Testing a New Theoretical Model for Attitudes Toward Immigrants: The Case of Social Workers' Attitudes Toward Asylum Seekers...
Article in Journal of Cross-Cultural Psychology · November 2015
DOI: 10.1177/0022022115613860
CITATIONS
5
READS
79
2 authors:
Some of the authors of this publication are also working on these related projects:
HBSC StudyView project Eugene Tartakovsky Tel Aviv University
40PUBLICATIONS 364CITATIONS
SEE PROFILE
Sophie Walsh Bar Ilan University
52PUBLICATIONS 500CITATIONS
SEE PROFILE
All content following this page was uploaded by Eugene Tartakovsky on 17 December 2015.
Journal of Cross-Cultural Psychology 2016, Vol. 47(1) 72 –96
© The Author(s) 2015 Reprints and permissions:
sagepub.com/journalsPermissions.nav DOI: 10.1177/0022022115613860
jccp.sagepub.com
Article
Testing a New Theoretical Model for Attitudes Toward Immigrants:
The Case of Social Workers’
Attitudes Toward Asylum Seekers in Israel
Eugene Tartakovsky
1and Sophie D. Walsh
2Abstract
The present study proposes a new threat–benefit theoretical model explaining attitudes of local people toward immigrants. Based on the theory of human values and extending Integrative Threat Theory, the threat–benefit model suggests that the local population perceives immigrants as both threatening and beneficial for the receiving society. The model assumes that appraisal of an immigrant group as threatening or beneficial for the receiving society influences opinions regarding immigration policy related to the immigrant group. The study assessed the new model investigating attitudes toward asylum seekers in a representative sample of 283 social workers in Israel. Results of the study support a conceptualization of immigrant appraisal involving four types of threats (economic, physical, social cohesion, and modernity) and four types of benefits (economic, physical, cultural diversity, and humanitarian), which represent different types of realistic and symbolic threats and benefits. Findings showed that appraisal of asylum seekers as beneficial or threatening to the receiving society mediated the connections between personal preferences for values of universalism, power, social security, and tradition and support for immigration policy directed either at defending immigrants’ rights or defending the receiving society. Application of the model for understanding attitudes toward different minorities as well as for creation of value-based interventions and programs aimed at reducing negative attitudes toward various stigmatized groups in society are discussed.
Keywords
threat–benefit model, attitudes toward immigrants, immigration policy, personal value preferences, asylum seekers, Israel
The stranger that dwelleth with you shall be unto you as one born among you, and thou shalt love him as thyself; for ye were strangers in the land of Egypt.
—Leviticus 19:34
Introduction
The present study develops a new theoretical model that aims to explain local people’s attitudes toward immigrants. The model conceptualizes attitudes toward immigrants as consisting of two
1Tel Aviv University, Israel
2Bar Ilan University, Ramat Gan, Israel Corresponding Author:
Sophie D. Walsh, Department of Criminology, Bar Ilan University, 5290002, Israel.
main elements: the appraisal of an immigrant group as both potentially beneficial and threatening to the receiving society, and opinions regarding different aspects of immigration policy, which are directed at either defending immigrants’ rights or defending the receiving society from the immigrants. The model further assumes that the appraisal of the immigrant group may influence local people’s opinions regarding the desired immigration policy for this group. Finally, the model conceptualizes the predictive nature of personal value preferences of members of the receiving society on their attitudes toward immigrants. Specifically, the model assumes that personal value preferences of members of the receiving society predict their appraisal of immigrants and, through this, their opinions regarding immigration policies. The threat–benefit theoretical model is tested investigating the attitudes of social workers toward asylum seekers in Israel.
The decision to focus on the attitudes toward asylum seekers was guided by the fact that the number of asylum seekers in the world is rapidly growing (Kritzman-Amir, 2012; Yaron, Hashimshony-Yaffe, & Campbell, 2013). Recent events in Europe and around the world have placed the issues surrounding asylum seekers and refugees in the forefront of news and politics.
Debates on policy regarding asylum seekers are heated both in Israel and in the world (Stratham, 2003). As such, an understanding of predictors of attitudes held by the receiving population toward asylum seekers may provide important suggestions for improving attitudes toward this group and thus promoting good intercultural relations all over the world. At the beginning of 2015, there were around 50,000 asylum seekers living in Israel. Ninety percent of them are from Eritrea and Sudan, and the rest are from other African countries (IPA, 2015). Asylum seekers constitute the most culturally distant immigrant group in Israel, and Israeli mass media has repeatedly reported conflicts between them and local people (African Refugee Development Center, 2012; Kritzman-Amir, 2009). Public debate in Israel has been heated around whether asylum seekers should be strongly encouraged to return to their countries of origin and around the issue of detainment centers for asylum seekers. However, despite the paramount importance of the issue, asylum seekers remain the least studied immigrant group, and to the best of our knowledge, no previous studies have focused specifically on the host society’s attitudes toward asylum seekers.
The present study focuses on social workers because this professional group, along with other human services professionals (e.g., educators, law enforcement officers, and medics), is at the forefront of working with asylum seekers. As most social workers are placed in the government or governmentally subsidized agencies, they are responsible for distribution of state welfare resources (Ayalon, Kaniel, & Rosenberg, 2008). In addition, in Israel and in many other countries, social workers affect policy regarding asylum seekers, through different nongovernment bodies working with asylum seekers, court rulings, and through parliament commissions (Kritzman-Amir, 2012).
Thus, the position of social workers in their contact with asylum seekers may be ambivalent:
On one hand, they are supposed to help asylum seekers to adjust and to defend their rights; on the other hand, they are supposed to defend the interests of receiving society (Ayalon et al., 2008).
As such, the focus on social workers may permit a better understanding of the complex system of attitudes toward immigrants existing in society.
Theoretical Background
Attitudes toward immigrants. Researchers differentiate between two aspects of attitudes toward immigrants: perception of immigrants by local people in the receiving country (Stephan, Diaz- Loving, & Duran, 2000; Stephan & Stephan, 1996) and the local people’s opinions regarding immigration policy (Berry, 1997; Bourhis, Moise, Perreault, & Senecal, 1997; Burns & Gimpel, 2000; Gorodzeisky, 2013; O’Rourke & Sinnott, 2006).
Perception of immigrants by local people. One of the most influential theories describing how local people perceive immigrants is Integrative Threat Theory (ITT; Stephan & Stephan, 1996).
This theory delineates four types of threats that immigrants (as an out-group) may possess for local people: realistic threat (competition for resources), symbolic threat (resulting from perceived incompatibility in cultural values), intergroup anxiety (out-group fear), and negative stereotypes.
