This is an open access article distributed under the terms of the Creative Commons
Attribution License (Political polarization is a defining feature of early 21st century Western societies. This polarization is evident for many topics but especially in relation to questions on immigration (Eatwell & Goodwin, 2018). Research has determined that personal political orientation is consistently related to individual attitudes toward immigrants and immigration (e.g., Ceobanu & Escandell, 2010; Sides & Citrin, 2007). On the well-known left-right scale of political stance, people on the right are more likely to oppose immigrants and favour more restrictive immigration policies (Citrin & Sides, 2008; Hainmueller & Hiscox, 2010; McLaren, 2001). However, less is known about the nature of the subjective ideological beliefs that is responsible for the differential responses to immigrants and immigration of the right and the left. The core political values and beliefs and how closely they are connected to political orientation, has not been considered in research (but see Piurko, Schwartz, & Davidov, 2011). In their review Hainmueller and Hopkins (2014, p. 244) conclude that the research on immigration attitudes “has been to a striking extent unintegrated with broader research on the sources of political attitudes, and with research on political partisanship and ideology specifically”.
The present study conducted among a large national sample in the Netherlands examines the ideological dimensions of social conformity and acceptance of inequality, and considers education as a key indicator of political sophistication (Bennett, 1996; Delli Carpini & Keeter, 1996). Based on the social cognition model developed in political psychology (see Jost, 2006, 2017), it is examined whether social conformity and acceptance of group-based inequality are two core ideological aspects of political orientation that explain the right-left difference in attitudes toward immigrants and migration policies for refugees. More importantly, however, the central and novel hypothesis tested is that these two ideological aspects are more strongly part of the political orientation of the higher than the lower educated. This expectation is based on the ideological sophistication perspective which argues that higher education is related to a stronger ideological crystallization of political beliefs (Bennett, 1996; Delli Carpini & Keeter, 1996; Highton, 2009; Hillygus, 2005). Thus, the impact of political orientation on immigrant attitudes is expected to be stronger among people with a higher education because for them the endorsement of social conformity and acceptance of group-based inequality are more strongly part of the ideological meaning of their political orientation.
In general, more positive attitudes towards immigrants are found among the political left and also among higher educated people who tend to have a more cosmopolitan orientation, experience little economic competition and cultural threat from migrants, believe that migrants make a valuable contribution to society, and more strongly value openness, change and cultural diversity (Ceobanu & Escandell, 2010; Hainmueller & Hopkins, 2014). However, in contrast to this research that has examined the independent importance of political orientation and of education for people’s attitudes, we propose an integrated model in which the ideological clarity of people’s political orientation, which is associated with their attitudes, depends on their level of education. Thus, by testing the expectation derived from the ideological sophistication perspective we go beyond the existing empirical studies on people’s attitudes towards immigrants and immigration.
Political Orientation
Social psychological research on the social cognition model has argued and demonstrated that two stable, core aspects capture the most important differences between the politically right and left (Jost, Glaser, Kruglanski, & Sulloway, 2003). Personal political orientation is manifested in a specific ideological configuration in which respect for tradition and acceptance of inequality are central (Jost, 2006, 2017). The first dimension concerns attitudes toward cultural tradition and social deviance, and the second one relates to questions of (in)equality and egalitarianism. Individuals on the right end of the political spectrum tend to prefer traditions and social conformity, and to accept inequalities between individuals and groups. In contrast, those on the left end more strongly embrace socio-cultural change and equality. Thus, right-wing people, compared to left-wing people, are more prone to endorse conformity and to accept inequality. Extensive empirical research in political psychology and in different Western countries (Piurko et al., 2011) confirms that these two core dimensions capture the most important ideological differences between both political orientations (Jost, 2006, 2017; Schwartz, Caprara, & Vecchione, 2010; Wetherell, Brandt, & Reyna, 2013).
