Structure and Determinants of the Ethnic
Hierarchy
A Study of 36 Ethnic Target Groups in the Netherlands
Master thesis by Hannah Soiné
University of Amsterdam Research Master Social Sciences
Student Number: 11237252 Email: hannah.soine@gmail.com
Supervised by: Dr. Bram Lancee, University of Amsterdam Second reader: Dr. Agnieszka Kanas, University of Amsterdam
Submission date and place: 10.08.2018, Amsterdam
Abstract: Using social distance towards ethnic groups as an indicator for prejudice, this study analyzes how and why ethnic prejudice varies across 36 ethnic groups in the Netherlands. To do so, a survey was conducted among the native Dutch working-age population that measures social distance towards the ethnic target groups. Ethnic target groups are ranked based on their social distance towards the native Dutch respondents. Contact with ethnic target groups and respondent’s level of right-wing authoritarianism are analyzed as potential determinants of this ranking. Results suggest that social distance increases across ethnic groups, with respondents perceiving the ingroup as least socially distant, followed by Western European, South American and Asian, African, East European and Middle Eastern groups. This ordering varies only slightly across different domains. Contact with the ethnic target group is associated with lower social distance, while high levels of right-wing authoritarianism in the respondent are associated with higher social distance.
1
Introduction
It is often relations between groups rather than individuals that govern social interactions. Social psychologists found that people favor members of the group they belong to over members of other groups, even if group membership is based on the most trivial differences like the outcome of a coinflip (Tajfel & Turner 1986). This is called outgroup prejudice and illustrates the importance of taking into consideration group memberships when analyzing social phenomena.
While a person can be a member of many different groups in society, this paper focuses on how ethnic group membership influences social life. Previous research has found that people are treated differently on the labor market (Bertrand & Mullainathan 2004; Booth, Leigh & Verganova 2011; Zschirnt & Ruedin 2016), in the doctor’s office (Derose et al. 2007; Ku & Matani 2001; Norredam et al. 2004) or on the housing market (Carpursor & Loges 2006) depending on which ethnic group they belong to. This suggests that prejudice and discrimination are rooted in ethnicity: Cultural and linguistic barriers between different ethnic groups often make interactions unpleasant compared to interactions within ethnic groups, which is why the latter are preferred by group members. This ingroup bias generally occurs together with outgroup prejudice, i.e. holding (mostly negative) stereotypes about outgroups threatening one’s ingroup (Aboud 2003). As a consequence, people tend to interact more with members of their ethnic ingroup and evaluate them more positively compared to members of ethnic outgroups.
However, not all outgroups are evaluated similarly negatively: Some ethnic groups have a lower status in society than others. This begs the question how people evaluate and why evaluations vary across ethnic groups. The concept of social distance is a promising approach to answer this question. It refers to “the degrees and grades of understanding and feeling that persons experience regarding each other” (Bogardus 1925a, p.299). The emphasis here lies on degrees
and grades: Social distance is a continuum, meaning that a person can perceive different social
distances towards different ethnic groups. Because a low social distance is associated with more positive and high social distance with more negative evaluations of an ethnic outgroup, an “ethnic hierarchy” (Hagendoorn & Hraba 1989; Hagendoorn 1995) is formed within a society. The ethnic ingroup is located on top of this hierarchy since social distance towards one’s ingroup tends to be small compared to ethnic outgroups. Unfortunately, the existence of an ethnic hierarchy may go hand in hand with discrimination against ethnic minority groups, particularly those at the bottom of said hierarchy.
2
It is important to investigate how and why evaluations of ethnic groups differ from each other. Only if a complete picture of the ethnic hierarchy in a country is available we can also investigate its structure and determinants. Focusing on said ethnic hierarchy to gain a better understanding of outgroup prejudice can benefit further research. Specifically, focusing on a group’s position in the ethnic hierarchy instead of the native-immigrant dichotomy often used in research about ethnic prejudice and discrimination results in a better fit between theory and data. For example, instead of using people’s attitude towards all immigrants to predict discrimination of Turkish immigrants on the labor market, now their attitude specifically towards Turkish migrants can be used instead. The results of this study have implications for policy makers as well: Ethnic groups in a country are different in terms of cultural and social background and they are treated differently by the ethnic ingroup, which policies need to take into account. Knowledge about structure and determinants of the ethnic hierarchy in the Netherlands can help policy makers with their decision about which groups to target and how to target them when they draw up policies. For example, with knowledge about the ethnic hierarchy it becomes easier to design programs aimed at improving the situation of the most disadvantaged ethnic groups because they can be easily identified.
In order to detect the ethnic hierarchy in a country and subsequently gain a better understanding of outgroup prejudice, existing research designs need to be expanded. Previous research suggests that different ethnic groups are discriminated against differently (Booth, Leigh & Verganova 2011; Zschirnt & Ruedin 2016). However, when investigating ethnic prejudice researchers only focus on relatively few ethnic groups, generally no more than six or seven at a time (Hagendoorn & Hraba 1989; Hagendoorn 1995; Snellman & Ekehammar 2005; Verkuyten & Kinket 2000; Verkuyten et al. 1996). This is insufficient since groups for which there is no data cannot be placed in the ethnic hierarchy. Additionally, inclusion of only such few ethnic groups may bias the results. For example, the ethnic hierarchy of Eastern European groups found by Hagendoorn (1995) would be interpreted differently if the left-out ethnic groups all ranked at the bottom of the hierarchy compared to if they were distributed throughout the hierarchy or all ranked at the top.
The first contribution of this study is therefore to gain a better understanding of how social distance varies across ethnic target groups in the Netherlands. What is the order of groups in the ethnic hierarchy of the Netherlands? How stable is it? Does it vary across different contexts? To answer these questions, social distance of respondents is measured towards a total of 36
3
ethnic groups. This is more than five times as many as previous studies have used, resulting in a more detailed and less biased picture of the ethnic hierarchy in the Netherlands.
A second contribution of this study is the insights it gives into determinants of social distance. Little is known about where social distance towards an ethnic group originates from. A study by Snellman & Ekehammar (2005) links it to social dominance orientation: People in whom this character trait is especially pronounced are more willing to rank ethnic groups than those who are low in social dominance orientation. Apart from that study, we rely on more general research about the emergence of prejudice to explain how social distance is generated: For example, more contact with the outgroup (Pettigrew 1998) as well as low levels of right-wing authoritarianism (Whitley 1999) have been linked to lower prejudice levels. This paper tests how these two characteristics change the perceiver’s evaluation of social distance.
Contact with ethnic outgroups is analyzed because it varies substantially across ethnic groups and therefore offers a promising explanation as to why these groups are evaluated differently. It is also a well-established determinant of other forms of prejudice (Allport 1954; Pettigrew 1998), making it a likely candidate for influencing social distance. Right-wing authoritarianism is analyzed because it is associated with willingly following authorities (Whitley & Kite 2009). Doing so can influence one’s perception of social distance since the ethnic ingroup (which constitutes the authority) likely promotes contact with and positive evaluations of the ingroup and distance towards and negative evaluations of ethnic outgroups.