Most of the empirical studies based on ITT have focused on the effects of realistic and symbolic threats on the emotions and behavior of local people toward immigrants.1 The studies have dem- onstrated that negative attitudes and emotions, as well as local people’s support for discriminative actions against immigrants, are related to perceiving them as a threat (Croucher, Aalto, Hirvonen, Sommier, & Hirvonen, 2013; Dustmann & Preston, 2007; Stephan et al., 2000; Stephan, Lausanne Renfro, Esses, White Stephan, & Martin, 2005; Stephan & Stephan, 1996; Velasco Gonzalez, Verkuyten, Weesie, & Poppe, 2008). In addition, it has been found that lower socioeconomic status (SES), lower education, higher religiosity, older age, and rightist political orientation are associated with a higher sense of immigrant threat (Berry, 1997; Bourhis et al., 1997; Burns &
Gimpel, 2000; Gang, Rivera-Batiz, & Yun, 2002; Gorodzeisky, 2013; Hainmueller & Hiscox, 2007).
However, despite its importance and popularity among researchers, ITT has several drawbacks.
The main weakness of the theory seems to be in its focus on the exclusively negative aspects of immigrant perception. Although studies conducted in different countries have indicated that anti-immigrant feelings are often strong in local populations, they have also demonstrated that most people in the receiving countries have some positive feelings toward immigrants, and a significant part of the local population supports continuing immigration to their country (Lee &
Fiske, 2006; Leong, 2008; Mayda, 2006; Velasco Gonzalez et al., 2008). Therefore, a compre- hensive theory explaining attitudes of local people toward immigrants should include positive aspects of perceptions of immigrants alongside their perception as a threat. Second, ITT does not delineate the antecedent psychological factors for perception of different types of threats. In the present study, we develop and test a new conceptualization, which includes appraisal of immi- grants as both threatening and beneficial (in different spheres of life) to the receiving society.
In addition, we connect appraisal of immigrants to general motivational goals of local people, as expressed in their personal value preferences, thus suggesting a psychological mechanism explaining the origin and maintenance of both positive and negative perceptions of immigrants.
Opinions regarding immigration policy. There are two separate research traditions consider- ing public opinions regarding immigration policy. Sociological research on opinions regarding immigration policy focuses on the views of the local people regarding rules of admission, natu- ralization, and allocation of rights to immigrants (Burns & Gimpel, 2000; Gorodzeisky, 2013;
Gorodzeisky & Semyonov, 2009; Levanon & Lewin-Epstein, 2010; O’Rourke & Sinnott, 2006;
Raijman, 2013; Raijman & Semyonov, 2004; Raijman, Semyonov, & Schmidt, 2003). On the contrary, most psychological studies on the issue focus on local people’s views regarding desired modes of acculturation of immigrants. For instance, acculturation theory (Berry, 1997) and Inter- active Acculturation Model (Bourhis et al., 1997) assume that dominant majority group members decide to what degree they want the immigrants to accept the majority culture and/or preserve their original culture. As such, each theoretical approach addresses some policy issues relevant for immigrants and the receiving society but ignores others. In the present study, we combine the two perspectives and suggest a new conceptualization of the opinions regarding different aspects of immigration policy. In addition, we suggest a theoretical model that explains why members of receiving society may have different views on immigration policy.
Personal value preferences. The present study is based on Schwartz’ theory of values (Schwartz, 1992, 2006; Schwartz et al., 2012) which defines values as desirable trans-situational goals that serve as guiding principles in people’s lives. In its latest formulation (Schwartz et al., 2012), the theory specifies a comprehensive set of 12 motivationally distinct values: power, achievement, hedonism, stimulation, self-direction, universalism, benevolence, humility, conformity, tradition, security, and face. In addition, the theory assumes the existence of dynamic relations between
these values: Pursuit of each value may conflict or may be congruent with the pursuit of other values. These conflicts and congruities yield an integrated structure of four higher order value types arrayed along two orthogonal dimensions: self-enhancement versus self-transcendence and openness to change versus conservation.
Researchers have assumed that attitudes toward minorities assert some values and contradict others. Specifically, researchers have argued that self-enhancement values (especially power) and conservation values (especially security and tradition) may be associated with negative attitudes toward minority groups, while self-transcendence values (especially universalism) and openness to change values (especially self-direction) may be associated with positive attitudes toward these groups (Sagiv & Schwartz, 1995; Schwartz, 2006). Several psychological mechanisms explaining the connection between values and attitudes toward immigrants have been suggested. Leong and Ward (2006) claim that local people with high preference for power, conservatism, and collectiv- ism values (which may be considered similar to the self-enhancement and conservation values of Schwartz) will hold more racist and discriminative attitudes toward immigrants due to a tendency to make greater distinction between in-groups and out-groups. The researchers further suggest that host society members with high levels of humanitarianism (which may be considered similar to Schwartz’ values of universalism and benevolence) are more sensitive to the difficulties experi- enced by disadvantaged groups. Schwartz (2007) suggests that universalism as a form of social justice and moral inclusiveness extends toward not only those from one’s in-group (as benevo- lence would) but also toward members of out-groups and, therefore, it may be the most important value predicting attitudes toward minorities.
The few empirical studies that have tested these theoretical assumptions mostly supported them. Power values have been associated with stronger prejudice and universalism and benevo- lence values with a weaker prejudices toward immigrants (Davidov, Meuleman, Billiet, &
Schmidt, 2008; Feather & McKee, 2008; Leong, 2008; Vecchione, Caprara, Schoen, Castro, &
Schwartz, 2012). We found only two studies that investigated the connection between personal values and opinions regarding immigration policy. In one study, values of self-transcendence and conservation were shown to be positively and negatively related, respectively, to willingness to allow immigrants into the country (Davidov et al., 2008). In another study, which used a theory of political values (Feldman, 1988), lower preference for the values of individualism, humanitarian- ism, and egalitarianism was found connected to attitudes related to border enforcement, reductions in the number of immigrants admitted, and lower immigrant eligibility for government services (Pantoja, 2006).
