the duty of the country to accept immigrants, the cost and benefit of immigration, economic burden and cultural war, racism, and discrimination. The core in these debates is the struggles of the majority population to adapt to the inflow of migrants. One question that arises, both on the theory level and the social level, is what determines people’s opinions about immigrants and immigration. Another issue is what are the primary mechanisms that affect people’s views. These questions generate the aims and the motivations behind this research.
1Autochtoon is a Dutch resident whose parents are both Dutch. This term is not officially used any more.
2 In the Netherlands this process happened earlier while Israel fully opened to international markets only in the beginning of the
There are three primary incentives for this research. First, analyzing the connection between individual characteristics and attitudes can reveal the part of the population who is being negatively or positively most affected. For example, higher opposition to immigration amongst skilled workers may highlight the negative effect immigration invoke on lower-skilled and low-income workers. On the other hand, people with high skills and high income may benefit from the increase in the lower-skilled workers that can work for lower salaries, and therefore, will have a moderate opposition. Moreover, analysis of opinions and attitudes towards immigration on the individual level can offer a way of addressing it. Personal views can reflect other processes like discrimination against immigrants, segregation and even the gap between native and immigrants in the job market. Understanding the source of these process can help to change them for the better. Finally, individual attitudes offer a broader understanding of the current political debates. People’s particular preference can shed light on the question why in recent years, even though the world seems to move towards globalization where the flow of goods and capital have high values, the flow of people have less endorsement and countries become more closed to immigration.
The aims of this research are, first, in line with previous studies on attitudes (Ceobanu & Escandell, 2010), to identify the characteristics in the native-born population in Israel and the Netherlands, that have an impact on attitudes towards immigrants. Second, to test whether an immigrant’s ethnicity can be accountable for different opinions. Therefore, unlike previous research, which usually uses the term immigrants as a uniform term in people consensus, I will use two different terms: immigrants from the same ethnic group as the majority and immigrants from a different ethnic group as the majority (Bessudnov, 2016). Third, to examine whether feeling threatened by the out-group can be an additional explanation for attitudes towards immigrants (Halperin, Canetti-Nisim, & Pedahzur, 2009). Finally, the question of whether these correlations are universal led me to compare the two countries to see if the broader socio-political and historical context may also be accountable for people’s attitude.
A review of the existing literature reveals different perspectives with different levels of analysis that have been used so far. In each level possible explanation to anti-immigration attitudes exists. The first level is the micro level, where individual characteristics shape attitudes towards immigration. At the micro level, one group of studies focus on the economic indicators. Using the "Labor Market Competition Theory," immigrants create competition in the labor market for individuals with similar skills, and this competition generates negative attitudes (Semyonov, Raijman, & Yom-Tov, 2002). The second group of theories, based on the "Social Identity Theory," focus on social and cultural indicators, claiming that social identity plays an important role generating inter-group relationships (Savelkoul, Tolsma, & Scheepers,
2015). An elaboration of this theory often looks at the nation as the individual in-group and therefore the individual symbolic concerns toward the country are an essential predictor of immigration attitudes (Jeong, 2013).
The second group of research focuses on indicators that may affect the relationship between individual characteristics and attitudes towards immigrants. One, using "Ethnic Hierarchy Theory," is the immigrant's ethnicity and the in-group preference in multicultural societies (Hagendoorn, Drogendijk, & Hraba, 1998; Kleinpenning & Hagendoorn, 1993; Snellman & Ekehammar, 2005). Second, based on “Reslistic Group Conflic Theory” and “Symbolc Threat Theory”, is the fear of the out-group, realistice and symbolic threat to the in-group(Schlueter & Scheepers, 2010; Schneider, 2008).
The third level focuses more on the macro structures and examines contextual factors as an explanation for individual attitudes. Within this set of studies, the immigration history of the country and the different existing circumstances in each country influence attitudes (Meyers, 2000; van Houdt, Suvarierol, & Schinkel, 2011). In this research, on the one hand, Israel, has a history of mainly Jewish immigration until recently, and permanent settlement is an option only for Jewish people. On the other hand, the Netherlands, has a history of mixed immigration waves and a screening process that focuses on economic and cultural criteria (Duyvendak, 2011; Joppke & Rosenhek, 2002).
In this research, I will try to scrutinize the possible role of each to explain the formation of attitudes in Israel and the Netherlands. For this purpose, data from the European Social Survey (ESS) Round 8 (conducted during 2016) is used. This dataset contains both Israel and the Netherlands and provides the newest data about attitudes towards immigration. I will make an attempt to contribute the existing literature on the formation of attitudes in both countries and try to answer four research questions. (1) Do native population individual characteristics (employment status, education level, national feeling) have an impact on attitudes towards immigration? (2) Do native population individual characteristics have a different effect on opposition towards immigrants from the same ethnic group and immigrants from a different ethnic group? (3) Does perception of threat (symbolic or realistic) serve as an additional explanation for the effect of individual characteristics on attitudes towards immigrants? And finally, (4) does contextual indicators on the country level also be accountable for anti-immigration attitudes?
The setting
Throughout the years, immigration has transformed the host counties in various ways. It is a trigger and a consequence of different processes (Portes, 2010). Micro-level individual factors and macro-level structure can both influences individual preferences on immigration and immigrants. Thus, to fully understand the perception of native-born on international migration, there is a fundamental need to review the history and the structure of each country.
Comparing the Netherlands and Israel is worthwhile because of the differences and the similarities between the countries. The comparison between the two countries can extend the understanding on how contextual indicators affects attitudes formation. For one, both countries have a way to distinguish the majority group from the minority group and reflect this differentiation on immigrants as well. Yet, the line is drawn based on different categories. In Israel, the line is based on ethnicity; Jewish and non-Jewish. In the Netherlands, the line is drawn using cultural indicators; western culture and non-western culture. Second, both countries experience diverse immigration inflow, but at different periods. Finally, in both countries, integration into the society, and the capabilities of the newcomers to integrate into the society is an essential factor affecting attitudes of native-born towards new-comers. However, the basis for measuring the fit of the immigrant to the society is different. In Israel, the integration is based on ethnic characteristics and acting by the Jewish religion, culture, and norms. In the Netherlands, economic and social features are the basis to measure integration processes (Barrett, 1998; Czaika & De Haas, 2013; Meyers, 2000; Simonsen, 2016; Willen, 2007; Zimmermann, Bauer, & Lofstrom, 2000).