To summarize, this paper presents and analyzes data on the ethnic hierarchy of 36 ethnic groups in the Netherlands and seeks to explain its determinants. In addition to updating and expanding research on the structure of the ethnic hierarchy in the Netherlands, this study aims to identify determinants of this hierarchy, which have scarcely been researched. The present paper attempts to fill this gap by investigating how two characteristics of the perceiver – namely the frequency of contact with an ethnic target group and their level of right-wing authoritarianism – influence how people perceive the ethnic hierarchy in the Netherlands. In doing so, the paper contributes to a better understanding of interethnic relations, which are crucial in contemporary societies.
Theoretical Background
Defining social distance
In his studies of social interactions, Robert E. Park (1924) noticed “grades and degrees of understanding and intimacy which characterize personal and social relations” (ibid., p.339). He
4
ascribed the lack of understanding and intimacy in interactions to “timidity and self-consciousness” and labeled it “social distance” (ibid.). Karakayali (2009) identifies four commonly used forms of social distance used in social science research. In this study, affective
social distance is used. This conceptualization emphasizes that social distance is subjective. It
is mainly based on the idea that “those who are socially close us are those we feel close to” (ibid., p.540, emphasis added). Therefore, social distance is short between people who are comfortable around each other, regardless of whether they share the same norms and habits or interact frequently.
Based on Park’s (1924) work, Emory Bogardus developed an instrument to measure affective social distance (Bogardus 1925a, 1933, 1947). He also investigated what explains variation in social distance between people. In one of his studies, respondents were asked how they developed antipathy towards an ethnic group (Bogardus 1925b). In most cases, even though they were explicitly instructed to avoid it, people based their reasoning on their negative experiences during only few encounters with members of a certain group, and then generalized them (ibid.). Prejudice is defined as “evaluations of an entire social group or of individuals because they are members of that group” (Whitley & Kite 2009, p.11) that “stem from reactions to the general social category” (ibid.), so Bogardus’s social distance scale can be considered a measure of ethnic prejudice.
Contemporary researchers continue using the tool Bogardus (1925a) developed – with some alterations modernizing or shortening it – to measure social distance (for examples see Hagendoorn 1995; Snellman & Ekehammar 2005; Verkuyten & Kinket 2000). Therefore, the present study also utilizes a modified version of this scale.
Linking social distance with ethnicity: The emergence of ethnic hierarchies
How can social distance help explain intergroup relations? Ethnic groups differ in terms of social distance people perceive towards them. According to Stephan & Stephan (1985), large differences between groups can lead to intergroup anxiety. The discomfort experienced during contact with outgroup members is ascribed to negative expectations about the outcome of an interaction. Reasons for such negative expectations can be previous interactions with the outgroup that have been perceived negatively or reliance on (negative) stereotypes when interacting with an outgroup (Whitely & Kite 2009). Interactions between different ethnic groups require more effort and are more prone to misunderstandings than interactions between people of ethnic groups who speak the same language and share similar norms and values.
5
Therefore, the social distance between people of different ethnicities is bigger than between people of the same ethnicity.
Hagendoorn (1995) explains the link between ethnicity and social distance through the status of ethnic groups: Ethnicity is linked to status in that it is “an indirect status criterion” (ibid., p.204) and low status is linked to negative stereotypes. These negative stereotypes are then linked to perceptions of higher social distance because they cause people to have negative expectations about the outcome of an interaction (Stephan & Stephan 1985). Therefore, social distance towards low-status ethnic groups associated with negative stereotypes is higher than the social distance towards high-status ethnic groups that have no such stereotypes associated with them, resulting in what Hagendoorn (1995) labels an ethnic hierarchy.
Previous empirical research suggests that ethnic hierarchies indeed exist in many different countries. As mentioned earlier, Bogardus (1925a, 1933) conducted several studies to validate and extend the social distance scale he developed. His results imply that there is a hierarchy of ethnic groups in terms of social distance towards them: Respondents in the United States felt more comfortable about contact with ethnic groups like the Scottish or Dutch compared to others like the Turkish or Japanese (Bogardus 1925b). Later on, other researchers have also found clear empirical evidence for the existence of ethnic hierarchies in the Netherlands, France, the United States and the former Soviet Union (Bessudnov 2016; Hraba et al. 1989; Hagendoorn 1995; Verkuyten et al. 1996).
The ingroup has generally been found to top the ethnic hierarchy, followed by Western European, Asian, Eastern European and Middle Eastern groups (Hagendoorn 1995; Snellman & Ekehammar 2005; Verkuyten & Kinket 2000). However, only a few groups from each region were used in these studies and from some regions like Africa or East Asia no groups at all were included. A hierarchy of 36 ethnic groups from different regions is an important source of new insights about the stability of an ethnic hierarchy: Does the rank order of regions hold if more than one or two groups are included per region and where are the lesser researched groups positioned in the Dutch society?
Previous research found that ethnic hierarchies are shared across ethnic groups within a country. For example, with the exception of putting their ingroup in first position, Ukrainians and Russians in the Ukraine rank the remaining three outgroups similarly (Hagendoorn 1995, p.221). The claim that ethnic hierarchies are somewhat universal in terms of who shares them is supported by findings that they are seemingly shared irrespective of gender (Snellman &
6
Ekehammar 2005) and level of prejudice (Verkuyten & Kinket 2000). They also at least partially transcend domains of contact: Hagendoorn (1995) found the same ethnic hierarchy when asking respondents how comfortable it would be for them to have members of different ethnic groups as friends and when he asked how comfortable respondents would be with them as marriage partners. The present study differentiates between three domains of contact in relation to social distance: Having a member of a specific ethnic group as (1) a neighbor, (2) a classmate for one’s child and (3) a colleague.
Most importantly, researchers find that there is a consensus about the ethnic hierarchy within as well as between ethnic groups in a country (Bessudnov 2016; Hraba et al. 1989; Hagendoorn 1995; Snellman & Ekehammar 2005; Verkuyten et al. 1996). Hagendoorn (1995) ascribes this surprising intergroup consensus to the different functionality of stereotypes for high- and low-status ethnic groups: High-low-status groups use negative stereotypes about the outgroup to “justify their advantages” (ibid., p.204) over them. They do so either due to their belief in a just world or because attribution bias leads them to attribute negative traits to lower-status groups. By contrast, low-status groups use negative stereotypes about other ethnic groups to fulfill their “need to distinguish themselves from ethnic groups in more or less similar positions” (ibid., p.205). These different functionalities of negative stereotypes lead both low- and high-status groups to arrive at the same ethnic hierarchy. Based on these theoretical considerations, I formulate the following hypotheses:
Hypothesis 1a: A hierarchy of social distance among ethnic groups exists within the Netherlands. The ingroup occupies the first rank, followed by Western European, Asian, Eastern European and Middle Eastern groups.
Hypothesis 1b: The ethnic hierarchy does not vary across the three domains of contact (neighbor, school and work).
Determinants of ethnic hierarchies
While there has been extensive research on the structure of ethnic hierarchies in different countries (see, for example Hagendoorn 1995; Verkuyten et al. 1996; Verkuyten & Kinket 2000), their determinants have received only sparse scholarly attention. Social distance is evaluated by ingroup members in relation to different outgroups. Therefore, social distance is
7
a consequence of two elements: The characteristics of the perceiver and the characteristics of the perceived. In this study we focus on the former1.
Characteristics of the perceiver can affect how one looks at members of other ethnic groups. They do so for example by influencing which characteristics are considered when evaluating contact with a member of an ethnic group. This difference in how comfortable one is with other ethnic groups is reflected in the social distance towards these groups. For example, previous research suggests that a person’s occupation (Bogardus 1925a) and level of social dominance orientation (Snellman & Ekehammar 2005) influence their perception of ethnic hierarchies. In this paper, two different individual-level factors are considered, namely how much contact the perceiver has with members of the outgroup and how high the perceiver is on right-wing authoritarianism.