The Threat–Benefit Theoretical Model
Appraisal of an immigrant group. A central premise of the threat–benefit theoretical model is a general tendency for perception of an immigrant group by local people as both beneficial and threatening for the receiving society. In extension of ITT (Stephan & Stephan, 1996), we suggest that local people’s perception of immigrants may be conceptualized as a three-level cognitive construct. At the least differentiated level, the immigrant group is perceived in general as either threatening or beneficial for receiving society. At a middle level, local people appraise the immi- grant group as constituting a realistic and symbolic threat as well as a realistic and symbolic benefit for the receiving society. Finally, at the most differentiated level, the realistic and symbolic threats and benefits of the immigrant group are appraised in different spheres of life. Given the specific characteristics of the asylum seekers from Africa in Israel (e.g., racial differences, low level of education, a high level of religiosity, and a dangerous situation in their homeland), in the present study, we examine four types of threats and four types of benefits. These threats and benefits were conceptualized and defined through a multistage procedure including analysis of public discourse, conducting and analyzing individual interviews with experts in the field and
focus groups with students, and through a pilot study with students (see the “Method” section for more details of the procedure). The four types of threats include physical threats, economic threats, threats to social cohesion, and threats to modernity. The four types of benefits include physical benefits, economic benefits, cultural diversity benefits, and humanitarian benefits.
Following ITT (Stephan & Stephan, 1996), we assume that physical and economic threats rep- resent realistic threats, and threats to social cohesion and to modernity represent symbolic threats;
economic and physical benefits represent realistic benefits; and humanitarian and cultural diver- sity benefits represent symbolic benefits. We assume that the two abstract levels of appraisal are culturally universal, while the specific types of threats and benefits may vary across different immigrant groups and receiving societies. For instance, asylum seekers have no voting rights;
therefore, they cannot constitute a political threat. However, other immigrant groups (e.g., immi- grants from the Former Soviet Union and Ethiopia who may vote immediately upon their arrival to Israel) may be perceived as constituting a political threat to the local population. In addition, the list of benefits used in the present study is also not exhaustive, and other immigrant groups may bring additional benefits to the receiving society. For example, immigrants from North America and Western Europe to Israel may be perceived as bringing benefits of modernity.
However, because these benefits seem to be irrelevant for asylum seekers, we did not include them in the present study.
Opinions regarding immigration policy. We suggest a conceptualization of the local people’s opinions regarding immigration policy as lying along two dimensions: (a) preserving the interests and rights of the immigrants (promoting their culture and providing them with rights and socio- economic benefits) and (b) defending interests of the local people (preserving the dominant culture and immediate economic, security, and other interests of the local population). The issues of preserving the immigrants’ rights versus defending the receiving country have been discussed previously (Ajzenstadt & Shapira, 2012; Ayalon et al., 2008; Canetti-Nisim & Pedahzur, 2003;
Dhont, Cornelis, & Van Hiel, 2010; Kritzman-Amir, 2009, 2012; Lucassen, 2005); however, this is the first study that operationalizes these concepts, measures the corresponding variables, and examines their connections to personal values and appraisal of immigrants. We assume that local people who perceive an immigrant group more positively (i.e., consider them beneficial for receiving society) will tend to support immigration policy directed at defending the immi- grants’ rights. On the contrary, local people who appraise an immigrant group negatively (i.e., consider them mainly as threatening) will tend to support immigration policy directed at defend- ing the state against immigrants.
The role of personal value preferences in forming attitudes toward immigrants. Based on the analysis of the existing literature, focus groups conducted with students, and interviews with profession- als working with asylum seekers, we have formulated the possible connections between different types of threats and benefits and personal value preferences (see Table 1).
Immigrants as a threat to the receiving society. (a) Physical threat: Asylum seekers are perceived as a threat to body and/or possessions of the local people. Physical threat is assumed positively related to the conservation values (especially security) because intrusion of an alien group disturbs the existing social order and harms the sense of stability and safety among the local population (Huddy, Feldman, Taber, & Lahav, 2005); (b) Economic threat: Asylum seekers are perceived as competing for jobs and welfare services. Economic threat is related to the self-enhancement values (especially power) and to the conservation values (especially security) because asylum seekers may be perceived as threatening the local’s dominance over resources and their economic security and stability (McLaren, 2003; Schneider, 2008); (c) Threat to social cohesion: Asylum seekers are perceived as threatening societal cohesion because they introduce their own customs
and behavioral norms (Huddy & Sears, 1995). This relates to conservation values (especially tradition and conformity), which express the motivational goals of preserving local customs, religion, and social order; (d) Threat to modernity: Asylum seekers may be perceived as a threat to modernity because they come from less developed and more conservative countries, and many of them hold traditional values and nonliberal political views. As such, they may be perceived as a threat to the existing social order and to the social achievements of the receiving society (Lucassen, 2005). Therefore, threat to modernity may be related to the conservation values (especially security) and to self-enhancement values (especially achievement).
Immigrants as a benefit to the receiving society. (a) Economic benefits: Asylum seekers generally work in undesirable jobs (described in literature as the 3Ds: dirty, difficult, and dangerous; Eller- man, 2005) and for a lower salary. Individuals who hold universalism and benevolence values are in general more open toward culturally different groups such as refugees and thus may also be more likely to see their economic benefits (Leong, 2008); (b) Cultural diversity benefits:
New cultural elements (e.g., food, clothes, and music) that asylum seekers bring with them may be perceived as enriching the receiving society and promoting motivational goals related to the openness to change values (Leong, 2008); (c) Humanitarian benefits: Accepting asylum seek- ers and helping them to adjust in the receiving society promote motivational goals related to universalism—achieving equality and helping other people (Leong & Ward, 2006); (c) Physical benefits: A perception of immigrants as nice to be with, interesting, and physically attractive may satisfy one’s motional goals related to enjoying life, achieving sensual gratification, excitement, and engaging in independent thoughts and actions (Fiske, Cuddy, & Glick, 2007) and thereby to promote values of hedonism, stimulation, and self-direction.
Table 1. Definitions of the Immigrant Threats and Benefits and Their Connections to Values.
Threats and benefits Definition Values threatened/advanced
Physical threat Immigrants are perceived as physically threatening because intrusion of an alien group harms the sense of stability/security.
Security
Economic threat Immigrants are perceived as threatening the local’s dominance over resources by competing for jobs and/or competing for welfare.
Power and security
Threat to social
cohesion Immigrants are perceived as threatening societal cohesion because they introduce their own customs and behavioral norms, which are different from those of the receiving society.
Tradition and conformity
Threat to modernity Most immigrants coming to the developed countries are more conservative than the local population and thus can be seen as a threat to modernity.
Security and achievement
Economic benefit Immigrants do low-paying work that locals do not want to do and can be highly motivated to work, enabling society to achieve higher wealth.
Universalism, benevolence, self-direction
Cultural diversity
benefit New cultural elements (food, clothes, music, etc.) that immigrants bring with them may be perceived as enriching the receiving society.