Migration to the Netherlands
Even though the movement of people was always part of the European history, the immigration history never became a part of the national self-understanding for the European countries. European states tend to see immigration as a challenge to their territory, their institutes, and their boundaries. Nevertheless, since the end of World War 2, millions of immigrants from developing countries have settled in Western Europe. The immigration history of the Netherlands is no different than that of the rest of the Western European countries and can be divided into three main periods.
After World War 2, the urgent need to reconstruct Europe raised the level of demand for work. This demand, however, could no longer be satisfied within the domestic regime. In the
pattern of most Western European countries, the Netherland looked for workers first in Southern European countries, later in Turkey and finally in North African countries. During the 1960s, the Netherlands had launched their first guest worker program, followed by roughly ten more in the coming years. 3 These agreements were meant to control immigration, not to create new flows. The basic idea behind the guest worker schemes was that the worker would stay as long as there are jobs for them (Hansen, 2003). All countries ended their "guest worker" programs in the early 1970s. The oil crisis in 1973 and the economic slowdown after that increased unemployment rate among natives as well as immigrants. Guest workers were no longer required as before.
The ending of the recruitment agreements, in the late 1970s, did not lead to decreasing flows of immigrants into the Netherlands, but it did lead to changes in the immigrant’s characteristics. The immigration community now included women and children, younger and older people. All efforts to limit family reunification or encourage repatriation failed. Likewise, the Dutch government increased efforts to stimulate return migration, but with little effect (Meeteren, Van de Pol, Dekker, Engbersen, & Snel, 2013). Consequently, since the 1980s immigration to the Netherlands has grown in terms of diversity and numbers.
The third period of immigration to the Netherlands started in the 1990’s with the rise of asylum seekers and refugees. In 1990, the largest immigrant groups were from the former Yugoslavia and the Soviet Union. At that time, 100,000 immigrants entered the Netherlands almost every year. Civil wars around the globe, the fall of the Iron Curtain, and the Maastricht Treaty signed in 19924 created new immigration movements. The peak was at the beginning of the year 2000, when more than 130,000 immigrants entered the Netherlands. According to Dutch statistics, half of the immigrants arriving at the beginning of 2000 were from non-European or non-western states and the other half was from the EU or other western countries like the United States and Canada (Ersanilli & Koopmans, 2010; Snel, de Boom, & Engbersen, 2004). In the early 2000s, the Netherlands saw a decline in immigration rate, but between 2007-2008 the influx was high again. Since 2011, migration flows into the Netherlands have been affected by humanitarian crises around the world, resulting in an increasing number of people attempting to reach Europe (International Migration Outlook 2017, 2017). In 2015, 205,000 immigrants entered the Netherlands.
3 In 1960, the Netherlands signed a recruitment agreement with Italy, with Spain in 1961, Portugal in 1963, in 1964 the Netherlands
signed the agreement with Turkey, followed by Greece in 1966 and Morocco 1969. With Yugoslavia and Tunisia, the agreements were signed in 1970.
Immigration movements and immigrant's characteristics are continually changing and affecting the host countries in various ways. Furthermore, different immigrant groups might provoke different attitudes and economic struggles may develop into cultural battles (Portes, 2010). The increased inflow since 2008 comes against the financial crisis and deeper unemployment amongst both native-born and immigrants, which increases the common perception of failed integration policies that did not achieve to make the immigrants part of the society. As a result, the influx was soon framed as a crisis and was experienced by the native population in the Netherlands as one outstretched event. Once seen as a crisis, immigrants became a direct enemy to the society (Jeandesboz & Pallister-Wilkins, 2016).
Migration to Israel
Israel is a traditional settler country for Jews, “The Home” of the Jewish people. Between the end of 1919 and the establishment of the Jewish State in 1948, the number of Jews living in Israel increased twelve-fold, mainly due to immigration. After the establishment of the Jewish state, immigration continued to be a big part in the formation of the nation. At the end of 2000, 40 percent of the Jews living in Israel were foreign-born 70 percent are either immigrants or second-generation immigrants. The goal of the Israeli immigration policies since 1948 has been to bring as many Jewish people as possible and prevent immigration of non-Jews. This was to increase the number of Jewish people living in Israel and maintain the Jewish majority. Nonetheless, the history of immigration to Israel does not end with the history of Jewish immigration (Gal & Leshem, 2000).
For several decades, because of the massive immigration of Jewish people and the labor force available from the Palestinian occupied territories,5 Israel did not consider labor migration. Until the war in 1967, Israel remained self-sufficient in her labor requirements. After the war, the demand for temporary labor expanded. The Palestinian people, from Gaza and the West Bank, were an essential source of labor in all the low-skill occupations, mainly agriculture, construction, and the service sector. Between 1967 and 1987 the agriculture and construction sectors in Israel became dependent upon Palestinian labor. The supply of working hands from the occupied Palestinian territories seemed unlimited. However, the uprising of the Palestinian people (the Intifada) in 1987 and the periodic closures of the Gaza Strip and the occupied territories in the West Bank by the Israeli Defence Forces, made the Palestinian workers less reliable in their availability, and consequently also less attractive to employers. The economy kept on growing, and industries were developing, but, while these workers got out of reach, the
demand remained high. Major employer groups put pressure on the Israeli government to find a solution. The result was new permits for temporary foreign workers who are not from the Palestinian occupied territories (Borowski & Yanay, 1997; Choen, 2002; Furst-Nichols & Jacobsen, 2011; Sabar, 1998).
By the beginning of the 1990s, foreign workers had become an important part of the labor force. At the end of 1990, the government issued a list of guidelines governing the employment of temporary labor immigrants, mainly to ensure their temporary stay (Borowski & Yanay, 1997). Foreign workers came primarily from three countries: Romania, Thailand, and the Philippines. Illegal workers came from a larger number of countries, including Poland, Bulgaria, Ghana, China, Sri Lanka, and Turkey (Papademetriou & Hamilton, 1995). At the end of the 1990s, the care sector, the agriculture, and the construction sectors were based mainly on foreign workers. By 2000, the proportion of foreign workers in the Israeli labor force was higher than in most OECD countries. According to Israeli statistics, the number of foreign non-Jewish people who entered Israel has been increasing since 2012, and at the end of 2016, the number was 100,000, only 85,000 of whom were legal (International Migration Outlook 2017, 2017).