Positive intergroup contact has been linked to lower levels of outgroup prejudice by Allport (1954) and later also Pettigrew (1998). The logic behind this relationship is simple: If people have personal contact with members of a particular group, they collect information and experiences related to them and do no longer rely solely on (negative) stereotypes in their evaluation of the group. A number of empirical studies support these assumptions (for a meta-analysis see Pettigrew & Tropp 2006).
Intergroup contact essentially determines what researchers call interactive social distance (Kadushin 1962; Karakayali 2009). The longer and more frequently people from different groups interact, the smaller the social distance between them. This type of social distance can but does not necessarily has to overlap with affective social distance as measured by Bogardus (Karakayali 2009). However, it is likely that people who have more positive interaction with each other also feel more comfortable with these interactions because there is less intergroup anxiety (Stephan & Stephan 1985). The following hypothesis emerges:
Hypothesis 2: The more personal contact someone has with members from an ethnic group, the smaller their social distance towards that group.
1 A different line of research is the analysis of characteristics of the perceived (i.e. the ethnic target group) as
determinants of the ethnic hierarchy. Because ethnic hierarchies rest on perceptions of social distance towards ethnic groups, it is likely that they are influenced by actual or perceived characteristics of that group. For example, a perceiver might feel comfortable with members of an ethnic group that is known for the relatively high socioeconomic status of its members, but uncomfortable with members of an ethnic group that is not.
8
Right-wing authoritarianism (short: RWA) is a set of attitudes defined by Altemeyer (1988) on the basis of Adorno et al.’s (1950) earlier research about what constitutes the authoritarian personality. It is characterized by three main traits: Authoritarian submission, authoritarian aggression and conventionalism (Altemeyer 1994, p.133). This means that people high in RWA are aggressive followers of established authorities. They will adhere to the norms and rules established by authorities and society and disapprove of individuals who break these rules. The “sister concept” of RWA, social dominance orientation (short: SDO), measures the belief that ingroups “should have a superior position in society and control over society’s resources” (Whitley & Kite 2009, p.238). Its relationship to social distance has been tested by Snellman & Ekehammar (2005) who found that it is tied to people’s inclination to rank ethnic groups. Since RWA and SDO are closely related, RWA may also influence social distance. Earlier research has shown that RWA is related to increased levels of prejudice against a number of different groups, including immigrants (Quinton et al. 1996). While it has not yet been evaluated in the context of social distance, people high in RWA likely have a bigger social distance towards outgroups than those low in RWA: If the ethnic majority group is recognized as an “authority”, people high in RWA will adhere to its norms and values and look down towards people who differ from them. Therefore, members of ethnic outgroups might be perceived as more of a threat to traditional norms and values and therefore more socially distant by people high in RWA compared to people low in RWA. At the same time, this willingness to submit to authority might also lead to a more biased perception of the ethnic ingroup, meaning that people high in RWA feel less social distance towards their own ingroup members than people low in RWA.
Hypothesis 3a: The higher a respondent is in RWA, the higher their social distance towards outgroup members.
Hypothesis 3b: When natives are being evaluated, higher RWA is associated with a smaller social distance towards that group. When non-natives are being evaluated, higher RWA is associated with a bigger social distance towards that group.
The following section of this paper details how the hypotheses formulated above were tested using data from a representative online-survey of the Dutch working-age population.
9
Method
Participants
To obtain the data for this study, an online survey of 1,017 participants was conducted in the Netherlands. The questionnaire was presented to respondents in Dutch and can be found in the appendix. The field period of this study was two and a half weeks in June 2018. Participants were rewarded with 0.5£ after completion of the survey.
All respondents and their parents were born in the Netherlands. Furthermore, only participants of working age (18-65 years) were sampled. Filter questions screened out people who did not meet these requirements. Respondents speeding through the questionnaire in less than 3.25 minutes were dropped from the dataset.
Table 1: Descriptive statistics for sample
Mean (or %)
Standard
deviation Minimum Maximum
Social distance (all domains) 3.310 1.470 1 7
Neighbor 3.567 1.635 1 7 School 3.234 1.598 1 7 Work 3.114 1.551 1 7 Age 44.27 12.88 20 65 Male 46.29% Education Primary 1.83% Lower secondary 24.27% Higher secondary 42.43% Tertiary 31.47% Employment status Employed 64.98% Self-employed 10.10% Unemployed 4.94%
Not active in labor force 19.98% Marital status
Never married 34.16%
Married, living together 58.00%
Divorced 6.02%
Separated 0.64%
Widowed 1.18%
Sample size 931
After also dropping cases with missing values on crucial variables2, the final sample size of this study amounts to N = 931. Gender, age, employment status and level of education quotas were
2 Cases were dropped if information about gender, age, educational status, employment status or marital status was
10
implemented to make the sample representative of the general Dutch working-age population. This is an important improvement compared to previous studies, which were mostly conducted with children or young adults. The average age of participants is 44.3 years and 46.29% of the final sample are males. More information about the composition of the sample, such as level of education, employment status and marital status can be found in Table 1.
Operationalization
Dependent variable: Social distance
The participant’s social distance towards ethnic target groups was measured with an item battery based on Bogardus’s (1925a) scale that was updated in terms of wording similar to Hagendoorn (1995) and Verkuyten & Kinket (2000). In three item batteries, respondents are asked whether they agree or disagree with the following three statements:
It would be pleasant for me (1) to have neighbors from… (2) if my children went to a school with pupils from… (3) to work together with colleagues from…
Every statement was followed by a list of ethnic target groups to evaluate. Respondents answered using a seven-point Likert scale ranging from strongly agree to strongly disagree. In order to prevent respondent fatigue, it was not feasible to present every respondent with all 36 ethnic groups in this study. Instead, five groups of six to nine countries that are somewhat geographically similar were formed. Every respondent was presented the ingroup (The Netherlands) and one randomly drawn country from each of the other five country pools to evaluate. An overview of the groups can be found in Table 2. Every respondent evaluated six ethnicities in the three domains of contact.
Table 2: Country pools from which six ethnicities were randomly drawn and presented to the respondent to evaluate
Pool Countries
Native The Netherlands
Western Europe France, Germany, Greece, Norway, United Kingdom, Belgium, Italy, United States, Spain
Eastern Europe Albania, Bulgaria, Poland, Romania, Russia, Macedonia South America & Africa Mexico, Suriname, Antilles, Ethiopia, Nigeria, Uganda
Asia China, Indonesia, Japan, South Korea, Vietnam, Malaysia, India Middle East & Pakistan Egypt, Iran, Iraq, Lebanon, Morocco, Turkey, Pakistan
Lastly, respondents who have answered the items assessing social distance with “don’t know” in least two out of three domains for all ethnic groups presented to them were excluded from the sample.
11
The countries were grouped and randomly drawn to keep the “comparison standard” between respondents as similar as possible even though they are presented with different groups. If the ethnic groups were not pooled, but randomly drawn from all 36 groups, someone presented with only Western European nations would likely rate e.g. Italy very differently from someone presented with Italy and five Middle Eastern countries. In contrast, it should make less of a difference in rating whether Italy is presented with Bulgaria, the Antilles, South Korea and Morocco compared to if it is presented with Poland, Nigeria, China and Iran.