These differences can explain why people on the right will be more opposed to the presence of immigrants and less in favor of immigration. Research using the well-known left-right scale of political stance has shown that people on the right are more likely to oppose immigrants and favor more restrictive immigration policies (Citrin & Sides, 2008; Hainmueller & Hiscox, 2010; McLaren 2001; Sides & Citrin, 2007). Immigrants and immigration are typically perceived as causing socio-cultural change and to challenge existing status hierarchies (Pratto & Lemieux, 2001), which are difficult to reconcile with the value of social conformity and group-based (in)equality beliefs. Thus, a more right-wing political orientation can be expected to be associated with less positive attitudes toward immigrants and immigration policies because it involves an ideological configuration with a stronger preference for social conformity and greater acceptance of group-based inequality (H1).
The Role of Education
A robust empirical finding in Western societies is that higher education is associated with more tolerant attitudes toward ethnic and racial minorities, and immigrants and immigration policies (e.g., Ceobanu & Escandell, 2010; Citrin & Sides, 2008; Coenders & Scheepers, 2003; Kunovich, 2004; McLaren, 2001; Meeusen, de Vroome, & Hooghe, 2013). This association cannot be fully explained by the greater tendency of higher educated people to respond in a socially desirable way (Heerwig & McCabe, 2009; Ostapczuk, Musch, & Moshagen, 2009; Wagner & Zick, 1995). Rather, people with a higher education tend to have more privileged positions and therefore are less likely to face and perceive competition over scarce resources (e.g. housing, jobs) from minorities and immigrants (Scheepers, Gijsberts, & Coenders, 2002). Additionally, education is associated with cognitive ability and flexibility and promotes a more enlightened world view (Bobo & Licari, 1989; Meeusen et al., 2013; Ohlander, Batalova, & Treas, 2005) which makes higher educated people more tolerant of ethnic and racial minorities and to generalize the principles of tolerance to immigrants (Gaasholt & Togeby, 1995; Henry & Napier, 2017). Research in different countries has found, for example, that higher educated individuals have lower prejudices because they less strongly endorse system-legitimating ideologies (e.g., Carvacho et al., 2013; Dhont & Hodson, 2014).
Despite the clear evidence for a positive association between education and more positive attitude towards minority groups and immigrants there is also an important inconsistency. Studies find that higher educated majority members are not more, or even less, in support of social policies designed to overcome ethnic and racial group-based inequalities (Jackman, 1978; Wodtke, 2012). The ideological refinement perspective (Jackman & Muha, 1984) explains this inconsistency by viewing education as endowing majority members with more advanced cognitive skills and ideological commitments to explain the status quo. For example, the association between measures of racism and opposition to minority policies tends to increase with education (Federico & Sidanius, 2002; Sears, van Laar, Carrillo, & Kosterman, 1997). The better educated tend to have a more well-understood and sophisticated set of political attitudes than the less well-educated. Thus, education might shape the nature of the ideological underpinnings of personal political orientation. Education implies political socialization which involves a better understanding of the values and beliefs that underlie political-ideological differences that helps people identify the ideologies that best fit their personality (Converse, 1964; Osborne & Sibley, 2012). Education also increases cognitive ability and is a strong correlate of political sophistication (Bennett, 1996; Delli Carpini & Keeter, 1996; Highton, 2009; Hillygus, 2005). And at the country level it has been found that when the educational level in the population increases, the ideological sophistication also increases (Tedin, 1987).
Individual higher education is related to a stronger ideological clarity and coherence which implies a tendency to seek out and accept information supporting one’s own political position, and to disagree stronger with one’s political opponents on specific issues (see Bartels, 2008; Jacoby, 1988; Judd, Krosnick, & Milburn, 1981). Political polarization tends to be strongest among the most educated and the higher educated have stronger prejudicial attitudes toward those at the other side of the political spectrum (Henry & Napier, 2017). Further, the negative relationship between openness to experience and conservative policy support is stronger for those higher on political sophistication (Osborne & Sibley, 2015) and for higher educated people (Osborne, Dufresne, Eady, Lees-Marshment, & van der Linden, 2017). And the association between conservatism and negative attitudes toward immigration is especially pronounced among those who more frequently engage in political discussions (Sides & Citrin, 2007).