Self-direction, stimulation, and hedonism
Humanitarian
benefit Accepting immigrants and helping them to adjust in the receiving society promote the motivational goals related to achieving equality, helping other people and seeing ourselves as humanitarian.
Universalism
Physical benefit Immigrants are perceived as interesting, good mannered, physically beautiful, and clean people, which are nice to be with.
Hedonism, stimulation, self- direction
Main Hypotheses of the Study
Hypothesis 1: We assume that social workers in Israel perceive asylum seekers as being both threatening and beneficial to the receiving society. We hypothesize that the social workers’
perception of the asylum seekers constitutes a three-level cognitive structure. In the most undifferentiated level of appraisal, there are two higher order factors (threats vs. benefits); in a middle level, there are four factors (realistic/symbolic threats and realistic/symbolic bene- fits); and in the most differentiated level of appraisal, there are four specific types of threats (physical, economic, social cohesion, and modernity) and four types of benefits (economic, cultural diversity, humanitarian, physical).
Hypothesis 2: A higher preference for the self-transcendence and openness to change values (especially universalism and self-direction) will be associated with a more positive appraisal of asylum seekers; that is, a higher perception of them as beneficial, and a lower perception of them as threatening for Israeli society. A higher preference for the conservation and self- enhancement values (especially tradition, conformity, security, and power) will be associated with a more negative appraisal of immigrants, that is, perceiving them as threatening.
Hypothesis 3: A tendency to perceive asylum seekers as beneficial for Israeli society will be associated with stronger support of policy defending immigrants’ rights, while perceiving them as a threat will be associated with supporting policy aimed toward defending Israeli society.
Hypothesis 4: Appraisal of asylum seekers as threatening or beneficial for Israeli society will mediate the relationship between personal value preferences and opinions on immigration pol- icy regarding asylum seekers. Our model is in line with the value-attitude-behavior hierarchy (Homer & Kahle, 1988) in which we are suggesting that opinions toward immigrant policy are attitudes-directed behavior intentions and hence may be assumed to follow appraisal (Figure 1).
Method Sample
A total of 283 social workers living across Israel and working in different social service organiza- tions participated in the study. Table 2 presents the sociodemographic characteristics of the par- ticipants comparing them with the characteristics of social workers as reported in the most recent existing survey of all social workers in Israel (Bar-Zuri, 2004). The results obtained indicate that the study sample is representative of the population of social workers in Israel regarding most of the sociodemographic characteristics: age, gender, family status, and immigration status.
However, among the study participants, there was a slightly higher proportion of social workers with a master’s in social work (MSW) or higher degree (43% as compared with 37%).2 On aver- age, the study participants had worked 10.7 years as social workers (SD = 9.6, range = 1-42).
Procedure
The study used self-report anonymous questionnaires. The questionnaires were distributed by students participating in a senior thesis seminar at the Tel Aviv University School of Social Work. The students distributed most of the questionnaires in their workplaces, which included dozens of organizations focusing on different fields of social work across Israel. In addition, the questionnaires were distributed through professional Internet forums and social networks. The goals of the study were explained, and a signed informed consent form was obtained from all the participants. Participants completed the questionnaires during their work hours or at home and returned them directly to the students in a sealed envelope or by email. The return rate was about 75%. The study was approved by the Tel Aviv University Ethical Research Board.
Instruments
Personal value preferences. The personal value preferences of the social workers were measured using the latest version of the Portrait Values Questionnaire (PVQ-R; Schwartz et al., 2012). This questionnaire consists of 57 items. Each item portrays a person’s goals, aspirations, or wishes that indicate the importance of a specific value. For each item, respondents indicate how similar the described person is to them on a 6-point scale, from 1 (not like me at all) to 6 (very much like me). Item examples are as follows: “It is important to him to obey all the laws” (conformity) and
“Being creative is important to her” (self-direction). Scores for each value are calculated as means of the relevant items. As recommended by Schwartz et al. (2012), to correct for individual differences in use of the response scales, each participant’s responses were centered on his or her own mean, that is, from each value score, the average of all 57 values was subtracted. Internal consistency of the scales was satisfactory and similar to those reported in previous studies (Cron- bach’s α = .64-.86).
Group’s appraisal. Appraisal of asylum seekers by social workers was measured by a newly cre- ated Threats–Benefits Inventory (TBI). The inventory consists of 45 items measured on a 5-point Figure 1. Theoretical model.
Table 2. Sociodemographic Characteristics of the Sample Compared With the Population of Social Workers in Israel.
Sociodemographic characteristic Study sample Population
Age 37.9 (SD = 10.3; range = 23-67) 40.2
Gender (% of females) 91 90
Family status (% of married or living with a partner) 68 66
Ethnicity (% of Jews) 100 96
Immigration status (% of foreign born) 15 15
Education (% with a master’s in social work or higher) 43 37
scale, from 1 (strongly disagree) to 5 (strongly agree). Threats are examined in four areas:
economic threats (“Asylum seekers drain welfare funds”), physical threats (“Asylum seekers commit many violent crimes against Israelis”), threats to social cohesion (“Asylum seekers are a threat to the Jewish character of Israel”), and threats to modernity (“Asylum seekers bring non- progressive rules of raising children, for example, physical punishment”). Benefits are measured in four areas: economic benefits (“Asylum seekers bring new knowledge and skills needed in the Israeli economy”), cultural diversity benefits (“Asylum seekers bring cultural diversity to our population and give us an opportunity to learn about cultures we might never learn about other- wise”), humanitarian benefits (“Accepting asylum seekers can help to save lives”), and physical benefits (“Asylum seekers are quiet and nice people”). Confirmatory Factor Analysis (CFA) was conducted to test the theoretically suggested three-level structure of TBI, and its results are presented below.
Immigration policy. Opinions regarding different aspects of immigration policy related to asylum seekers were measured by a newly created Immigration Policy Questionnaire (IPQ). The ques- tionnaire consisted of 21 items measured on a 5-point scale, from 1 (strongly disagree) to 5 (strongly agree). The items formed two scales: policies directed toward defending the particular- istic and distinct interests of Israelis and Israeli society (11 items; for example, “The Israeli government needs to build detention facilities for asylum seekers and to put all asylum seekers there”) and policies directed toward defending asylum seekers’ rights (10 items; for example,
“The Israeli state should ensure that employers pay asylum seekers the minimal salary and cover all the obligatory insurances”). CFA was conducted to test the theoretically suggested two-factor structure of IPQ, and its results are presented below.