Intergroup contact
To record contact with the ethnic target group, the measure of Hindriks et al’s (2014) was adapted and shortened. Participants indicate how often they have personal contact with members each of the ethnic target groups they were asked about in the social distance item batteries. Their answer options were never (1), yearly, quarterly, monthly, weekly and daily (7).
Right-wing authoritarianism
Right-wing authoritarianism was measured using a shorter and slightly less extremely-worded version of Altemeyer’s (1988) right-wing authoritarianism scale that Zakrisson (2005) developed and validated. It consists of 15 items, seven of which are reverse-scored. Respondents are asked to indicate on a seven-point Likert scale whether they strongly agree or strongly disagree with statements such as “There are many radical, immoral people trying to
ruin things; the society ought to stop them”. There was a don’t know option included in the
questionnaire. For analysis, the original scale was inverted so high values represent high right-wing authoritarianism. The Cronbach’s alpha of the scale in this sample is 0.748. Table 3 contains the item wording and other descriptive information.
Table 3: Items of the right-wing authoritarianism scale
Item wording Mean Std. dev.
Item-rest corr.
N
Our country needs a powerful leader, in order to destroy the
radical and immoral currents prevailing in society today. 3.057 1.691 0.443 5163
Our country needs free thinkers, who will have the courage to stand up against traditional ways, even if this upsets many people.
3.979 1.807 0.180 5184 The "old-fashioned ways" and "old-fashioned values" still
show the best way to live. 3.005 1.484 0.601 5265
Our society would be better off if we showed tolerance and
understanding for untraditional values and opinions. 3.494 1.592 0.435 5243
God’s laws about abortion, pornography and marriage must be strictly followed before it is too late, violations must be punished.
12
The society needs to show openness towards people thinking differently, rather than a strong leader, the world is not particularly evil or dangerous.
3.694 1.533 0.463 5194 It would be best if newspapers were censored so that people
would not be able to get hold of destructive and disgusting material.
5.393 1.620 0.246 5174
Many good people challenge the state, criticize the church
and ignore "the normal way of living". 4.049 1.446 0.113 4843
Our forefathers ought to be honored more for the way they have built our society, at the same time we ought to put an end to those forces destroying it.
3.252 1.497 0.557 5083
People ought to put less attention to the Bible and religion,
instead they ought to develop their own moral standards. 3.259 1.753 0.123 5240
There are many radical, immoral people trying to ruin
things; the society ought to stop them. 2.484 1.222 0.346 5255
It is better to accept bad literature than to censor it. 3.236 1.403 0.278 5068 Facts show that we have to be harder against crime and
sexual immorality, in order to uphold law and order. 2.455 1.405 0.453 5257
The situation in the society of today would be improved if
troublemakers were treated with reason and humanity. 3.926 1.568 0.179 5123
If the society so wants, it is the duty of every true citizen to
help eliminate the evil that poisons our country from within. 3.271 1.463 0.484 5029
Cronbach’s alpha 0.748 Note: Items in italics are reverse-scored.
Analysis
For the analysis, the data was transformed into long format, so that it contains six observations per respondent with one social distance rating each. Since there were cases in which respondents had only rated one or two of the ethnic groups and answered the rest with “don’t know”, the final sample size of ratings in this dataset is N = 5,295. An overview of the social distance ratings can be found in Table 4.
Repeated measures analysis of variance (ANOVA) and Bonferroni-corrected pairwise comparisons are used to determine whether there are significant differences between mean social distance scores of ethnic groups. Repeated measures ANOVA is used because in the long data format, the up to six observations per person can be considered repeated measures and are not independent of each other. To test whether the ethnic hierarchy is similar across the three domains of neighbors, school and work, the ordering of ethnic groups is compared between them. Large differences in the rank of an ethnic group in a specific domain compared to its rank in the aggregated social distance hierarchy across all domains are highlighted and discussed. Lastly, to investigate the effect of right-wing authoritarianism and contact with the ethnic target group on social distance, OLS regressions adjusted for clustered standard errors are used.
13
Standard errors are clustered within respondents since the transformation of the data to a long format results in several observations per respondent with the rating of one ethnic group each.
Results
In Table 4, the average social distance scores of respondents towards each of the 36 ethnic groups is displayed. Higher scores represent a higher social distance towards the group. Table 4: Social distance towards 36 ethnic groups (averaged across all three domains)
Mean Standard
deviation Min. Max. N
Netherlands 2.081 0.936 1 5 909 Belgium 2.528 0.998 1 6 102 United Kingdom 2.659 1.025 1 5 84 United States 2.844 1.210 1 7 81 Norway 2.982 1.204 1 7 101 Greece 3.031 1.195 1 7 85 Germany 3.050 1.133 1 7 101 Italy 3.085 1.198 1 7 102 Spain 3.107 1.211 1 7 115 France 3.120 1.170 1 7 110 Suriname 3.217 1.204 1 7 171 Japan 3.283 1.357 1 7 116 China 3.285 1.146 1 7 153 Indonesia 3.315 1.340 1 7 110 Antilles 3.324 1.284 1 7 142 Macedonia 3.458 1.372 1 7 159 Vietnam 3.583 1.403 1 7 131 South Korea 3.599 1.449 1 7 103 India 3.661 1.329 1 7 137 Mexico 3.686 1.173 1 7 122 Malaysia 3.697 1.492 1 7 125 Pakistan 3.710 1.559 1 7 112 Iraq 3.738 1.555 1 7 131 Uganda 3.799 1.557 1 7 135 Lebanon 3.811 1.501 1 7 126 Egypt 3.813 1.391 1 7 132 Albania 3.845 1.440 1 7 128 Iran 3.851 1.411 1 7 131 Russia 3.900 1.390 1 7 145 Turkey 3.903 1.518 1 7 134 Bulgaria 3.933 1.581 1 7 135 Nigeria 3.972 1.600 1 7 156 Romania 3.991 1.601 1 7 154 Poland 4.022 1.336 2 7 152 Ethiopia 4.105 1.550 1 7 152 Morocco 4.232 1.524 1 7 113 Total 3.310 1.470 1 7 5,295
14
With a mean social distance score of 2.081, natives are evaluated as the group that is socially closest to the respondents. This was expected since we only surveyed respondents with a Dutch ethnic background. By contrast, Moroccans receive the highest average social distance scores across all ethnic groups with a mean of 4.232. This is about double the distance towards the native ethnic group, showing that there is a large variation of social distance scores across ethnic groups.
In terms of geographical regions, the Western European groups are on top of the ethnic hierarchy with low average social distance scores. They are followed by Suriname, the Antilles and most of the countries in the “Asia” pool. African, Middle Eastern and Eastern European groups receive the highest average social distance scores. Previous studies found a similar ordering of geographical regions, even though not all of them contained all regions: The ingroup is generally the most favored group followed by Southern Europeans, the former colonies Suriname, Indonesia and the Antilles and then Middle Eastern countries and former Yugoslavia (Hagendoorn 1995; Snellman & Ekehammar 2005; Verkuyten & Kinket 2000).