Because the ideological differences between the political left and right are more fully understood by the well-educated compared to the poorly educated, higher education amplifies the role of political orientation for social attitudes (Sniderman, Tomas, Piazza, Tetlock, & Kendrick, 1991). Research has found that the effect of political orientation on opposition to social policies for minorities increases as a function of intellectual sophistication (Reyna, Henry, Korfmacher, & Tucker, 2006; Sidanius, Pratto, & Bobo, 1996). Furthermore, the correlation between political conservatism and the endorsement of group-based inequality has been found to be stronger among higher compared to lower educated people (Sidanius et al., 1996). Higher political sophistication means that the core ideological beliefs of social conformity and group-based (in)equality can be expected to more clearly define the meaning of right versus left political orientation among the higher compared to the lower educated (H2). We examine this expectation by comparing the associations between personal political orientation and the two ideological beliefs among higher and lower educated individuals.
Immigrants and Refugee Policies
Ceobanu and Escandell (2010) argue that attitudes toward immigrants and toward migrant-related policies do not necessarily correspond and should be examined separately. Empirically both attitudes can be intermingled and closely related or rather be relatively independent. The degree of interrelation depends on the societal context and on whether both attitudes are measured in a similar way (i.e. scales used, and target group). In the current study we tried to differentiate between attitudes towards immigrants and attitudes towards policies by using different scales and different target groups. We assessed the attitude toward immigrants in terms of thermometer-like general feelings toward the four largest and most prominent immigrant-origin groups in the Netherlands: of Turkish, Moroccan, Surinamese and Antillean origin. Furthermore, for the attitude toward policies we used Likert-type scales and focused on people’s support for more restrictive policies for refugees which was strongly debated during the recent refugee crisis in Europe. In this way we try to distinguish the measurement of both attitudes which allows us to test the robustness of our findings across both outcomes that differ in measurement (group-based feelings and attitudes) and target group (immigrant-origin and refugees).
In Summary
The present study examines the predictions that a stronger endorsement of social conformity and acceptance of group-based inequality are both associated with less positive attitudes toward immigrants and refugee policies (H1), and, more importantly, that these two ideological beliefs are more strongly part of the political orientation of higher compared to lower educated people (H2). The latter hypothesis is based on the political sophistication proposition and implies that endorsing social conformity and accepting group-based inequality are expected to be more strongly defining parts of the ideological meaning of political orientation of people with higher than a lower educated.
We test these expectations among a national sample of majority members in the Netherlands. Similarly to other western countries, there are in this country strong debates on immigration and there is a pattern of political polarization with the rise of right-wing populist political parties. In the most recent polls (January, 2020) these parties attract around 20% of the votes, with similar percentages for the parties on the political left. In parliament, around 57% of seats are occupied by right-wing political parties and 43% by left-wing parties.
Currently, the four largest immigrant-origin groups constitute about 1.3 million people, or 7.6% of the Dutch population (Central Bureau of Statistics [CBS], 2019) and these groups are visibly different from the majority in terms of religion, cultural practices and/or skin color. Turks and Moroccans are mainly Muslim and have a history of labor migration. The Surinamese are Christians or Hindus from the former Dutch colony of Suriname, and Antilleans are from the Caribbean Islands. The Moroccans and Antilleans are perceived most negatively in the Netherlands and face the most negative stereotypes, mainly because these two groups are associated with normative deviance and relatively more often involved in criminal behavior (Andriessen, 2016; CBS, 2018). Between 2015 and 2019, the number of refugees who applied for asylum in the Netherlands was approximately 150 thousand.