The new questionnaires were built using a mixed emic–etic approach and a multistage proce- dure recommended by Millward (2012): (a) Items were generated from public discourse on asylum seekers in Israel. The researchers together with a team of students analyzed newspaper and Internet items to initially brainstorm themes used to describe asylum seekers. (b) Five focus groups with students and six interviews with experts in the field were conducted to discuss the initial items. In the focus groups, students were asked to brainstorm what words come to mind when they think of asylum seekers. Following this initial trigger, they were asked in what ways they thought asylum seekers impact on Israeli society and in what ways they think other people in Israeli society see that asylum seekers impact on Israeli society. The final questions were more specific: In what ways could asylum seekers be seen as a threat to Israeli society? In what ways could asylum seekers be seen as a benefit to Israeli society? Interviews with experts were along similar lines but without a trigger and focusing on how they think asylum seekers are seen in the eyes of the Israeli public. (c) “Theoretical sensitivity” (Glaser, 1978) was used by the authors with their knowledge of immigration literature and theory to conceptualize the framework of the questionnaires and to formulate the questionnaires’ items. (d) A pilot study was carried out with 172 students from two major universities in Israel to verify the structure and internal consistency of the new questionnaires. Principal components factor analysis with varimax rotation was carried out on the new questionnaires to confirm hypothesized clustering of items. For TBI, total variance explained by the threats and benefits scales was 59.1%, with most items loading higher than .4.
Threat scales were positively correlated with each other (r = .63-.66); benefit scales were posi- tively correlated (r = .49-.71), whereas threat scales and benefit scales were negatively correlated with each other (r = −.29 to −.61). For IPQ, total variance explained by the two scales was 46.6%, with most items loading higher than .6. The state defense and immigrants’ rights scales were negatively correlated (r = −.64). Eigenvalues of all scales in both questionnaires were greater than 1. Reliability of all the scales was good (Cronbach’s α for the TBI scales ranged from .72 to .85; Cronbach’s α for both IPQ scales was .87). Further details regarding the student sample are available from the authors.
Data Analysis
To evaluate the hypothesized structure of the social workers’ appraisal of asylum seekers and of their opinions regarding immigration policy, CFAs were conducted using AMOS (SPSS, 2014).
After the structure of the two theoretical constructs was verified, the composite indexes for the general threats and benefits factors, as well as for the two aspects of immigration policy (defend- ing the state and defending the immigrants’ rights), were used as oblique factors in the Structural Equation Modeling (SEM) testing the connections between values, immigrant appraisal, and immigration policy. In the initial model, values that were significantly correlated with at least one component of appraisal or policy were included as exogenous variables predicting two com- ponents of group appraisal (threats and benefits) and two components of immigration policy (defending the state and defending the immigrants’ rights). All values were permitted to correlate with each other. The two components of group appraisal were connected to the two components of immigration policy, which permitted testing of the mediation effects. Four sociodemographic variables (age, gender, education, and religiosity) were included in the model as exogenous vari- ables. The initial SEM model was tested, and modifications were made to improve the model’s goodness of fit. The covariance structure of all the models was evaluated with multiple fit indexes:
comparative fit index (CFI), the adjusted goodness of fit index (AGFI), the nonnormed fit index (NNFI; also known as the Tucker–Lewis index), and root mean square error of approximation (RMSEA). The following values were regarded as indicating a reasonable fit: CFI > .90 (Bentler, 1995), AGFI > .90 (Baumgartner & Homburg, 1996), NNFI > .95 (T. A. Brown, 2014), and RMSEA < .08 (T. A. Brown, 2014; Hooper, Coughlan, & Mullen, 2008).
Results
To evaluate the internal structure of the TBI, we conducted CFA for the initial three-level theo- retical model. In this model, the 45 TBI items measured eight first-order factors representing four threats and four benefits. In their turn, the eight first-order factors represented four second-order factors reflecting realistic and symbolic threats and realistic and symbolic benefits. Finally, the four second-order factors represented two third-order factors reflecting threats and benefits in general (Figure 2). The goodness of fit indexes of this model indicated that the data poorly fitted the theoretical model (Table 3, Model A). Examination of the modification indexes led to dropping four items (all measuring economic benefits); however, the revised model’s fit was still not good (Table 3, Model B). Because TBI includes a large number of items and our sample was only moderate in size, this resulted in a low subjects-to-variables ratio (1:6.9), which might reduce the model’s goodness of fit indexes. To solve this problem, we used a procedure of item parceling (T. A. Brown, 2014; Little, Cunningham, Shahar, & Widaman, 2002). At the first stage of parceling, to confirm that each TBI item loaded on its theoretically assigned factor, we conducted CFA for each of the four types of threats and four types of benefits separately, where each type of threat and benefit was presented as a latent variable, and each questionnaire’s item loaded on the corre- sponding latent variables. The results of all the analyses indicated a good fit; in addition, loadings of most of the items on their corresponding factors were >.40 (Appendix A).3 After confirming that the theoretically assigned items indeed belong to the corresponding factors, we calculated means of the items measuring the four threats and four benefits and used them as observed vari- ables for the eight first-order factors in our model (Figure 2). The model demonstrated good fit (Table 3, Model C) and, therefore, confirmed the theoretically hypothesized three-level structure of the local people’s appraisal of immigrants.4
Next, we used CFA to test the theoretically suggested two-factor structure of the IPQ.
Our theoretical model included two oblique correlated factors, with each of the 21 IPQ items loaded on its corresponding factor. The goodness of fit indexes of this model indicated a poor fit;
therefore, we examined the modification indexes. This led to dropping the two items leaving 19 items in the model. The modified model yielded a good fit (Table 4, Model A), and loadings of all items in the model were >.40 (Appendix B). Because the two oblique factors were highly correlated (estimate = −.744), we evaluated a one-factor model, in which the 19 IPQ items were loaded on a single oblique factor. Goodness of fit indexes of the one-factor model indicated a poor fit (Table 4, Model B). Thus, the theoretically suggested two-factor structure of the local people’s opinions regarding immigration policy was confirmed.
At the third stage of analysis, we used SEM to test the suggested theoretical model of the connections between values, appraisal of immigrants, and opinions regarding immigration policy (Figure 1). The composite indexes for the higher order threats and benefits factors, as well as for the two aspects of immigration policy (defending the state and defending the immigrants’ rights) were calculated during the CFA described above, and they were used in SEM testing the connections Figure 2. The threat–benefit model of the immigrants’ appraisal: Path diagram with estimated
standardized coefficients.
Note. All coefficients have a level of significance p < .001.
Table 3. Confirmatory Factor Analyses of the Theoretical Model of Appraisal of Immigrants: Goodness of Fit Indexes.