Figure 1 depicts the social distance means and their 95% confidence intervals for all ethnic target groups. This graph suggests that the ingroup is the only group with a social distance mean significantly different from all others. The other confidence intervals overlap with the ones of other target groups, suggesting there is no significant difference between them in terms of social distance rating. However, there is enough of a gradient that even a large confidence interval like that of Pakistan does not overlap with all other confidence intervals, suggesting that the social distance towards Pakistanis is different from a few, but not all other ethnic groups.
15
Figure 1: Mean social distance scores of 36 ethnicities with 95% confidence intervals
Repeated measures ANOVA was used to test whether the differences between the social distance mean scores are significant. It indicates that there is a significant difference in social distance means between the 36 ethnic groups, F(35, 4350) = 80.65, p < 0.0001. Because so many different ethnic groups are included this information by itself is only of limited value. Using post-hoc Bonferroni-corrected pairwise comparisons reveal that some ethnic groups differ significantly from almost all others in terms of social distance. Overall, 34.4% of the pairwise comparisons are significant, i.e. the means of the two compared groups differ significantly from each other. For example, the mean social distance that respondents feel towards people from the Netherlands is significantly lower than the social distance towards all other groups except for Belgians and Brits. On the other side there are also some ethnic groups that differ significantly from less ethnic groups like Pakistan, which differs significantly only from five other ethnic groups (the Netherlands, Norway, the United Kingdom, Belgium and the United States). Especially within regions (or “country pools”) there are not many significant differences in the social distance means. This could be explained by the experimental design, i.e. by grouping them into country pools from which one is randomly drawn. After running another repeated measures ANOVA on social distance and the region (i.e. the country pool) of the ethnic group that was rated (F(5, 4380) = 511.71, p < 0.0001), the Bonferroni-corrected post
16
hoc pairwise comparisons revealed that all regions are significantly different from each other3 with only one exception: There is no significant difference in the social distance means of Eastern European countries compared to Middle Eastern countries. Both of them are located at the bottom of the hierarchy.
Table 5: Ethnic hierarchies across domains
Position All domains Neighbor School Work
1 Netherlands Netherlands Netherlands Netherlands
2 Belgium Belgium Belgium Belgium
3 United Kingdom United Kingdom United Kingdom United Kingdom 4 United States Norway United States United States
5 Norway United States Norway Italy
6 Greece Greece Germany Spain
7 Germany Germany Greece Greece
8 Italy Italy Spain Germany
9 Spain France Italy Norway
10 France Spain China Suriname
11 Suriname Indonesia Suriname France
12 Japan Japan France China
13 China Suriname Japan Antilles
14 Indonesia China Antilles Indonesia
15 Antilles Antilles Indonesia Japan
16 Macedonia Macedonia Macedonia Macedonia
17 Vietnam Vietnam India Korea
18 South Korea South Korea Vietnam India
19 India Malaysia South Korea Pakistan
20 Mexico Egypt Pakistan Vietnam
21 Malaysia Mexico Mexico Iraq
22 Pakistan Iraq Iraq Malaysia
23 Iraq Uganda Lebanon Lebanon
24 Uganda India Egypt Mexico
25 Lebanon Pakistan Malaysia Iran
26 Egypt Lebanon Iran Turkey
27 Albania Albania Uganda Uganda
28 Iran Iran Albania Albania
29 Russia Romania Bulgaria Russia
30 Turkey Russia Romania Bulgaria
31 Bulgaria Nigeria Russia Egypt
32 Nigeria Turkey Nigeria Poland
33 Romania Bulgaria Poland Nigeria
34 Poland Poland Turkey Ethiopia
35 Ethiopia Ethiopia Ethiopia Morocco
36 Morocco Morocco Morocco Romania
3 All pairwise comparisons are significant at the p < 0.05 level except for the Eastern European & South-America
17
The rank ordering of ethnic groups across the three domains of neighbors, school and colleagues is remarkably similar (see Table 5). The top three spots are always occupied by Netherlands, Belgium and the United Kingdom and Morocco and Ethiopia can be found in the bottom three. When compared to the rank order across all domains, only five of the 36 groups are off more than two ranks in the neighbor domain, five in the school domain and twelve in the work domain. In the neighbor domain, the total number of rank deviations from the average ranking of ethnic groups4 amounts to 40, in the school domain it is 46 and in the work domain 62.
In the neighbor domain, Indonesia, Romania and Egypt are more than two ranks above their rank across all domains and only India and Pakistan are more than two ranks below it. In the school domain, only China and Romania rank more than three places higher compared to their rank across all domains, while Malaysia, Turkey and Uganda rank more than three ranks below it. In the work domain, the groups that are rated more than three ranks above their average domain ranking are Italy, Spain, Turkey, Pakistan and Iran. The good position of the first three in the can be explained by the immigration history of these groups in the Netherlands: Many Italians, Spaniards and Turks came to the Netherlands in the 1960s as guest workers within the framework of agreements for labor recruitment between their countries and the Netherlands. This might have shaped stereotypes of these groups as particularly hard workers, which explains their relatively high rank in the work domain. When it comes to groups that are ranked lower in the work domain compared to their rank across all domains, no particular pattern emerges: Egypt, Norway, Mexico, Japan, Vietnam, Uganda and Romania are the seven groups more than three ranks below their average domain rank in the work domain.
To test which perceiver characteristics influence social distance, we estimated a series of regression models with standard errors clustered within respondents. In Table 6, the results are presented.
In model 1, only right-wing authoritarianism is included as an independent variable to predict social distance. Higher scores on the right-wing authoritarianism scale are related to larger social distance. This relationship is significant at p < 0.001. Model 2 uses contact with the ethnic
4 This was calculated by first taking the absolute value of the deviation of an ethnic group from the average position
in the ethnic hierarchy (e.g. Norway is ranked fifth in the hierarchy across all domains, but ninth in the work domain, meaning there is a four-rank difference between them). Then the deviations of all ethnic groups were summed for each domain.
18
group in question as the sole independent variable. As expected and contrary to RWA, contact with an ethnic group reduces social distance towards that group significantly.
Table 6: Regression models of social distance towards 36 ethnic groups (standard errors clustered within respondents)
Model 1 Model 2 Model 3 Model 4
Right-wing 0.423*** 0.366*** 0.504*** authoritarianism (0.044) (0.048) (0.053) Contact with -0.277*** -0.274*** -0.150*** ethnic group (0.013) (0.012) (0.019) Age (centered) 0.002 0.003 (0.003) (0.003) Gender (1 = Male) 0.092 0.066 (0.067) (0.067) Level of education (Ref.: Lower secondary) Primary -0.097 -0.187 (0.255) (0.273) Higher -0.069 -0.078 secondary (0.087) (0.087) Tertiary -0.140 -0.161+ (0.097) (0.097) Employment status (Ref.: Employed) Self-employed -0.075 -0.063 (0.119) (0.117) Unemployed -0.058 -0.001 (0.161) (0.165)
Not in labor force -0.135+ -0.128
(0.081) (0.082)
Marital status
(Ref.: Not married)
Married, living 0.079 0.081 together (0.077) (0.077) Divorced 0.207 0.212 (0.145) (0.145) Separated -0.334 -0.335 (0.269) (0.271) Widowed 0.532*** 0.539*** (0.133) (0.125) Native (1 = Natives) 3.491*** (0.309) Native X RWA -0.789*** (0.056) Constant 0.957*** 4.066*** 2.026*** 1.093*** (0.246) (0.057) (0.284) (0.313) N 5295 5295 5295 5295
Note:Standard errors in parentheses.