Method
Data
In the spring of 2016 the data were gathered by I&O Research, a research agency that maintains a panel consisting of approximately 22,000 panel members. A random sample of 2,609 people was drawn from this panel and asked by I&O Research via e-mail for participation in the research. The respondents filled out an online questionnaire in February 2016. The response rate of 45% was similar to other research in the Netherlands (Stoop, 2005) and resulted in a sample of 1,168 respondents. The respondents’ age varied from 16 to 90 years (M = 52, SD = 15), and women were slightly underrepresented (47%). Since this study investigated the feelings and attitudes of native Dutch majority members, we excluded the respondents of whom at least one parent was not born in the Netherlandsi. A weight was applied to the sample resulting in the final sample (N = 1,155) that is representative for the Dutch majority population in terms of age, gender, education and region. For these demographic characteristics, the weighted sample differed very little from the original sample and was similar to the national statistics (https://www.cbs.nl/en-gb/society/population).
Measures
Negative feeling toward immigrant-origin groups was measured by the well-known feeling thermometers toward people of Moroccan, Turkish, Surinamese, and Antillean background. The four feeling thermometers were measured on 10-point scales. Using feeling thermometers with wider ranges of responses than Likert-type scales generates a more reliable measure (Alwin, 1997). This scale was reverse coded, so that the highest value represented the most negative attitude (ρ = .78).
Negative attitude toward refugee policies was measured with five Likert items (7-point scales; 1 ‘completely disagree’ to 7 ‘completely agree’). The respondents were asked to what extent they agreed with five statements on policies toward refugees. The items were ‘Regarding the refugee crisis, the attention should be primarily given to well-being of refugees’, ‘The Netherlands should continue receiving refugees’, ‘The Netherlands is perfectly capable of taking in more refugees’, ‘The Netherlands should close its borders as much as possible for refugees’ and ‘The policies regarding refugees should be much more restrictive’. The first, second and third item were reverse coded, in order for a higher value to represent a more negative attitude (ρ = .85).
Education was assessed in terms of the highest level completed or to current education level for those still studying, and it was measured in categories varying from 1 ‘no education/primary school’, 2 ‘lower secondary vocational training’, 3 ‘high school: first three years’, 4 ‘vocational training’, 5 ‘high school: pre-university education or propaedeutic certificate at University of Applied Sciences or University’, 6 ‘undergraduate degree at the University of Applied Sciences or University’, and 7 ‘graduate degree’. The level of education had a low correlation with political orientation (r = .07, p < .001). For the main analysis, education was categorized in three levels: low, medium and high education levelii. The first, second and third answer option constituted the category of low education. The fourth and fifth categories were recoded as medium education. Medium education was operationalized by the requirement of the Dutch Ministry of Education for students aged 18 to have obtained either a HAVO or VWO qualification – both part of the Dutch secondary education system that prepares the students for either the University of Applied Science or for University – or at least a MBO level 2 qualification – basic tertiary vocational training. The sixth and seventh answer option constituted the higher education category. Information about age and gender was available in the dataset and were included as control variables. Age was measured in years and gender was recoded in 0 ‘male’ and 1 ‘female’.
Political orientation was measured with the well-known and validated political self-placement scale (Jost, 2006). The question asked to the respondent was ‘Regarding politics in general, do you consider yourself as left-wing, center-left, center, center-right or right-wing?’. The answers were given on a 5-point scale with value 1 representing a left-wing political orientation and value 5 representing right-wing political orientation.
Social conformity was measured with four items (7-point scales) taken from previous research on attitudes toward immigrants in the Netherlands and Italy (Sniderman & Hagendoorn, 2007; Sniderman, Peri, Rui, De Figuerido, & Piazza, 2000). The items focus on the conformity aspect of right-wing authoritarianism and were, ‘One should better watch out for people who act differently than the mainstream’, ‘Rules are made to be adhered to, and people should not try to change them’, ‘People should conform to the conventional norms and rules’, and ‘Customs and traditions should be preserved and cherished instead of adjusted and changed’ (ρ = .69).