Models χ2 df χ2/df p CFI AGFI NNFI RMSEA (90% CI)
A. The initial theoretical
model (45 items) 1,841.693 929 1.982 .000 .821 .744 .810 .059 [.055, .063]
B. A modified model
(41 items) 1,492.623 772 1.933 .000 .846 .772 .837 .058 [053, .062]
C. A modified model
with eight parcels 27.774 11 2.525 .004 .984 .925 .959 .074 [.040, .108]
Note. CFI = comparative fit index; AGFI = adjusted goodness of fit index; NNFI = nonnormed fit index; RMSEA = root mean square error of approximation; CI = confidence interval.
between the variables in the study. In the initial model, values predicted the two aspects of the local’s appraisal of asylum seekers (threats and benefits), which in turn predicted opinions regarding the two aspects of immigration policy (defending the state and defending the immi- grants’ rights). In addition, direct connections between all values and the two policy aspects were included in the model to test for mediation effects. All values were permitted to correlate with each other. Finally, four sociodemographic variables (age, gender, education, and religiosity) were included in the model as predicting the values, as well as the appraisal and policy variables.
Table 5 presents Pearson correlation coefficients between all the variables in the model.
The initial model demonstrated poor fit, and modifications were made to improve the model’s goodness of fit. The resulted best-fit model (Figure 3) included four values (power, social security, universalism, and tradition), which predicted the appraisal of asylum seekers as both threatening and beneficial, which in turn predicted attitudes toward immigration policy as directed at defending the state and defending the rights of the asylum seekers. In addition, social security values directly predicted immigration policy of defending the state, while the direct predictions of all other values of the two components of immigration policy were not significant. Finally, among the four sociodemographic variables, only age and religiosity were significantly related to other variables in the study. The resulting model demonstrated excellent fit (χ2 = 10.31; df = 10;
χ2/df = 1.031, p = .414; CFI = 1.000; AGFI = .960; NNFI = .999; RMSEA [CI] = .010 [.000, .066]).5 The model explained a substantial proportion of variance in appraisal of asylum seekers as a threat (R2 = .43) and benefit (R2 = .39) as well as in immigration policy directed at defending the state (R2 = .67) and defending the asylum seekers rights (R2 = .55).
To examine the mediating role of the appraisal of immigrants as threatening and beneficial for the local population, we decomposed the total effects of the two sociodemographic variables (age and religiosity) and the four values (power, social security, universalism, and tradition) on the opinions regarding immigration policy into indirect effects (Table 6) and direct effects (Table 7). The results indicate that indirect effects of all the predicting variables on the opinions regarding immigration policy were significant. In addition, a direct effect of the social security values on the aspect of immigration policy directed at defending the state was significant.
Therefore, perceived threats and benefits fully mediated the effect of the universalism, tradition, and power values on the opinions regarding immigration policy. In addition, the effect of the social security values on the aspect of immigration policy directed at defending the asylum seekers’ rights was also fully mediated by the perceived threats and benefits. However, the effect of the social security values on the aspect of immigration policy directed at the defense of the state was only partly mediated by perceiving asylum seekers as threatening the receiving society. Values partly mediated the effects of age and religiosity on the immigrant appraisal.
At the last stage of analysis, we compared the appraisal of asylum seekers as threatening versus their appraisal as beneficial for the receiving society. In addition, we compared the opinions reflecting immigration policy directed at defending the state with those reflecting immigration Table 4. Confirmatory Factor Analyses of the Theoretical Model of Opinions Regarding Immigration Policy: Goodness of Fit Indexes.
Models χ2 df χ2/df p CFI AGFI NNFI RMSEA (90% CI)
A. The two-factor theoretical
model (19 items) 304.506 147 2.071 .000 .949 .968 .940 .062 [.052, .071]
B. A one-factor model
(19 items) 580.262 148 3.921 .000 .859 .709 .837 .102 [.093, .111]
Note. CFI = comparative fit index; AGFI = adjusted goodness of fit index; NNFI = nonnormed fit index; RMSEA = root mean square error of approximation; CI = confidence interval.
Table 5. Pearson Correlation Coefficients. AGERSDSTHEACPOFASPSSTRCOHUBEUNASST 1234567891011121314151617181920 11 2.051 3.38**.081 4.08−.02.031 5.05.06.24**−.071 6−.12*−.14*.06.00.14*1 7−.24**−.03−.16**−.23**−.21**.13*1 8−.24**−.08.04−.04.03.07.18**1 9−.05−.16**−.02−.03−.30**−.09.09.19**1 10−.12−.02−.04.06−.20**−.07−.01.20**.081 11−.03.07−.02−.03−.21**−.23**.07−.03−.12*.17**1 12.20**−.07−.07.18**−.23**−.19**−.03−.15**−.07.15*.14*1 13.09−.07−.09.44**−.26**−.03−.19**−.15*.06−.05−.08.23**1 14.07.04−.15*.08−.36**−.27**−.14*−.20**−.20**.05.07.09−.011 15.08.01−.05−.01−.01−.12*−.27**−.31**−.36**−.15*−.08−.06−.05.19**1 16.05.11.06−.10.30**−.19**−.15*−.10−.19**−.28**−.03−.10−.18**−.23**.051 17.09.19**.15*−.16**.35**.06−.15**−.27**−.55**−.36**−.13*−.30**−.29**−.17**.24**.12*1 18−.06.10.14*−.19**.20**.07.00−.08−.27**−.12*−.06−.33**−.29**−.07.12*.14*.50**1 19.14*−.17**−.10.24**−.22**−.06−.01.08.24**.21**.09.41**.31**.08−.20**−.10−.55**−.68**1 20−.02−.16**−.16**.27**−.24**−.01.05.13*.21**.31**.11.35**.30**.12*−.19**−.21**−.57**−.58**.77**1 21−.14*.08.08−.14*.22**.12.02.06−.24**−.08−.09−.34**−.31**−.05.03.06.45**.65**−.58***−.46** Note. A = age; G = gender (1-male; 2-female); E = education; R = religiosity; SD = self-direction; ST = stimulation; HE = hedonism; AC = achievement; PO = power; FA = face; SP = security personal; SS = security social; TR = tradition; CO = conformity; HU = humility; BE = benevolence; UN = universalism; AS = defending the rights of asylum seekers; S = defending the state; T = threats; B = benefits. *p < .05. **p < .01.***p < .001
policy directed at the defending the asylum seekers’ rights. We calculated two paired-samples t tests using the composite indexes for the higher order threats and benefits factors, as well as for the two aspects of immigration policy (defending the state and defending the immigrants’ rights).