+ p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001
In model 3, both independent variables as well as the control variables age, gender, level of education, employment status and marital status of the respondent are added. Age is centered
19
around the sample mean. The effect of RWA on social distance decreases slightly after adding controls, but it remains positive and highly significant. Similarly, more frequent contact with the ethnic group is still associated with significantly less social distance towards that group after adding control variables to the model. The control variables themselves do not have much of an effect on social distance: Firstly, widowed people exhibit significantly higher social distance than those who are unmarried, but since the sample only contained very few widowed people not too much weight should be placed on this finding. Secondly, not being active on the labor market (e.g. students, permanently disabled, housewives) is associated with slightly lower social distance, even though this relationship is only significant at p < 0.1.
To test whether the influence of contact is constant across ethnic groups, an additional regression model with clustered standard errors was run: It includes social distance as the dependent variable and all control variables, but in addition also an interaction term between contact and ethnicity of the target group (reference category: Dutch). The results are depicted in Table A2 in the appendix. The main effect of contact on social distance remains negative and significant (p < 0.05). However, none of the interaction terms are significant, suggesting that the influence of contact on social distance does not vary with ethnicity, i.e. contact with someone French decreases social distance towards the French to a similar extent as contact with someone Japanese decreases social distance towards the Japanese.
In model 4, an interaction term between the right-wing authoritarianism score of the respondent and the migration background of the evaluated ethnic group was added to the variables from the previous model. The interaction term is negative and significant with p < 0.001. This indicates that while higher RWA is connected to higher social distance towards non-native ethnic groups, this relationship is reversed for the ingroup (b = 0.504 - 0.789 = -0.285): Respondents that are high in RWA actually exhibit less social distance towards natives than respondents who are low in RWA. This is in line with expectations as RWA captures willingness to conform and since there has been an uptick in popularism and prejudice against immigrants as of late. People who want to conform (i.e. are high in RWA) favor the ingroup more than those who do not want to conform and have a lower social distance towards natives. At the same time, they are also more prejudiced against ethnic outgroups and therefore have a bigger social distance towards them.
20
Discussion
This study found that Dutch respondents differentiate to some extent between the 36 ethnic target groups when evaluating social distance towards them: In partial support of hypothesis 1a, 34.4% of pairwise Bonferroni-corrected comparisons between the mean social distance towards ethnic groups are significant and suggest that the social distance mean of the groups are significantly different from each other. An ethnic hierarchy exists in which natives are situated on top, meaning the social distance respondents feel towards them is lowest compared to the other ethnic groups. This was expected since all respondents are Dutch natives and therefore should be biased in favor of their ingroup. Natives are followed by other Western European groups and then South-American and Asian groups in the ethnic hierarchy. The lower part of the hierarchy consists of Eastern European, African and Middle Eastern groups with respondents exhibiting an especially large social distance towards Moroccans and Ethiopians. This is in line with earlier research (Bogardus 1925b; Hagendoorn 1995; Verkuyten & Kinket 2000) but adds some new insights: This is the first study to include African groups and we observe that Ethiopians and Nigerians rank in the bottom sixth of the ethnic hierarchy in all three domains, while Ugandans are doing slightly better at an average rank 24. South Asian groups which – with the exception of Indonesians – have also not received attention with regards to their position in the ethnic hierarchy yet rank in the upper half of the hierarchy after the Western European groups. According to Hagendoorn (1995), this ordering can be explained by the social status of ethnic groups. The groups at the bottom of the hierarchy have a lower status and therefore more negative stereotypes associated with them. These negative stereotypes make communication with the group uncomfortable and increases social distance.
While there is a hierarchy among the ethnic groups, the consistency of that hierarchy across domains proposed in hypothesis 1b is not as strong as expected: While the ordering of groups in the ethnic hierarchy was mostly similar across the three domains neighbors, school and work, there were some noteworthy exceptions. Especially in the work domain, ethnic groups displayed differences in rank compared to the overall hierarchy. Overall, 29 of the 36 ethnic groups in the work domain did not correspond to their rank in the overall hierarchy across all domains. Especially ethnic groups with a longer history in the Netherlands (e.g. as guest workers) ranked higher in the work domain compared to other domains. Earlier studies that found no or only minor differences between domains. However, due to the low number of ethnic groups they include in their studies it is “harder” for ethnic groups to differ between domains if there are only ten instead of 36 different positions they could occupy.
21
As expected, characteristics of the ranking individual affect their social distance towards other ethnic groups: More frequent contact with an ethnic group is associated with lower social distance towards it, which supports hypothesis 2. This finding is in line with contact theory (Allport 1954), which states that interaction with an outgroup can reduce prejudice towards and anxiety to interact with that group. Frequent contact with an ethnic group leads respondents to rely less on (negative) stereotypes when evaluating a group since they have their own personal experiences to rely on instead.
Right-wing authoritarianism was also found to be related to social distance in this study. In support of hypothesis 3a, respondents high in RWA exhibit an overall higher social distance towards ethnic groups. However, RWA is related to higher social distance only when judging non-native groups. When the evaluated ethnic groups are the Dutch, the effect of RWA on social distance is reversed: Higher RWA is associated with lower social distance towards people with a Dutch background, which supports hypothesis 3b. This is in line with expectations as RWA captures the willingness to submit to authority and all respondents were Dutch natives. Therefore, people high in RWA are likely to be especially biased towards their ingroup and hostile towards ethnic outgroups, especially given the populist narrative of many mainstream media outlets that present ethnic outgroups as an economic and cultural threat to natives. The two main advantages this study design has over previous studies are twofold: Comparing 36 ethnic groups at once makes it possible to assess where in the hierarchy previously unresearched groups rank and how this new expanded hierarchy compares to the ones of previous studies. Secondly, because the general working-age population was sampled instead of only children or adolescents, the results are generalizable to a bigger population than their predecessors. Additionally, this study confirms that ethnic hierarchies are not only used by children and adolescents, but also shared among adults in the Netherlands. However, the tradeoff for this expanded study design in terms of ethnic target groups is that only people with a Dutch ethnic background could be surveyed. Therefore, no conclusions can be drawn about the generalizability of the results to ethnic minority groups in the Netherlands.
While contact and right-wing authoritarianism were thoroughly analyzed as determinants of social distance in the present study, there is a multitude of other potential factors that have not been analyzed in this paper and might affect social distance between ethnic groups. Further research should explore this avenue, especially with a focus on how the characteristics of the characteristics of the ethnic target group influences social distance towards them: The size of an ethnic group in a country and the year and circumstances in which the first wave of
22
immigrants of that group arrived might influence their position in the ethnic hierarchy. Alternatively, differences in cultural norms and values between in- and outgroup (e.g. how emancipated an ethnic group is believed to be) likely also influences a group’s position in the ethnic hierarchy.
Besides being a useful starting point for further researching these open questions, the results of this study can also potentially be helpful for improving integration efforts: They highlight ethnic groups like Ethiopians, Moroccans, Nigerians and Poles that would benefit the most from measures aimed at reducing stereotypes about them and fostering contact between them and natives. Such social distance-lowering measures can help to better integrate ethnic minority groups into society and lower discrimination rates of these groups.
References
Aboud, F. E. (2003). The formation of in-group favoritism and out-group prejudice in young children: Are they distinct attitudes? Developmental Psychology, 39(1), 48–60.
Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Stanford, N. R. (1950). The
Authoritarian Personality. New York, NY: Harper.
Allport, G. W. (1954). The Nature of Prejudice. Cambridge, MA: Perseus Books.
Altemeyer, B. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco, CA: Jossey-Bass.
Altemeyer, B. (1994). Reducing prejudice in right-wing authoritarians. In: The Psychology of
Prejudice: The Ontario Symposium (Vol. 7, pp. 131–148). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Bertrand, M., & Mullainathan, S. (2004). Are Emily and Greg more employable than Lakisha and Jamal? A field experiment on labor market discrimination. American Economic
Review, 94(4), 991–1013.
Bessudnov, A. (2016). Ethnic Hierarchy and Public Attitudes towards Immigrants in Russia.
European Sociological Review, 32(5), 567–580.
Bogardus, E. S. (1925a). Measuring Social Distance. Journal of Applied Sociology, 9, 299–308. Bogardus, E. S. (1925b). Social Distance and Its Origins. Journal of Applied Sociology, 9, 216–
226.
Bogardus, E. S. (1933). A Social Distance Scale. Sociology & Social Research, 17, 265–271. Bogardus, E. S. (1947). Measurement of Personal-Group Relations. Sociometry, 10(4), 306–
311.
Booth, A. L., Leigh, A., & Varganova, E. (2012). Does Ethnic Discrimination Vary Across Minority Groups? Evidence from a Field Experiment. Oxford Bulletin of Economics and
Statistics, 74(4), 547–573.
Carpusor, A. G., & Loges, W. E. (2006). Rental discrimination and ethnicity in names. Journal
23
Derose, K. P., Escarce, J. J., & Lurie, N. (2007). Immigrants and Health Care: Sources of Vulnerability. Health Affairs, 26(5), 1258–1268.
Hagendoorn, L. (1995). Intergroup biases in multiple group systems: The perception of ethnic hierarchies. European Review of Social Psychology, 6(1), 199–228.
Hagendoorn, L., & Hraba, J. (1989). Foreign, different, deviant, seclusive and working class: Anchors to an ethnic hierarchy in the Netherlands. Ethnic and Racial Studies, 12(4), 441– 468.
Hraba, Joseph, Hagendoorn, Louk, & Hagendoorn, Roeland. (2011). The ethnic hierarchy in The Netherlands: Social distance and social representation. British Journal of Social
Psychology, 28(1), 57–69.
Kadushin, C. (1962). Social Distance Between Client and Professional. American Journal of
Sociology, 67(5), 517–531.
Karakayali, N. (2009). Social Distance and Affective Orientations. Sociological Forum, 24(3), 538–562.
Ku, L., & Matani, S. (2001). Left Out: Immigrants’ Access to Health Care and Insurance.
Health Affairs, 20(1), 247–256.
Norredam, M., Krasnik, A., Sorensen, T. M., Keiding, N., Michaelsen, J. J., & Nielsen, A. S. (2004). Emergency room utilization in Copenhagen: a comparison of immigrant groups and Danish-born residents. Scandinavian Journal of Public Health, 32(1), 53–59.
Park, R. E. (1924). The Concept of Social Distance as Applied to the Study of Racial Attitudes and Racial Relations. Journal of Applied Sociology, 8, 339–344.
Pettigrew, T. F. (1998). Intergroup Contact Theory. Annual Review of Psychology, 49(1), 65– 85.
Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic test of intergroup contact theory.
Journal of Personality and Social Psychology, 90(5), 751–783.
Quinton, W. J., Cowan, G., & Watson, B. D. (1996). Personality and Attitudinal Predictors of Support of Proposition 187—California’s Anti-Illegal Immigrant Initiative. Journal of
Applied Social Psychology, 26(24), 2204–2223.
Snellman, A., & Ekehammar, B. (2005). Ethnic hierarchies, ethnic prejudice, and social dominance orientation. Journal of Community & Applied Social Psychology, 15(2), 83–94. Stephan, W. G., & Stephan, C. W. (1985). Intergroup Anxiety. Journal of Social Issues, 41(3),
157–175.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of Intergroup Relations (pp. 7–24). Chicago, IL: Nelson-Hall.
Verkuyten, M., Hagendoorn, L., & Masson, K. (1996). The Ethnic Hierarchy Among Majority and Minority Youth in The Netherlands. Journal of Applied Social Psychology, 26(12), 1104–1118.
Verkuyten, M., & Kinket, B. (2000). Social distances in a multi ethnic society: The ethnic hierarchy among Dutch preadolescents. Social Psychology Quarterly, 75–85.
Whitley, B., & Kite, M. (2009). The Psychology of Prejudice and Discrimination (2nd ed.). Boston, MA: Cengage Learning.
24
Whitley Jr, B. E. (1999). Right-wing authoritarianism, social dominance orientation, and prejudice. Journal of Personality and Social Psychology, 77(1), 126.
Zakrisson, I. (2005). Construction of a short version of the Right-Wing Authoritarianism (RWA) scale. Personality and Individual Differences, 39(5), 863–872.
Zschirnt, E., & Ruedin, D. (2016). Ethnic discrimination in hiring decisions: A meta-analysis of correspondence tests 1990–2015. Journal of Ethnic and Migration Studies, 42(7), 1115– 1134.
25
Appendix
A1: Questionnaire used for data collection (question numbers in bold and notes in italics added for convenience, not visible to respondents)
Beste respondent,
Bedankt dat u de tijd heeft genomen om deze vragenlijst in te vullen. Het zal u ongeveer 7 minuten kosten en uw antwoorden zullen ons helpen om te begrijpen hoe mensen tegen verschillende groepen in Nederland aankijken.
De verzamelde gegevens zullen uitsluitend voor wetenschappelijke doeleinden gebruikt worden en worden anoniem verwerkt. Indien u vragen of opmerkingen over de vragenlijst heeft, kunt u contact opnemen met Dr. Bram Lancee via b.lancee@uva.nl.
We starten deze vragenlijst met een paar vragen over uzelf.
Q1 Wat is uw geslacht?
Man Vrouw
Q2 Wat is uw geboortejaar?
Dropdown menu from 1900 until 2018
Q3 In welk land bent u geboren?
Nederland Anders
Q4 In welk land is uw vader geboren?
Nederland Anders
Q5 In welk land is uw moeder geboren?
Nederland Anders
Q6 Wat is uw hoogst voltooide opleiding?
Als u uw hoogste opleiding in het buitenland heeft voltooid, kies dan het Nederlandse niveau dat hier het meest mee overeen komt.
Basischool
Vmbo-b/k, mbo1, lts, mts, ambachtsschool, huishoudschool Vmbo-g/t, havo-onderbouw, vwo-onderbouw
Mbo2, mbo3, mbo4 Havo, vwo, hbs
Hbo-bachelor, wo-bachelor Hbo-master, wo-master, doctor Anders: ___________________
Q7 Welk van de volgende omschrijvingen beschrijft het best uw huidige situatie?
26 Zelfstandige
Student Werkloos
Permanent ziek of arbeidsongeschikt Gepensioneerd
Huishoudelijk werk, zorgen voor kinderen of andere leden van het huishouden Anders
Q8 Note: Only if answer to previous question was “Werknemer”, “Zelfstandige” or “Werkloos”: Wat is uw beroep?
Indien u meer dan een beroep heeft, geef dan uw hoofdberoep aan. ______________________________________
Q9 Wat is uw burgerlijke staat?