Acceptance of group-based inequality was measured with six items (7-point scales) based on a short version of the social dominance measure developed by Pratto, Sidanius, Stallworth, and Malle (1994) which was validated and translated to Dutch by Van Hiel and Duriez (2002). The items were ‘Some groups are simply not equal as compared to others’, ‘We should treat groups as equal as possible’ (reverse coded), ‘Some groups are simply more valuable than other groups’, ‘Equality between groups should be our ideal’ (reverse coded), ‘Sometimes it is necessary to hold back other groups to get ahead in life’, and ‘There would be fewer problems in society if all groups would be more equal’ (reverse coded) (ρ = .68).
Analytical Strategy
The analyses were conducted in Mplus Version 7.3 (Muthén & Muthén, 2010). With confirmatory factor analysis it was first examined whether the items of the different measures loaded on separate latent constructs. The measurement models were fitted using structural equation modelling with the estimator MLR (maximum likelihood estimation with robust standard errors) to accommodate non-normality of the measures. To test whether the measurement and structural model fitted the data properly, common fit statistics, such as Tucker-Lewis Index (TLI), comparative fit index (CFI), and the root mean square error of the approximation (RMSEA), were used. To test whether an adjusted model fitted the data better than the unadjusted model, the Sattora-Bentler Scaled Chi-square Test was used (Satorra & Bentler, 2010).
Measurement Model
The dependent variables – negative feelings toward immigrants and negative attitudes toward refugee policies - as well as endorsement of social conformity and acceptance of group-based inequality were treated as latent constructs. One adjustment was made in all factor structures. For the group-based inequality measure, the residual variances of the three recoded items (asking about equality) were allowed to covary and the residual variances of the other three items (asking about dominance) were also allowed to covary. Allowing these residual variances to covary accounts for the two sub-dimensions of equality and dominance (Hindriks, Verkuyten, & Coenders, 2014; Ho et al., 2012). The modified 4-factor model, with separate factors for the four constructs, provided the best fit for the data as compared to the other factor models (see Model E in Table A in the Supplementary Materials). Three modifications with regard to the residual covariances were made. Firstly, the covariance between the feeling thermometers toward Surinamese and toward Antilleans was relaxed. Due to their Caribbean origin and similar colonial history, Surinamese and Antilleans have more in common with each other than with the Turks and Moroccans who have a history of migrant labor. Secondly, the residual covariance between the feeling thermometers toward Moroccans and toward Antilleans was also relaxed. These two groups have in common that they are perceived most negatively in the Netherlands (Andriessen, 2016). Thirdly, the residual covariance between two items of the attitude toward refugee policies was relaxed. These items have in common that they both refer to more restrictive policy measures on border control. We furthermore tested whether the measurement of the latent constructs was comparable across the education levels of low, medium and high and the findings can be found in Table B in the Supplementary Materials.
Results
Descriptive Findings
The means and standard deviations of political orientation and of the latent variables obtained from Mplus are presented in Table 1. Further, the means of the different variables for the three educational levels are shown in Table 2, and the means for support for social conformity and for group-based inequality were tested against each other to examine whether there were significant differences between the three levels (Table C, in the Supplementary Materials). Both means differed significantly, with people with a high education demonstrating lower support for social conformity and for group-based inequality. Furthermore, higher educated persons had less negative feelings towards immigrants and a less negative attitude toward refugee policies (Table 2).
Table 1
Correlations for the Main Measures (N = 1,155)
| Variable | 1 | 2 | 3 | 4 | M | SD |
|---|---|---|---|---|---|---|
| 1. Negative feelings toward immigrants | - | 4.83 | 2.21 | |||
| 2. Negative attitudes refugee policies | .78*** | - | 3.35 | 0.92 | ||
| 3. Political orientation (left to right) | .34*** | .35*** | - | 3.02 | 1.01 | |
| 4. Social conformity | .42*** | .48*** | .24*** | - | 5.23 | 1.04 |
| 5. Group-based inequality | .56*** | .58*** | .29*** | .20* | 4.40 | 0.87 |
Note. Political orientation is a single self-placement scale, the others are latent measures.