In Israel, just as in the Netherlands, the continuing flow of migrants was against mixed economic, social, and security situation. At first, the increased use of foreign workers derived from Israel geopolitical situation, most of the decisions to import workers were justified by security needs.6 The second uprising of the Palestinian in 2001 created an even more problematic security situation with increasing debates on the need to ensure Israel as the home of the Jewish people. Furthermore, framing the country as a close in-group with strict roles and definitions makes the out-group perpetual enemy. In addition, the increased use of immigrants in specific sectors raises the debate about the connection between immigration and unemployment amongst low skilled native-born. Since 2010, the non-Jewish immigrants seem more as a threat and burden and less as a working hand.
6 The Israeli geopolitical situation, and the constant war with the Palestinians prevents the Israeli government from issuing work
Theoretical Framework
Individual-Level Explanation to Attitudes Formation
To outline the theoretical framework and to explain individual preferences on the micro level, I drew upon two main explanatory indicators: the economic characteristics of the individual and the cultural non-economic ones (Mayda, 2005; Scheve & Slaughter, 2001).
A set of studies and theories argue that material economic concerns lie at the center of anti-immigrant attitudes. An individual with a different skill level and a different position in the labor market may have reasons to view immigration in different ways (Facchini & Mayda, 2009; Hanson, Scheve, & Slaughter, 2007; Mayda, 2006; Scheve & Slaughter, 2001). The first economic indicator is employment status. The most common theory to explain that is the "Labor Market Competition Theory." According to this theory, people will resent immigrants with similar skill levels that may pose economic competition to them (Citrin & Wong, 1997; Facchini & Mayda, 2012). Scholars who used this framework, like Dustmann and Preston (2004), suggested that personal labor market concerns are a significant determinant of attitudes towards immigration. Immigration induces shifts in the national labor market and creates national wage changes mostly among the low skilled native population. Therefore, they are the ones that most probably will be negatively affected by the immigration inflows (Scheve & Slaughter, 2001). However, a large portion of the studies using this theory did not consider other mechanisms that may affect attitudes (Dustmann et al., 2004).
The second economic individual indicator is the level of education. Several studies already concluded that the more educated people are, the less anti-immigration they are (Citrin & Wong, 1997; Dustmann & Preston, 2007; Mayda, 2005; Scheve & Slaughter, 2001; Zimmermann, Bauer, & Lofstrom, 2000). There are three ways to explain the connection between high education and low anti-immigration attitudes. The first explanation relies on the labor market competition theory. Immigrants are not a competition to the highly educated people. However, this does not explain finding indicating that highly educated people have higher opposition to low skilled immigrants than to high skilled (Hainmueller, Hiscox, & Margalit, 2015). The second explanation is that better-educated people are more likely to follow debates regarding immigration. These debates usually contain more than one viewpoint and have a broader picture of the consequences of immigration on the country. Individuals that pay more attention to the discussion tend to have less extremist views (Dustmann et al., 2004). The third explanation links education and a higher level of ethnic and racial tolerance. Research
shows that education, especially at the college level, generates a more diverse network. This implies that the more educated people have a more complex view of other culture, ethnicities, and race. They thus tend to have a more accepting attitude towards immigrants and minorities (Citrin & Wong, 1997; Espenshade & Calhoun, 1993; Espenshade & Hempstead, 1996).
Those discussions lead to two hypotheses. On the individual level, employment and higher education will have a positive influence on opposition towards immigrants (both immigrants from the same ethnic group and immigrants from different ethnic groups) (H1). On the national level, however, I do not expect to see much of a difference between the two countries for two reasons. First, according to the OECD numbers and other national and international research, native-born in both countries have more or less similar economic characteristics (OECD, 2017a, 2017b); my descriptive statistics show the same results as well (See Table 1). Consequently, when considering only the effect of economic indicators, with no other interaction or mediation effect, then I do not expect to see a significant difference. Second, the immigration inflow in both countries in the last decades have similar characteristics: low skilled migrants, mainly from the sub-Saharan countries or the Mediterranean area; refugees from different humanitarian crisis around the world; and low skilled work immigrants from the less developed countries. Hence, when examining the effect of immigrants on the native population, it seems clear that the main economic impact will be on the low skilled worker in both countries. Moreover, since 2008 both counties have struggled to recover from the financial crisis, creating nearly the same economic conditions in the country. The above discussion suggests a second hypothesis, the positive influence of employment and higher education on opposition towards immigrants will be the same in both countries (H2).
The non-economic factors influencing attitudes’ formation are broad. Yet, “Social Identity Theory” is one of the most elaborate theories on in-group- out-group relations (Sides & Citrin, 2007). The theory intended to explain different behaviors using group processes. Accordingly, all group behaviors such as favoritism and discrimination are part of social identity processes. The main aim of the individual is to create a positive social identity for himself. Thus, social identity is a favorable comparison between the in-group and the out-group. The motivation to pursue and obtain a positive social identity is always present, since people need to obtain a positive social identity, they make an effort to gain it (Mangum & Block, 2018). According to the theory, when the in-group has great emotional significance, then the tendency to have in-group favoritism in higher. Therefore, even the minimal intergroup discrimination is often involved with maximum motivation to have differentiation. The interplay between superior and inferior and the group’s movement between those categories, together with the
social need to keep and maintain the favorable superior position is what we can see as the interplay between majority and minority, between native-born and immigrants.
The definition of social identity can be applied on the national level as well. “The National Identity Theory” argues that in the modern world, the country is perceived as an in-group and an objective of allegiances. National identity gives people a sense of who they are relative to others and infuses them with a sense of purpose and home (Hjerm, 1998). Additionally, the national identity is a normative content that people use to create 'us' and 'them.' Thus, it is based on similarity to some people and differences from others. Immigration is firmly connected to national identity. It challenges it and may undermine it. "If national identity means self-definition and belonging to the national polity, then immigration cuts to the heart of this concept” (Hjerm, 1998, p. 88). The main finding of Wright (2011), for instance, is that mainstream citizens do seem to reshape their definition of the national community along narrower lines in response to heightened immigrants inflows.