Ongehuwd
Getrouwd, samenwonend Gescheiden
Gescheiden van tafel en bed Weduwe of weduwnaar
De volgende vragen gaan over hoe prettig u zich voelt in situaties met mensen van verschillende ethniciteiten. Probeer uw eerste ingeving te volgen en niet te lang over uw antwoorden na te denken.
Q10 In hoeverre bent u het eens met de volgende uitspraak?
Ik zou het prettig vinden om buren te hebben afkomstig uit … Sterk mee eens Eens Enigszi ns mee eens Niet eens, niet oneens Enigszi ns mee oneens Oneens Sterk mee oneens Weet ik niet Nederland Etniciteit 1 Etniciteit 2 Etniciteit 3 Etniciteit 4 Etniciteit 5
Note: The order the ethnicities appear in was randomized. Because we cannot ask respondents to rate 30 ethnicities, each respondent gets to rate the Netherlands and one random ethnicity from each of five “pools”:
1. Frankrijk, Duitsland, Griekenland, Noorwegen, Verenigd Koninkrijk, België, Italië, Verenigde Staten, Spanje
2. Albanië, Bulgarije, Polen, Roemenië, Rusland, Macedonië
3. Mexico, Suriname, Nederlandse Antillen, Ethiopië, Nigeria, Oeganda 4. China, Indonesië, Japan, Zuid-Korea, Vietnam, Maleisië, India
27
5. Egypte, Iran, Irak, Libanon, Marokko, Turkije, Pakistan
Q11 In hoeverre bent u het eens met de volgende uitspraak?
Ik zou het prettig vinden als mijn kinderen naar school gaan met scholieren afkomstig uit …
Sterk mee eens Eens Enigszi ns mee eens Niet mee eens, niet mee oneens Enigszi ns mee oneens Oneens Sterk mee oneens Weet ik niet Nederland Etniciteit 1 Etniciteit 2 Etniciteit 3 Etniciteit 4 Etniciteit 5
See previous question for notes.
Q12 In hoeverre bent u het eens met de volgende uitspraak?
Ik zou het prettig vinden om samen te werken met collega’s afkomstig uit …
Sterk mee eens Eens Enigszins mee eens Niet mee eens, niet mee oneens Enigszins mee oneens Oneens Sterk mee oneens Weet ik niet Nederland Etniciteit 1 Etniciteit 2 Etniciteit 3 Etniciteit 4 Etniciteit 5
See previous question for notes.
De volgende stellingen gaan over de algemene publieke opinie met betrekking tot uiteenlopende sociale kwesties.
28
Q13 In hoeverre bent u het eens met de volgende uitspraken?
Sterk mee eens Eens Enigs zins mee eens Niet mee eens, niet mee oneens Enigsz ins mee oneens Onee ns Sterk mee onee ns Weet ik niet
Ons land heeft een sterke leider nodig om de radicale en immorele stromingen in de hedendaagse
samenleving te vernietigen. Ons land heeft vrije
denkers nodig, die de moed zullen hebben om tegen tradities in te gaan, zelfs als dit veel mensen stoort. De “ouderwetse manieren” en “ouderwetse waarden” laten nog steeds de beste levenswijze zien. Onze samenleving zou beter af zijn als we
tolerantie en begrip zouden tonen voor niet-traditionele waarden en meningen. Gods wetten over abortus, pornografie en het huwelijk moeten strikt nageleefd worden voordat het te laat is, overtredingen moeten bestraft worden.
Onze samenleving moet zich openstellen voor mensen die anders denken, in plaats van een sterke leider, de wereld is niet bijzonder slecht of gevaarlijk.
Het zou het beste zijn om kranten te censureren zodat mensen geen destructief en walgelijk materiaal in handen kunnen krijgen. Veel goede mensen dagen de staat uit, bekritiseren de kerk en negeren “de
normale manier van leven”. Onze voorouders zouden meer geëerd moeten worden voor de manier waarop ze onze
samenleving gevormd hebben; tegelijkertijd zouden we een eind moeten
29 maken aan de krachten die
haar vernietigen. Mensen zouden minder aandacht aan de Bijbel en religie moeten besteden en in plaats daarvan hun eigen morele standaarden moeten ontwikkelen.
Er zijn veel radicale, immorele mensen die proberen dingen te
verpesten; de samenleving zou hen hiervan moeten weerhouden.
Het is beter om slechte literatuur te accepteren dan te censureren.
De feiten tonen aan dat we harder moeten optreden tegen criminaliteit en seksuele onzedelijkheid om de rechtsorde te handhaven. De situatie in de
hedendaagse samenleving zou verbeterd worden als onruststokers met rede en menselijkheid behandeld zouden worden.
Als de samenleving dit van ons vraagt, is het de plicht van iedere echte burger om te helpen met het
elimineren van het kwaad dat ons land van binnenuit vergiftigt.
Q14 Hoe vaak heeft u persoonlijk contact met mensen uit…
Nooit Jaarlijks Driemaandelijks Maandelijks Wekelijks Dagelijks Nederland Etniciteit 1 Etniciteit 2 Etniciteit 3 Etniciteit 4 Etniciteit 5
Note: Same ethnic groups presented as in Q10 – 12.
Q15 In hoeverre bent u het eens met de volgende uitspraken?
Sterk mee eens
Eens Niet mee eens, niet
mee oneens
Oneens Sterk mee oneens
Weet ik niet
30 Ik ben trots op mijn
etnische achtergrond. Ik identificeer mezelf sterk met mijn etnische groep. Ik voel me sterk verbonden met mijn etnische groep. Mijn etnische identiteit is een belangrijk onderdeel van mezelf.
31
Table A2: Regression models of social distance towards 36 ethnic groups (standard errors clustered within respondents) with and without contact & ethnic target group interaction
Model 1 Model 2 Age (centered) 0.004 0.004 (0.003) (0.003) Gender (1 = Male) 0.097 0.093 (0.070) (0.069) Level of education
(Ref.: Lower secondary)
Primary -0.302 -0.315 (0.282) (0.284) Higher Secondary -0.125 -0.129 (0.088) (0.087) Tertiary -0.310** -0.314** (0.097) (0.096) Employment status (Ref.: Employed) Self-employed -0.086 -0.086 (0.122) (0.121) Unemployed -0.032 -0.027 (0.175) (0.174)
Not in labor force -0.148+ -0.149+
(0.083) (0.083)
Marital status
(Ref.: Not married)
Married, living together 0.172* 0.174*
(0.078) (0.078) Divorced 0.310* 0.320* (0.147) (0.147) Separated -0.326 -0.299 (0.328) (0.320) Widowed 0.552*** 0.546*** (0.141) (0.139)
Contact with ethnic group -0.155*** -0.170*
(0.021) (0.086) Ethnicity (Ref.: NLD) FRA 0.402** 0.118 (0.136) (0.552) DEU 0.493*** 0.269 (0.128) (0.559) GRC 0.304+ -0.134 (0.160) (0.570) NOR 0.210 0.355 (0.140) (0.569) GBR 0.063 -0.339 (0.130) (0.556) BEL -0.002 -0.034 (0.115) (0.554) ITA 0.430** 0.500 (0.133) (0.562) USA 0.184 0.024 (0.146) (0.562) ESP 0.472*** 0.599 (0.128) (0.546)