*p < .05. ***p < .001.
Table 2
Means, Standard Deviation, and Range for the Main Measures per Education Level
| Variable | Range | Low education
|
Medium education
|
High education
|
|||
|---|---|---|---|---|---|---|---|
| M | SD | M | SD | M | SD | ||
| Negative feelings towards immigrants | 0-10 | 5.61 | 2.21 | 5.25 | 2.21 | 4.49 | 2.15 |
| Negative attitudes refugee policies | 1-7 | 3.72 | 0.86 | 3.40 | 0.95 | 2.94 | 0.94 |
| Political orientation (left to right) | 1-5 | 3.05 | 1.04 | 3.08 | 1.10 | 2.88 | 1.11 |
| Social conformity | 1-7 | 5.71 | 0.97 | 5.25 | 0.93 | 4.54 | 1.07 |
| Group-based inequality | 1-7 | 4.73 | 0.53 | 4.36 | 0.93 | 4.13 | 1.18 |
Note. nlow education level = 381; nmedium education level = 483; nhigh education level = 292.
Further, negative feelings toward immigrants and toward refugee policies were strongly associated, indicating that Dutch respondents who had more negative feelings toward groups of immigrants also favored more restrictive refugee policies (Table 1). Additionally, these feelings and attitudes were associated with social conformity, group-based inequality, and political orientation. A more right-wing political orientation was associated with higher support for social conformity and greater acceptance of group-based inequality in the full sample and among respondents with relatively high education but not among respondents with a low or medium level of education.
Structural Model
The standardized regression coefficients of the main associations are presented in Figure 1. Please note that the associations of political orientation with social conformity and group-based inequality are presented with lines instead of arrows to indicate that these two constructs are considered to constitute the two core ideological beliefs of political orientation. Two models were estimated: the first model with the main association without considering level of education, and the second model including this variable (see Tables D1 and D2, in the Supplementary Materials).
Figure 1
Standardized regression coefficients for the main paths of the model tested.
Note. The associations between political orientation and social conformity and group-based inequality are presented per education level (low, medium, high).
*p < .05. **p < .01. ***p < .001.
The first model estimated the associations between political orientation with its two underlying ideological beliefs and negative feelings toward immigrants and negative attitudes toward refugee policies simultaneously. As expected (H1), respondents with a stronger right-wing political orientation had more negative feelings toward immigrant-origin groups (β = 0.083, SE = 0.042, p = .023) and supported social conformity and group-based inequality more (β = 0.275, SE = 0.040, p < .001, and β = 0.312, SE = 0.051, p < .001, respectively). Stronger support of social conformity was associated with more negative feelings toward immigrants (β = 0.317, SE = 0.054, p < .001), and the same was found for the acceptance of group-based inequality (β = 0.489, SE = 0.059, p < .001). Furthermore, the association between political orientation and negative feelings toward immigrants was partly due to support for social conformity (βindirect = 0.087, SE = 0.018, p < .001), and acceptance of group-based inequality (βindirect = 0.153, SE = 0.032, p < .001). The findings were similar for the prediction of the attitudes toward restrictive refugee policies. For example, respondents with a more right-wing political orientation had more negative attitudes toward refugee policies (β = 0.100, SE = 0.040, p = .013) and this was partly due to stronger support of social conformity (βindirect = 0.097, SE = 0.019, p < .001) and to acceptance of group-based inequality (βindirect = 0.155, SE = 0.033, p < .001).