The t tests indicated that social workers perceive asylum seekers as more beneficial than threat- ening for Israel: M(SD)benefits = 2.82(.55); M(SD)threats = 2.24(.61); t(282) = 9.86, p = .000.
In addition, the t tests indicated that social workers supported more strongly immigration policy directed at defending the asylum seekers’ rights than the policy directed at defending the country:
M(SD)AS = 3.47(.86); M(SD)state = 2.46(.80); t(282) = 11.2, p = .000.
Discussion
In this study, we formulated and empirically tested a new model explaining local people’s atti- tudes toward immigrants. The results have confirmed that (a) local people appraise immigrants as both threatening and beneficial for receiving society, (b) threats and benefits are perceived either as realistic or as symbolic, (c) immigrant perception as threatening or beneficial for receiving society influences opinions regarding immigration policy, (d) personal value preferences affect individuals’ appraisal of immigrants as representing a threat or benefitting the host society, and (5) the threat–benefit appraisal mediates the relationship between personal values and the indi- viduals’ opinions on immigration policy.
A major contribution of the present study is that it demonstrated that immigrants are perceived not only as a threat but also as a benefit for the receiving society. Moreover, the results obtained indicated that social workers appraised asylum seekers as more beneficial than threatening, and they were more strongly oriented to defending the asylum seekers’ rights than defending the receiving society. This clearly extends previous literature on attitudes toward immigrants that has tended to focus on negative aspects of appraisal, in particular threat and competition (Croucher et al., 2013; Dustmann & Preston, 2007; Fitzgerald, Curtis, & Corliss, 2012; Stephan et al., 2000;
Stephan et al., 2005; Stephan & Stephan, 1996; Velasco Gonzalez et al., 2008).
Figure 3. The best-fit structural equation model including sociodemographic variables, values, appraisal, and immigration policy.
Note. χ2 = 10.31; df = 107; χ2/df = 1.031, p = .414; CFI = 1.000; AGFI = .960; NNFI = .999; RMSEA (CI) = .010 [.000, .066]. CFI = comparative fit index; AGFI = adjusted goodness of fit index; NNFI = nonnormed fit index;
CI = confidence interval.
Table 6. Standardized Indirect Effects: Estimates, Bootstrap Standard Errors, and the Level of Significance. VariablesReligiosityAgeUNTRSESPO S.065 (.026); p = .006.016 (.010); p = .013−.341 (.041); p = .002.084 (.036); p = .012.125 (.036); p = .003.021 (.011); AS−.027 (.015); p = .006−.037 (.018); p = .020.281 (.044); p = .004−.100 (.032); p = .004−.120 (.036); p = .003−.049 (.020); Note. UN = universalism; TR = tradition; SES = security social; PO = power; S = defending the state; AS = defending asylum seekers.
Table 7. Standardized Direct Effects: Estimates, Bootstrap Standard Errors, and the Level of Significance. VariablesReligiosityAgeUNTRSESPOT T.122 (.045); p = .009−.504 (.045); p = .004.095 (.054); p = .087.168 (.051); p = .004 B−.085 (.038); p = .025.385 (.060); p = .004−.179 (.051); p = .004−.188 (.056); p = .003−.111 (.042); p = .011 S−.067 (.054); p = .209.019 (.034); p = .558.117 (.045); p = .006.067 (.041); p = .071.534 (.067); p = .002−.187 (.067); AS.076 (.064); p = .238−.023 (.042); p = .637−.075 (.048); p = .142−.050 (.056); p = .350−.223 (.087); p = .007.438 (.082); Note. Empty cells represent no direct effect. UN = universalism; TR = tradition; SES = security social; PO = power; T = threats; B = benefits; S = defending the state; AS = defending asylum seekers.
The present study provides some answers to the question of why it is that various people in a receiving society perceive the same immigrant group so differently. Why for some people they are experienced as a threat, whereas for others their presence signals positive contributions and benefits to the society. We suggest that the existence of an out-group and its definition as a threat or a benefit plays an essential role in enabling the fulfillment and attainment of personal motivational goals (as expressed in personal value preferences). Threat/benefit appraisal of an immigrant group as “the other” enables the local people to put into practice their values and to experience consistency and concordance between values and attitudes (Haddock & Maio, 2007;
Hitlin, 2003, 2007). For example, for those who highly value power, a need to control people is justified through a perception of an out-group as threatening. On the contrary, people for whom the motivational goals related to understanding, appreciation, tolerance, and protection of the welfare of all people are important (as reflected in their high preference for the universalism values) may appraise immigrants more positively, because for them experiencing the out-group as a threat would arouse dissonance with their high preference for universalism values.
The present study sheds light on the possible psychological mechanism explaining the mediating role of immigrants’ appraisal in the connection between value preferences and opinions regarding immigration policy. Our findings indicate that those members of receiving society for whom it is important to care for others and to help them, tend to appraise immigrants as more beneficial and less threatening because this perception justifies their support for immigration policy directed at defending the immigrants’ rights, which in turn helps them to attain important motivational goals expressed in their preference for universalism values (Davidov et al., 2008;
Pantoja, 2006; Schwartz, 2007). On the contrary, people who highly value social security, power, and tradition and, therefore, for whom preservation of status quo and controlling others is impor- tant may tend to perceive immigrants as less beneficial and more threatening to receiving society because this perception justifies their support for immigration policy directed at defending the receiving society against immigrants.
The results of our study shed some new light on the psychological understanding of in- and out-group relationships in general. Traditional social psychology has considered in-group relations as crucial for the individuals’ well-being and attaining important goals in life, and as intrinsically positive in the experience of the individual. On the contrary, the out-group is consid- ered as largely threatening in the individuals’ experience, and the relations with out-groups are considered as colored by competition and hostility (R. Brown & Hewstone, 2005; Sherif, 1967;
Tajfel & Turner, 1986). The results of our study demonstrate the out-group is no less crucial than the in-group for attainment of personal motivational goals. Our findings indicate that a negative appraisal of the out-group (asylum seekers in the present case) may be necessary for attaining the motivational goals expressed in the values of power, security, and tradition. On the contrary, a positive appraisal of the out-group is needed for attaining the motivational goals expressed in the values of universalism. Attainment of the motivational goals expressed in these values is impossible without the existence of “the other” because only the existence of positively and/or negatively appraised other can allow their actualization.