With regard to social identity theory and the national identity theory, I framed two hypotheses. First, on the individual level, national attachment will have a negative effect on opposition towards immigration (both immigrants from the same ethnic group and immigrants from different ethnic groups) (H3). Second, on the country level, I expect that native-born Israeli will indicate higher national attachment feeling that forms higher opposition towards immigrants. Israel is a country that was built for the Jewish people under their right to have a Jewish state where they can determine its cultural and political character. Having a strong national identity is also a way to ensure the future survival of the majority (Gavison, 1999). The ethnocultural terms determine who is allowed and who is not allowed to immigrate. The use of these as a means to forestall future threat and the secure the Jewish majority led to ethnic discrimination and hierarchy (Joppke & Rosenhek, 2002; Yonah, 2004). In the Netherlands, a country that has already existed for several hundred years, the future survival of the nation seems less in risk than in Israel, so the need to have a strong national identity that fiercely rejects everyone else is also weaker than in Israel (Givens & Luedtke, 2005). Thus, I predict that the negative effect of national attachment on opposition towards immigrants will be stronger in Israel than in the Netherlands (H4).
Ethnic Hierarchies Theory: Explaining the Role of Different Ethnicities
in Attitudes Towards Immigrants
On the second level of indicators, I summarize the theoretical framework for this level using the literature on ethnic hierarchy. People have a way of ranking ethnic groups systematically (Berr & Kalin, 1979; Duckitt, 1992; Kleinpenning & Hagendoorn, 1993). In 1960, Pettigrew found that white South African students had contact preferences, starting with Afrikaners, then English South Africans, Jews, Colored7, and Indians (Pettigrew, 1998). Similar to his finding, Barry and Kalin (1979) found in their research a more general ethnic hierarchy. Hagendoorn, Drogendijk and Hraba (1998) found, in a study conducted among Dutch students, that western Europeans were ranked first, while people from the Far East and African countries were last. These studies, among others, provide evidence of what Hagendoorn (1995) labels as "Ethnic Hierarchy.”
Similar to prejudice and discrimination, ethnic hierarchies are based on stereotypes and in-group preference and are motivated by emotional and physical security. When stereotypes, as a cognitive element, are a shortcut for judging the other group, and prejudice is a practical element of these cognitive elements, both become an instrument for ethnic hierarchies. According to Hagendoorn and Sniderman (2001), other factors that may affect ethnic ranking are the socio-economic status of the out-group and perceived threats from the out-group. People believe in their cultural superiority. Their ethnocentrism and stereotypes will lead to ranking the out-groups based on how close the out-group is to the in-group and how socially desirable the out-group. The assessment of how close the out-group is and how beneficial it is may be based on the economic situation or the social position of the out-group and the in-group (Berr & Kalin, 1979).
There are two processes in the creation of ethnic hierarchies: in-group preference, where the need to maintain positive social identity justifies in-group favoritism, and consensus on the rank of each of the out-groups amongst the group members, which usually is based on cultural differences between the two groups (Hagendoorn et al., 1998). Although in every society there is a different combination of ethnic groups and positions, the tendencies to create hierarchies are universal and can be observed in many different societies (Hagendoorn et al., 1998; Kleinpenning & Hagendoorn, 1993; Snellman & Ekehammar, 2005). Nevertheless, ethnic hierarchies reflect the political and economic situation of the group and the country and may also reflect the deep-seated ethnic and national differences in the country.
In line with the ethnic hierarchy theory, I used two dependent variables, immigrants from the same ethnic group and immigrants from a different ethnic group. First, it would be a mistake to assume that all respondents are grasping the concept of 'immigrant’ or ‘immigration’ in the same way. Second, the ethnic hierarchies that individuals possess are essential to the understanding of attitudes formation. On the individual level, I hypothesize that the effect of individual characteristics will be different depending on the immigrant's ethnicity. Individual economic characteristics will have a smaller positive effect on opposition to immigrants from different ethnic groups, while national attachment will have a stronger negative effect on opposition towards immigrants from the same ethnic group (H5).
Next, on the country level, I hypothesis that in both countries, the overall opposition towards immigrants from different ethnic groups will be higher than the opposition towards immigrants from the same ethnic group (H6). In each society, however, the meaning of ‘immigrant from a different ethnic group’ is unique and important to this study.
In the Netherlands, the changes over time in the immigrant's characteristics created new and different categories to classify the immigrants. Classifying immigrants in the Netherlands was often done by the terms "Allochtoon" and "Autochtoon" which, until 2016, was the official way to differentiate native white Dutch from others living in the Netherlands.8 Autochtoon is a Dutch resident whose parents are both Dutch. Allochtoon refers to persons with one or both parents born outside of the Netherlands. Even though the terms are no longer in the official language, they are still socially applied. The official abandonment of the term also formed new terms like "Western Allochtonen" when referring to immigrants with a western background and more successful integration, or “expect” when referring to immigrants with high skills who were invited to work in the Netherlands.9 The social use of allochtoon is meant to draw attention to one’s ethnicity with implication to the superiority of the native Dutch ethnicity (Wekker, 2016). In non-official language, the term allochtoon is usually applied on a racial or ethnic basis. Mainly, it refers to non-Western groups that seem to be less integrated (i.e., Antilleans, Moroccans, and Turks). This distinction defines the majority and minority groups in the Netherlands. Furthermore, this also creates the distinction between immigrants from the same ethnic group; EU members with Western culture and other western immigrants who either fit the Dutch culture or can easily be integrated to it. And immigrants from different ethnic groups; non-western immigrants.
8 In November 2016 the Dutch government decided to drop the use of these terms, however, in the society these terms are still
being used (Adviesraad overheid stopt met termen allochtoon en autochtoon,”2016).
In Israel, the national definition of immigration and immigrants is different from other countries, which provides the understanding of who are the immigrants 'from different ethnic groups’ (Chaloff, Jonathan & Mestres-Domènech, 2011). There is a fundamental need to understand the three main categories in the Israel debate over immigration and immigrants to understand the concept of immigrants from the different ethnic group. The first one refers to Jewish immigrants or their family members. This is known as "Aliah," and the immigrants are called "Olim".10 This immigration process of Jewish people is codified in the Law of Return (from 1950), which covers a heterogeneous group that includes Jews, family members of Jews and individuals with a Jewish connection. The other two are non-Jewish immigrants; foreign workers (who usually enter legally) and refugees (who enter illegally) (Shamai & Amir, 2016). In the discourse of the Israel society, this distinction between Jewish immigrants and non-Jewish immigrants is fundamental. First, only the non-Jewish immigrants have the possibility for a permanent stay in Israel. Second, the difference is part of the separation between the majority and minority group in Israel. Thus, it also distinguishes the immigrants that are from the same ethnic group, Jewish immigrants, from immigrants that are from the different ethnic groups, non-Jewish immigrants.