In the next analysis, a multi-group model for the three education levels (low, medium and high education) was estimated (Figure 1). This allowed us to test the political sophistication hypothesis (H2) that support for social conformity and for group-based hierarchy are more strongly part of the ideological configuration of the higher compared to the lower educated. There were indeed differences in the associations of political orientation with the support of social conformity and acceptance of group-based inequality across the education levels (Table E, in the Supplementary Materials). Wald tests indicated that the association between political orientation and social conformity differed significantly between respondents with a low education and a high education and between respondents with a medium education and a high education. Thus, as expected, support for social conformity was more clearly a defining aspect of the meaning of the right-wing political orientation among respondents with a higher level of education. A slightly different interaction effect was found for the association between political orientation and acceptance for group-based inequality. The association between a more right-wing political orientation with a higher score on group-based inequality differed significantly between respondents with a medium education level and with a high education level, but less so between respondents with a low and a high education level.
Discussion
Many studies have examined majority members’ attitudes towards immigrants and immigration policies in relation to people’s political orientation and level of education as two independent predictors. We tried to go beyond this research by focusing on an integrated model that proposes that the ideological configuration of political orientation is more well-developed among people with a higher level of education. Specifically, based on the ideological sophistication proposition we tested whether political orientation is more clearly defined in terms of social conformity and group-based inequality beliefs among higher educated individuals compared to lower educated individuals. Education is a strong correlate of political sophistication (Bennett, 1996; Delli Carpini & Keeter, 1996; Highton, 2009; Judd et al., 1981) and has similar effects on social attitudes as sophistication (Osborne et al., 2017). Political sophistication helps people to identify policies that suit their personality and involves a crystallization of political attitudes with a tendency toward political polarization (Bartels, 2008; Henry & Napier, 2017; Jacoby, 1988).
We found that higher educated participants support social conformity and group-based inequality less than lower educated people, which correspond with the somewhat higher left-wing political orientation of the higher educated participants. Yet, both ideological beliefs were more strongly linked to personal political orientation of higher compared to lower educated individuals. This indicates that education does not only affect the average endorsement of ideological political beliefs, but importantly also seems to help people to connect their political orientation more closely to their personal values about sociocultural traditions and (in)equality (Jacoby, 1988; Jost 2017). This finding is similar to research demonstrating that the negative association between openness to experience and support for conservative policies is stronger for higher educated people (Osborne et al., 2017) and the more politically sophisticated (Osborne & Sibley, 2015; Reyna et al., 2006; Sidanius et al., 1996). The current finding that the explained variance of the structural model was larger for the higher compared to the lower educated participants further supports this interpretation.
In agreement with research on the social cognition model (Jost, 2006, 2017) and with cross-national research in Western societies (Piurko et al., 2011), we found that the importance attached to social conformity and the acceptance of group-based inequality were two separate aspects of the ideological configuration of the right-left political stance. Thus, although the meaning of a left-wing and right-wing political orientation might change over the years, might not fully capture the political cleavages that characterize many western societies (Middendorp, 1992), might differ for economic and for cultural issues (Knutsen, 1995), and might not be fully similar across countries, there were two ideological aspects distinguishing between left-wing and right-wing people: beliefs about conformity and traditions and beliefs about inequalities. People on the political right tend to more strongly endorse social conformity and to accept inequalities, whereas a left-wing orientation is associated with openness to change and preference for greater equality (Jost, 2017). Furthermore, both ideological aspects were associated with negative feelings toward immigrant-origin groups and negative attitudes toward policies for refugees. This indicates that concerns about immigrants and refugees causing socio-cultural changes and challenging existing structural relations underlie the right-left difference in these attitudes.