The high negative correlation between the perceived threats and benefits obtained in the present study has indicated that social workers tend to perceive asylum seekers in a nondifferentiated manner, considering them as either a threat or benefit in general as opposed to a fine perception of them as being a threat in some spheres of life and a benefit in other. We suggest that this indis- criminate perception may be due to the nature of asylum seekers from Africa in Israel as a racially different and culturally distant group, with which the majority population (including social workers) have few personal contacts (Ajzenstadt & Shapira, 2012). A more differentiated appraisal may be anticipated in research examining attitudes toward other immigrant groups, with whom contacts are more frequent and intimate; these groups may be perceived as simultaneously threatening and beneficial in different spheres of life. Low levels of contact between social workers and asylum
seekers may also explain the fact that the personal security values did not affect either appraisal or policy opinions. We would expect that this relationship may be different with other immigrant groups with whom the host population has greater levels of contact.
Limitations of the Research
The present study has several limitations. It examines attitudes toward a particular group of immigrants in a particular cultural and social context; therefore, further research is needed to examine the proposed model in additional cultural environments. The research involves cross- sectional data, and while the predictive nature of relationships between study variables rests on a theoretical and conceptual base, longitudinal research is needed for establishing causality. The study focuses on social workers who, although critical in implementing policy and responsible for distribution of state welfare resources (Ayalon et al., 2008), represent only a part of Israeli society. In the present study, social workers saw asylum seekers as more beneficial than threaten- ing for Israeli society and stronger supported defending the asylum seekers rights than defending the receiving society. However, this pattern of appraisal and opinions regarding immigration policy toward asylum seekers may reflect professional self-selection, socialization, and confor- mity to the professional values (Israeli Association of Social Workers [IASW], 1994; National Association of Social Workers [NASW], 2000).
Social workers’ values have been found to be different from those in the general population, in that they have a relatively high preference for the benevolence, universalism, and self- direction values, and a relatively low preference for the hedonism, achievement, power, and security values (Knafo & Sagiv, 2004). In addition, because of the norms of professional ethics (Dekel & Peled, 2000), social workers may have a more positive attitudes toward immigrants.
This bias probably leads to narrowing the variance in the value preferences and attitudes among social workers compared with the general population. This factor might decrease the size of connections between values, immigrants’ appraisal, and policy opinions. The fact that we found strong connections in the study’s sample provides a strong support for the model, as these connections may be even stronger in a general population, where the variance in the values (and probably also in attitudes) are greater.
The fact that the vast majority of the study sample was female (representative of the Israeli community of social workers) also constitutes a limitation. While gender was not a significant predictor in the final model, correlations showed male social workers tended to perceive asylum seekers as more threatening and supported immigration policy defending the state more strongly than female social workers did. However, previous studies show that the correlations between gender and values (Schwartz & Rubel, 2005) and gender and attitudes (Vecchione et al., 2012) are low. Therefore, it is improbable that the gender bias of the social workers would affect the results of the model testing.
The study examines attitudes toward asylum seekers (as opposed to other immigrant groups such as foreign workers or diaspora immigrants who are often part of greater public consensus due to the benefits they provide to the society or the similarity they hold to the local population).
Recent literature suggests that when there are widely accepted social norms, the connections between personal value preferences and attitudes are weaker (Davidov, Meulemann, Schwartz,
& Schmidt, 2014; Kuntz, Davidov, Schwartz, & Schmidt, 2015; Schwartz & Butenko, 2014). As such, we believed that it was better to test our model investigating the attitudes toward asylum seekers than toward other immigrants groups. However, further research involving a nationally representative sample, examining additional immigrant groups and involving a more equal gender split will explore the robustness of the proposed model. Finally, external validity of the new measures should be assessed in further research.
Implications
Public opinion around the world has often emphasized the negative aspects of migration, such as the contribution of immigrants to crime and social and economic deterioration (Fitzgerald et al., 2012; Higgins, Gabbidon, & Martin, 2010; Simon & Sikich, 2007). These negative atti- tudes can heighten intergroup tension and conflict creating an ever deepening vicious cycle of lack of trust and hostility. Results of the present study suggest that encouraging perceptions of immigrants as benefitting the society (economically, morally, culturally, and physically) through the media and the educational system can promote greater acceptance of and understanding toward newcomers. They also show that some benefits can be found even for groups considered culturally distant and mainly disliked or rejected by the host population. In addition, results point to the importance of value-based interventions and education (Bardi & Goodwin, 2011;
Biesta, 2010) as a means to change appraisal and societal attitudes toward immigration policy.
Findings show that values of universalism are related to greater levels of acceptance and positive appraisal of asylum seekers. Educating toward such values (as well as awareness of the need for caution around values of security, power, and tradition) could lead to greater levels of acceptance not only of asylum seekers but also of “the other” in general, which is critical in today’s reality of global multicultural contacts.
More widely, we believe that while the new threat–benefit model has been developed in the area of immigration, it may have meaning and implications concerning additional minority and stigmatized groups in society. As such, the model may well prove relevant for understanding how individuals appraise and develop opinions toward policy concerning the gay community (e.g., legalization of gay marriage and equal legal gay rights) or toward people with physical or emotional challenges (e.g., policy surrounding positive affirmation in the workplace and accessibility for individuals with physical challenges). The model suggests that personal value preferences may predict opinions toward various minority and stigmatized groups (Kuntz et al., 2015) through the mediation of positive and negative appraisal. Therefore, education around values of acceptance and potential benefits of the groups for the larger society may be able to influence opinions supporting greater tolerance toward different minority groups in society.
Appendix A
Confirmatory Factor Analyses of the Four Types of Threats and Four Types of Benefits: Goodness of Fit Indexes and Standardized Regression Weights (Loadings).
Items and goodness of fit indexes Loadings
Economic threats: χ2 = 24.367; df = 13; χ2/df = 1.874; p = .028; CFI = .976; AGFI = .950; NNFI = .962;
RMSEA (CI) = .056 [.018, .089]
Asylum seekers send money home and thus damage the Israeli economy .645
Asylum seekers take work from locals and thus hurt them economically .705 Children of asylum seekers overload Israeli schools and decrease the quality of education for
Israeli children
.598
Asylum seekers raise apartment prices in Israel .536
Asylum seekers drain our welfare funds .460
Use of the cheap labor force of asylum seekers harms economic development of Israel .575 Asylum seekers overload social services who then have less resources to help Israelis in difficulty .668 Physical threats: χ2 = 25.650; df = 19; χ2/df = 1.350; p = .140; CFI = .991; AGFI = .961; NNFI = .987;
RMSEA (CI) = .035 [.000, .067]
Asylum seekers are involved in a lot of drinking and drugs .625
Asylum seekers spread infectious diseases .669
Asylum seekers help enemies of Israel .646
(continued)