The Mediation Role of Realistic Group conflict and Symbolic Threat
To build the theoretical framework of the mediation role of threat perception, I draw upon the literature of different threat theories. “Concern about economic well-being and national identity underlay reaction to immigrants in Western Europe.” (Sniderman, Hagendoorn, & Prior, 2004, p. 36). Threat and fear play a vital role as a mechanism of anti-foreigner attitudes and negative reaction to immigration. Different attitudes and behaviors between groups are aimed to maintain the group resources, keeping the power position, and preserving the culture and identity of the group. As a general definition, out-group threat occurs when one group feels that its well-being is challenged by the actions, beliefs or characteristics of another group. The core of the threat theory is that “competition for resources leads to attempts at the exclusion of one group by another” (Olzak, 1992, p. 163). In this view, minorities and immigrants are potential competition over limited resources (Manevska & Achterberg, 2013). Threat perception can be on the individual level and the group level (Sniderman et al., 2004).One of the earliest theories of intergroup relations and threat is the "Realistic Group Conflict Theory." When two groups are competing over the same scarce resources, the success of
10The Hebrew for immigration is “Aliah” and the Hebrew word for immigrants is “Olim”. Both terms are used only when referring
one group means a threat to the well-being of the other. The resources may be money, land, power or privileges (Blake, Mania, & Gaertner, 2006). Additionally, the greater the sense of threat and fear of competition, the greater the motivation to discriminate the out-group (Semyonov et al., 2002). The second concept is the "Symbolic Threat." Case, Greeley and Fuchs, (1989) define cultural and symbolic threat as the beliefs that the out-group has different values and different norms, and that those values and norms are a threat to the values and norms of the in-group. The group differentiates in values, beliefs, moral, and cultural behavior can create competition, prejudice, and negative attitudes. Theories of racial threat and racial discrimination are a subset of the symbolic threat theories (Hopkins, 2010).
The conflict between groups is composed of situational-objective conditions, in addition to a subjective evaluation of the threat that derives from the out-groups. For that reason, threat perception should be examined as a mediator between the socioeconomic characteristics and attitudes. Only a few studies had examined the influence of threat as a mediator. Some of the studies found a vital role of threat in mediating between the effect of education and anti-immigration attitudes. Other research found that threat perception played a key role in the translation of socioeconomic factors into xenophobia (Raijman, Semyonov, & Schmidt, 2003; Scheepers, Gijsberts, & Coenders, 2002; Verberk, Scheepers, & Felling, 2002). I argue that the perception that one group’s gain is another group’s loss shapes attitudes towards immigrants. The gain and loss various amongst different individuals depending on their socioeconomic position. Therefore, I hypothesize (H7) that the relationship between the individual characteristics and opposition towards immigrants (both immigrants from the same ethnic group and immigrants from different ethnic groups) will be partly mediated by threat perception in both countries. On the country level, I do expect to see differences. In Israel, fear of persecution, anti-Semitism, and constant war over resources and land are part of the national identity. The Jewish state of Israel was built primarily as a result of persecution. Jewish people living in North-African countries felt they were no longer able to live there and the rise of the Nazi party in Germany created the same feeling amongst Jewish people living in European countries (Ofer, 1995). In addition, the unstable continuing situation with the Palestinian creates insecurity feeling. In the Netherlands, the fear of persecution is not a part of the national identity, and there is no immediate security threat. Fear of persecution creates low tolerance towards out-groups. The need to keep a safe haven also generates a higher threat perception as a defense mechanism. Hence, I expect to see higher threat perception in Israel, threat perception, symbolic and realistic will have a stronger negative effect on opposition to immigrants from different ethnic groups in Israel (H8).
Data and Methods
The Dataset
The dataset for this research comes from the European Social Survey (ESS) round 8, conducted during 2016. The survey, which focuses mainly on attitudes, also contains background characteristics such as demographics and labor market positions (Markaki & Longhi, 2013). The main aim of the survey is to measure the attitudes, beliefs and behavioral patterns of the diverse European population and to examine citizen perspectives and judgments of key aspects of the society (The ESS Data Team at NSD, 2017). To achieve the coverage required there is a combination of repeated key items, which are the core section, in addition to rotating models. Another goal of the ESS is to achieve equivalence and make sure the data are comparable. Therefore, all the questions are translated for each language used by more than 5 percent of the population in the country (Sides & Citrin, 2007). A third goal of the ESS is to achieve maximum representation of the population in the country. Therefore, individuals are selected in a random probability method. The data are collected via hour-long face-to-face interviews using a standardized interviewing approach. Countries are weighted equally, so the results do not depend on the population size or the sample size (The ESS Data Team at NSD, 2017).
This research focuses on two countries out of the 23 that the ESS round 8 covered, Israel and the Netherlands. In Israel, the fieldwork period was between the 10th of September 2016 and the 8th of February, 2017. The geographic coverage for this round was the State of Israel and the Jewish population resident in the West Bank11. In Israel (IL=11) there are 2,557 valid interviews. In the Netherlands, the fieldwork period was between the 1st of September, 2016 and the 31st of January, 2017. The geographic coverage was the Netherlands, divided into 12 units. In the Netherlands (NL=13) there are 1,681 valid interviews (European Social Survey (2017), 2017).
My research is focused on the perception of the majority groups on immigration in Israel and the Netherlands. And so, I made a sample selection within the data. To select the majority group from the ESS, I excluded respondents who indicated that they are not from the majority group. To do this I used one question for the ESS round 8, Do you belong to a minority ethnic group in [country]?. I deleted all the respondents who answered Yes (N=627). The second sample selection I did was to look only at respondents who were born in the country, meaning they are not immigrants, assuming that if they are immigrants they may have different opinions
11The Palestinian residents of these territories are subjects of the Palestinian Authority and are not part of Israeli society neither by
on immigrants and immigration. I followed the question, were you born in [country]? I deleted all the respondents who answered No (N=723). In addition, missing analyses removed 352 observations, meaning 8.6 percent of the total observations. The final sample size of valid respondents is 1,472 in the Netherlands, and 1,636 in Israel.