Yet, the findings indicate that significant direct associations between political orientation and the attitudes toward immigrants and refugee policies remained. This can be interpreted in three ways. First, our measures did not assess the full breath of the core ideological aspects pertaining to attitudes toward tradition versus social change and attitudes toward inequality versus egalitarianism (Jost, 2006). For example, we assessed tolerance of social dissenters and not the importance attached to tradition and order. And we assess the acceptance of group-based inequality and not equalitarianism or defense of the status quo. Second, we examined political orientation in terms of the associations with its two core aspects but ideological sophistication can, for example, also be conceptualized in terms of consistency across different political issues and domains. Those with more developed political views can be expected to respond more consistently to different societal issues and developments. Third, according to Jost (2006) political orientation consists of deeply held moral values (core features) together with beliefs that are more changeable and that relate to specific issues (peripheral features; such as immigration). This means that attitudes toward immigrants and refugee policies will not only depend on the underlying ideological configuration but also on the specific societal concerns and current public debates about questions of immigration.
Using different target groups and measurement scales it was possible to empirically differentiate between feelings toward immigrant-origin groups and attitudes toward policies for refugees. This supports the notion that these are distinct domains of study (Ceobanu & Escandell, 2010). However, we also found that feelings toward immigrants and attitudes toward refugee policies were strongly correlated and that the pattern of associations was similar. For example, acceptance of group-based inequality was a stronger predictor than endorsement of social conformity for both negative feelings toward immigrants and negative attitudes toward refugee policies. Thus, although there are important legal differences between categories of migrants, in public perception the term might also include refugees with the result that attitudes toward immigrant-origin groups and refugee policies overlap. In the context of the Netherlands, or in the period directly after the so-called refugee crisis, these attitudes appear to have a lot in common and future research could examine when and why these two attitudes might be less or more strongly connected.
Future research should also examine the generalizability of the present findings to other national contexts. We focused on the Netherlands and although the importance of ideological beliefs for people’s political orientation appears to be quite similar across Western societies (Piurko et al., 2011), there are country differences related, for example, to immigrant groups and issues of immigration and related policies more generally that might be important to consider (e.g., Mayda, 2006; Valentino et al., 2019). For example, whereas the expected difference in ideological meaning between higher and lower educated participants was clearly supported for the value of social conformity it was somewhat less clear for the acceptance of inequality, and for the highest and lowest educational category in particular. This might be due to the Dutch context in which inequalities are less visible and less accepted than in for example the United States or Great Britain (Eatwell & Goodwin, 2018).
Furthermore, we used cross-sectional data which means that we cannot draw firm conclusions about the direction of influence. It is also possible that individuals with a negative attitude towards immigrants and refugee policies develop a stronger and more coherent right wing political orientation. However, research indicates that political orientation with its underlying ideological beliefs is a rather stable individual attribute that is formed relatively early in life and that underlies people’s attitudes toward a range of issues (e.g., size of the government, military spending, foreign policy) (Jost, 2017; Rico & Jennings, 2016). Thus, the direction of influence that we tested seems more likely than the reversed direction of influence.
Finally, future research could examine the ideological sophistication proposition among immigrant and minority groups. It is likely that also among minority groups higher education is related to a stronger ideological crystallization of political views that have an impact on social attitudes. For example, research on the so-called integration paradox has found that more highly educated immigrants develop more negative attitudes towards the majority because they are more aware of, and have a better understanding of, processes of discrimination and reduced opportunities in society (Verkuyten, 2016; Wodtke, 2012). Education enables immigrants and minorities to become more informed social critics who can seek to challenge discrimination and advocate policies that redress group disadvantages.
Conclusion
In agreement with the ideological sophistication proposition we have demonstrated that the ideological meaning of political orientation depends on the level of education. Although higher educated individuals less strongly value social conformity and accept group-based inequality less, their political orientation was more strongly connected to these two core ideological aspects (Jost, 2006, 2017). The findings suggest that political orientation is ideologically more clearly defined among higher compared to lower educated majority members, and that the corresponding ideological beliefs are associated with attitudes towards immigrants and refugee policies. These findings add to the existing literature on negative attitudes towards immigrants and refugees by considering the core ideological beliefs of people’s political orientation and how the meaning of this orientation differs between individuals with a higher and lower level of education.