The European Social Survey has advantages and limitations that need to be addressed. First, round 8 is the newest data available that measure public attitudes about immigration. Furthermore, it was conducted during 2016, after the major immigration “crisis” in Europe and Israel. The waves of immigrants and refugees at that time were followed by heated debates about the impact of these migrants and the responsibilities of the European countries (Holmes & Castañeda, 2016). Individual opinions are most likely influenced both by the actual immigration wave and by the debates over them. Hence, using the last round from 2016 is a significant advantage for this research. The second advantage is the quality of the data. The ESS has a secure amount of data collected. It has sufficient sample size and countries are weighted the same, which ensures that the data can be used to produce general statements about the general population. Third, since the ESS is designed to have optimal international comparability of the data, the adaption to each country ensures also the quality of the translation and the ability to compare between two or more countries (The ESS Data Team at NSD, 2017).
There are also limitations to the European Social Survey round 8. First, unlike round 7, which includes a rotating model that tested specific attitudes towards immigration and immigrants, round 8 has fewer questions on immigration, leaving this research with fewer indicators. Second, it is important to be aware of the limitation in prediction attitudes and beliefs when using cross-sectional data. In a cross-sectional study, data are collected on the whole study population at a single point in time. Therefore, it provides a snapshot of the situation at this given point in time with no past or future. Moreover, in cross-sectional data, the outcome and the cause are simultaneously assessed which makes it hard to determine what came first. There is a need to be careful when reaching any conclusions regarding cause and effect in the cross-sectional research (Knaub, 2015).
Methods
To test the hypotheses in this research, the data were analyzed using three methods; multivariable OLS regression, interaction model and mediation model. First, since I used continuous variables, I used multivariable OLS regression to estimate the direction and size of the effect the independent variables have on the dependent variables. At that point, in order to compare between the countries, I used interaction models and examined the different effects
each variable has in each country. Finally, I tested the mediation hypothesis using the four steps of the mediation model (Baron & Kenny, 1986). First, I established the correlation between the independent variables (individual characteristics) and the dependent variables (opposition towards immigrants) using OLS regression. Then I tested the causal variables correlation with the mediator. I examined the effect of education level and employment on realistic threat perception and the effect of emotional attachment to the country on the symbolic threat perception. Next, I showed that the mediator affects the dependent variables. Realistic threat and symbolic threat affects opposition to immigrants from the same ethnic group and different ethnic groups. The last step was to establish that threat perception mediates the relationship between individual indicators and attitudes towards immigrants by controlling for the mediation in the equation (Baron & Kenny, 1986). After running the regression in all the models, I checked for distribution of the dependent variable and linearity to validate the linear assumption using scatter plot. I also checked for multicollinearity by testing the correlation between all the independent variables using VIF (Mean VIF=1.23, See Appendix A).
Measurements
Dependent Variables
The dependent variables, based on previous studies in the Netherlands and Israel, measure how many and which newcomers individuals are willing to accept (Hanson et al., 2007; M McLaren, 2001; Quillian, 1995; Scheve & Slaughter, 2001; Sides & Citrin, 2007). I have two dependent variables using two questions from the ESS. One question tests opposition to immigrants from the same ethnic group: To what extent do you think [country] should allow people of the same race or ethnic group as most of [country]’s people to come and live here? The answer to this question is scaled from 1 (Allow many to come and live here) to 4 (Allow none). The second question tests opposition to immigrants from a different ethnic group is: How about people of a different race or ethnic group from most [country]? The answer to this question is scaled the same (European Social Survey (2017), 2017).
Independent Variables
The independent variables are personal background and demographic characteristics, based on different individual-level models connecting individual characteristics and attitudes towards immigrants. I divided the individual characteristics into two categories: the economic characteristics and the cultural characteristics (Mayda, 2006; Schneider, 2008).
For the economic indicators, I used the individual’s position in the labor market and his or her education level (Markaki & Longhi, 2013). The employment status is measured using the question: Employment status- Doing in the last 7 days: 'paid work’. The answers are recoded into dummy variables. Respondents are either employed = 1 (yes), or unemployed= 0 (no) (European Social Survey (2017), 2017). The individual level of education is measured by the highest level of education successfully completed (Markaki & Longhi, 2013). The question I used, from the ESS, is: What is the highest level of education you have successfully completed? The answers to this question are recoded from 1 (No education) to 5 (Tertiary education) (European Social Survey (2017), 2017). For the cultural characteristics, I used one question that measures people’s degree of closeness to their country (Mangum & Block, 2018). Respondents were asked: How emotionally attached do you feel to [country]? The answer to this question is scaled from 0 (Not at all emotionally attached) to 10 (Very emotionally attached) (European Social Survey (2017), 2017).
Mediation Variable
In this research, I used two central concepts of threat: realistic threat and symbolic threat based on the two amin theories, "Realistic Group Conflict Theory" and "Symbolic Threat" (Brader, Valentino, & Suhay, 2008). To measure the perception of realistic threat, I used one question from the ESS: Would you say it is generally bad or good for [country]’s economy that people come to live here from other countries? The answer to this question is scaled from 0 (immigration is bad for the economy) to 10 (immigration is good for the economy) (European Social Survey (2017), 2017). The answers to this question were recoded from 0 (immigration is good for the economy) to 10 (immigration is bad for the economy) in order to be able to measure the increase in economic threat. To measure symbolic threat, I used one question from the ESS and will refer to culture as containing symbols, norms, and values. Respondents were asked: Would you say that [country]’s cultural life is generally undermined or enriched by people coming to live here from other countries? The answers to this question were scaled from 0 (immigration undermines cultural life) to 10 (immigration enriches cultural life) (European Social Survey (2017), 2017). The answers to this question were recoded as well from 0 (immigration enriches cultural life) to 10 (immigration undermines cultural life) in order to be able to measure the increase in symbolic threat (Markaki & Longhi, 2013).
Control Variables
I used three control variables. The first one is Age. Attitudes may change over time and due to different life course depending on personal experience. Therefore, it is vital to add age as a control variable (Card, Dustmann, & Preston, 2005). This variable is between 15 years (the
minimum age in the ESS) to 97 years ( the oldest respondent) (European Social Survey (2017), 2017). The Second control variable is Gender. Since men and women tend to be affected differently by different social changes and tend to react differently to different situations, gender is an important control variable. Gender is coded (1=female 0= male). The third control variable is Living in an Urban Area. Because immigrants usually settle in big cities, the living area might reflect the chance of respondents to meet and socialize with immigrants in their daily life (Brenner & Fertig, 2006). This variable was recoded to be a dummy variable. The question I used is: Which answer best describes the area where you live? The answers were 1 (A big city) 2 (The suburbs or outskirts of a big city) 3 (A town or a small city) 4 (A country village) 5 (A farm or home in the countryside) (European Social Survey (2017), 2017). I recoded 1- 2 to 1 (living in an urban area) and 3-5 (living in the countryside or small town).
Results
Descriptive Statistics
Table 1 displays a descriptive overview of the variables for each country. The description of the variable can already hint at some differences between the countries that may contribute to the conclusions. As for the control variables, on average the respondents in Israel are younger, and the majority lives in urban areas. Regarding gender, in both countries, the respondents are divided equally.
Regarding the distribution of the independent variables. In the economic characteristic variables, there is hardly any difference between the two countries;12 the education level is almost the same in both countries when the average is lower secondary education (with Sd of approximately 1). Employment status has about equal distribution as well. On the other hand, the data show variation in the cultural variable. In Israel, the mean of national attachment is 8.6 (with Sd of 2.3) while in the Netherlands is 7.1 (with Sd of 1.8). Threat perception is higher in Israel than in the Netherlands. However, the differences are noteworthy only in the symbolic threat perception (symbolic threat perception in Israel = 4.6 and in the Netherlands =4).
Concerning the dependent variables, in the Netherlands, both opposition towards immigrants from the same ethnic group and immigrants from different ethnic groups is somewhat beneath the average. Yet, opposition towards immigrants from different ethnic groups is relatively higher. In Israel, there is hardly any opposition to immigrants from the same ethnic group (mean=1.5, with 1 being the lowest level on the scale). However, opposition towards immigrants from different ethnic groups is higher (the mean is almost 3 when the highest level on the scale was 4).
Multivariable OLS Regression Results- Hypothesis Testing
Table 2 shows the results of four multivariable structural models summarizing the first step of my research. Table 3 displays the results of the second step, interaction models. Table 4 and 5 (See Appendix B, C) presents the results of the mediation models, the third step. Before discussing the hypotheses, I will briefly discuss the results of the background variables. The effect of the control variable is somewhat similar in all the models tested in table 2. Age has no statistically significant impact except in one model (M6). Gender has no significant influence, but women have lower opposition. Living area has a substantial impact in all the models(p<.05);
people who live in urban area tend to be less opposed to immigrants. This fits other studies that had tested opposition to immigration (Mayda, 2006; Scheve & Slaughter, 2001).
Next, I turned to the finding on the hypotheses. My first hypothesis is that employment and higher education will have a positive influence on opposition towards immigrants (both immigrants from the same ethnic group and immigrants from different ethnic groups) (H1). Table 2 model 1 and 5 shows that the higher the education, the lower the opposition to both immigrant groups. Employment, on the other hand, has mixed results. Model 1 indicates that employed people have higher opposition to immigrants from different ethnic groups (b=0.023). Model 5 illustrates that working people have lower opposition to immigrants from the same ethnic group (b=-0.003). Nevertheless, both results are not significant. The results partially confirm the hypothesis.
Next, I hypothesize that the positive influence of employment and higher education on opposition towards immigrants will be the same in both countries (H2). The results in table 3 partially confirm my hypothesis. In both countries, education level has the same impact only on opposition towards immigrants from different ethnic groups (in Israel=-0.126, in the Netherlands =-0.142). Yet, the effect on opposition towards immigrants from the same ethnic group is not the same. In Israel, employment has a positive effect on opposition to both groups. In the Netherlands, employment reduces opposition towards immigrants from the same ethnic group and increases opposition towards immigrants from different ethnic groups.
Then, I ascertained significant and negative associations between national attachment and opposition towards immigrants from both groups (H3). The analysis had mixed results,
shown in table 2 model 2 and 6. On opposition towards immigrants from different ethnic groups, national attachment had a negative effect b=0.013 (model 2), while on opposition towards immigrants from the same ethnic group, national attachment had a positive effect b=-0.063 (model 6), people who indicate having strong attachment feeling had lower opposition towards immigrants from the same ethnic group as their own. Therefore, the third hypothesis was partially confirmed.
I further expected that the negative effect of national attachment on opposition towards immigrants is stronger in Israel than in the Netherlands. As indicated by the results in table 3, national attachment in Israel does have a stronger impact. Yet, even though national attachment has a substantial effect in Israel, on opposition towards immigrants from the same ethnic group the impact is positive (in Israel = -0.099 and in the Netherlands= -0.011), while the impact on opposition towards immigrants from different ethnic groups is negative (in Israel =0.042 and in the Netherlands= -0.004).
In hypothesis five, I estimated that the effect of individual characteristics would be different depending on the immigrant ethnicity (H5). Looking at table 2 model 5 and 10, education and employment have a positive impact on both immigrant groups. By contrast, the national attachment has a negative effect on opposition towards immigrants from the different ethnic groups and positive effect on opposition towards immigrants from the same ethnic group. Therefore, hypothesis 5 is correct only on for the individual cultural characteristics. In addition, the evidence in table 2 model 5 and 10 supports the hypothesis (H6) that in both countries, the opposition towards immigrants from different ethnic groups will be higher.
The mediation assumption was that the relationship between the individual characteristics and opposition towards immigrants (both immigrants from the same ethnic group and immigrants from different ethnic groups) will be partly mediated by threat perception in both countries (H7). To prove mediation effect, I use the four-step model by Baron and Kenny (1986). First, I established the relationship between the independent and dependent
variables as shown in table 2. After that, I demonstrated that the independent variable is correlated with the mediator (model 1 in table 5, see Appendix C). As expected, both education and employment reduce realistic threat perception. However, model 2 in table 4 (see Appendix B) displays that national attachment also reduces symbolic threat, while the opposite was expected. In the third step, present in Table 5 (See Appendix C), I verified that the mediator has effects on the dependent variables. Both symbolic threat and realistic threat increases the opposition towards immigrants from the same ethnic group and immigrants from different ethnic groups. The final step confirmed the partial mediation effect by controlling for the mediator (Baron & Kenny, 1986). Consistent with the results reported above, these indirect associations proved that threat perception has a mediation effect (H7). In table 2 model 3 and 8 realistic threat had decreases the impact of education and employment in reducing opposition by 50 percent. In model 4 and 9, symbolic threat has a mediation effect only between national attachment and opposition to immigrants from different ethnic groups.
Finally, I found a stronger negative effect of threat perception on opposition to immigrants from different ethnic groups in Israel (symbolic threat b=0.143 vs b=0.065 and realistic threat b=0.143 vs b=0.07), while in the Netherlands the different effects threat perception has on opposition to immigrants from the same ethnic group and immigrants from different ethnic groups, is not significant (H8).
Conclusions and discussion
This study set out to identify the characteristics in the native-born population in Israel and the Netherlands that have an impact on attitudes towards immigrants. It also aimed to examine the effect of economic and cultural individual characteristics on attitudes towards immigrants from the same ethnic group and immigrants from a different ethnic group. Moreover, the study wished to test if threat perception may be an additional explanation to the effect of individual indicators on the opposition towards immigrants. Finally, the study compared Israel and the Netherlands to explore the effect of contextual indicators on the correlation. I began my paper with a brief introduction to the historical and current immigration situation in each country, stating that without understanding the context, examining native-born opinions is misleading. Then, I briefly discussed the theoretical framework for pointing out possible explanations that may lead to different attitudes among different people. I further introduced the model of my research and the data I used to answer my research questions.
Individual- Level Analysis
When testing the effect of employment status on opposition to immigrants from different ethnic groups and immigrants from the same group (H1), employment reduces the opposition when looking at immigrants from the same ethnic group. Yet, it increases the opposition when looking at opposition to immigrants from different ethnic groups. That is, the economic position or the competition does not have a significant impact. If we argue that the labor market competition theory is the primary explanation, then people should care more for their self-interest and fear for their economic position and competition. Immigrants from different ethnic groups are also perceived as low-skilled immigrants. As a result, native-born should have higher opposition to immigrants that may compete with them over jobs or threaten their wages, meaning we should have seen higher opposition towards immigrants from the same ethnic group. And yet, this is not the case (Hello, Scheepers, & Sleegers, 2006). Attitudes do appear to be related to education level; the higher the education, the lower the opposition. This has proven to be a significant indicator in past research (Ceobanu & Escandell, 2010; Hainmueller & Hiscox, 2010). However, based on my results and the results of studies that tested similar questions, there is a higher chance that the positive correlation between education and lower opposition is because education tends to increase cultural and ethnic tolerance (Citrin & Wong, 1997; Espenshade & Calhoun, 1993; Espenshade & Hempstead, 1996).
As for the cultural indicators (H3), the higher the national attachment, the higher the opposition towards immigrants from different ethnic groups but not towards immigrant from the same ethnic group (See Appendix E). Hence, confirming that national attachment has a negative influence on opposition yet only towards immigrants from different ethnic groups. Using social identity theory, if social identity is correlated with national identity, and national identity means self- definition and belonging, then immigrants from different ethnic groups, unlike immigrants from the same ethnic group, get in the middle of this concept, and people with a higher attachment to the country will only have higher opposition to immigrants from different ethnic groups (Citrin & Wong, 1997; Manevska & Achterberg, 2013; Sides & Citrin, 2007).
I also found that individual characteristics have a different effect depending on the immigrant ethnicity (H5). Various researches have shown that individual attitudes toward immigrants are known to vary when attitude holders’ attention is directed to specific characteristics of immigrants such as race, ethnicity or class. Using Hagendoorn (1993) explanation of ethnic hierarchy, it is possible to assume that immigrants from different ethnic groups were ranked lower and had negative stereotypes attached that invoke higher opposition.
Moreover, in-group preference motivated by emotional and physical concerns will lead to ranking the out-groups based on how closely they are to the in-group, and how socially desirable they are (Bessudnov, 2016). Therefore, individual characteristics have a different effect on opposition to immigrants from different ethnic groups and immigrants from the same ethnic group.
My finding also implies that threat perception has a mediation effect on the correlation between opposition to immigrants and individual characteristics, and can be used as an additional explanation (H7). Threat perception among native-born, whether an individual threat or a group threat, appears to be the most supported theory in the literature so far. In line with other studies, when I added threat perception to the regression to test my hypothesis, the results validate the threat hypothesis (Berg, 2015; Brown, Vivian, & Hewstone, 1999; Hopkins, 2010). As many studies have indicated before, the need of the majority group to maintain the group’s status, resources, and power position leads to fear of the immigrant's costs to the nation or the individual (Manevska & Achterberg, 2013; Olzak, 1992). So, when immigrants enter the country, they immediately create perceived threat to the majority group, regardless the individual position (Citrin & Wong, 1997; Espenshade & Calhoun, 1993; Sides & Citrin, 2007).
Contextual Analysis
Contextual indicators can serve as an explanation for the relationship between national attachment and opposition to immigrants, higher and lower. But, it is hard to reach a direct conclusion on the effect on the relationship between economic indicators and opposition. First, because in both countries economic characteristics have mixed effects (H2). Second, because in both countries, the national and individual financial situation is almost the same according to OECD numbers (OECD, 2017a, 2017b). Additionally, the immigrant inflows have similar characteristics. In contrast, in Israel national attachment has a stronger negative effect on opposition towards immigrants from a different ethnic group(H4). This effect is different than in the Netherlands and lies in two contextual mechanisms. One, a history of persecution, years in the diaspora, and the present ongoing conflict with the Palestinians produces higher national concern and stronger national attachment. Second, same as Lee and Ottati (2002) that showed in their research that in-group favoritism and out-group prejudice significantly predict attitudes towards immigration. In Israel, the deep-seated and strong social and national identity is a way to ensure the survival of the majority. Strong national identity generates the feeling among the majority that they are superior, and this invokes negative attitudes and greater political hostility towards immigrants (Borowski & Yanay, 1997; Rosenhek, 2000; Semyonov et al., 2002).
