¹ Autonomous University of Nuevo Leon (Mexico).
² University of Salamanca (Spain).

Correspondence

ABSTRACT

This study analyzes the cognitive representations of ethnic minorities in a sample of Spanish undergraduate students. As a general hypothesis it was predicted that perceived differences in social status shape these representations and expectations for contact. In consecutive group interviews, participants were quizzed about their knowledge, experiences and expectations associated with social interactions with people belonging to ethnic minority groups. The information obtained from the participants was analyzed by using a mixture of quantitative and qualitative techniques in order to discover underlying dimensions in their responses. The results confirmed our prediction by displaying different associations between different types of contact (e.g., positive, negative) and the representations of minorities according to their position in the perceived social hierarchy. Findings are discussed in terms of their potential implications for contact interventions.

Key words: Intergroup contact; cognitive schemas; ethnic minorities; prejudice; social identity.

RESUMEN

El presente estudio hace un análisis de las representaciones cognitivas sobre minorías étnicas en una muestra de jóvenes universitarios españoles. Como hipótesis general se señala que las diferencias de estatus social percibidas ejercen un papel relevante en tales representaciones y en las expectativas de contacto. En varias entrevistas grupales los participantes fueron cuestionados sobre sus experiencias, conocimientos y expectativas asociadas a las interacciones con miembros de grupos étnicos minoritarios. La información recabada fue analizada a partir de un cruce de técnicas cualitativas y cuantitativas para verificar las dimensiones subyacentes en sus respuestas. Los resultados confirmaron esta hipótesis mostrando distintas nociones de contacto (ej., positivo, negativo) asociadas a las representaciones de las minorías según su posición en la escala social percibida. Los hallazgos son discutidos en cuanto a su implicación para futuras intervenciones de contacto intergrupal.

Palabras clave: Contacto intergrupal; esquemas cognitivos; minorías étnicas; prejuicio; identidad social.

Introduction

Contemporary studies in the intergroup contact paradigm (Allport, 1954) have focused on the cognitive structures that influence communication dynamics between members of different groups (Cameron, Rutland, Turner et al. 2011; Lin, Zhang & Harwood, 2004). The way in which outgroup members are represented can determine, sometimes to a large extent, the success or failure of these meetings. This is due to the fact that interactions are made on the basis of preexistent information deeply anchored in people's schemas of thought (Fiske & Taylor, 1991; Harwood, McKee & Lin, 2000).

The mental scaffolding underlying intergroup interactions seems more relevant in contexts of ethnic majorities and minorities because it is expected that a frequent and normalized contact with outgroups would contribute to eventually improving the relations between groups (Lee, 2001). However, despite the assumption that physical proximity in these contexts would enhance contact opportunities somehow, there is an established tendency among groups to avoid this type of encounter; specifically with members from highly stigmatized groups or with those considered to be socially inferior (Martinez, 2000).

In multiethnic societies, a propensity to rank ethnic groups with the majority group at the top and the rest of the minorities in positions placed closer or further away according to the largest ingroup criteria, has been shown to exist in diverse countries and settings (e.g., Hagendoorn, Drogendijk, Tumanov & Hraba, 1998; Hraba, Hagendoorn & Hagendoorn, 1989; Snellman & Ekehammar, 2005). Different aspects such as ethnocentrism, stereotyping, and perceptions of threat play an important role in the way this ranking is built, and mirror the interrelated social psychological and structural processes that shape group identities (Emessik & Mackie, 1989; Hraba et al., 1989; Stephan & Stephan, 1985). Members of high-status groups have access to antagonistic, normative, and ideologic intergroup representations that maintain and reinforce their privileged position by means of conventional processes of symbolic influence (Lorenzi-Cioldi & Clémence, 2001; Staerklé, Clémence & Spini, 2011). The fear of losing status would lead these members to form relationships only with those outgroup members who are perceived to be more socially desirable, similar, or less vilified (Osbeck & Moghaddam, 1997).

In contemporary research, there has been recurrent interest in finding out how perceptions of asymmetry in status, power, and access to resources may affect members' dispositions for contact (Gomez-Berrocal & Navas, 2000; Tropp & Pettigrew, 2005). Since Allport's conditions included equal status for group members in order to guarantee optimal outcomes, this structural imbalance could diminish or nullify the potential benefits of the interaction. As mentioned by Martinez (2000): "It is likely that, when group relations are scarce, status differences are noticeable, and the perceptions about the outgroup are stereotyped, a hypothetical anticipation of contact might bring about anxiety" (p. 35). Therefore, it is expected that an analysis of how the cognitive representations of outgroups are configured in a context of unequal structures would contribute to a better understanding of the underlying beliefs about contact among members of the ingroup, and based on such information, more accurate predictions for contact effects in future interventions could be formulated.

The purpose of this study is to examine how the schematic representations of ethnic minorities are organized in members of the majority ingroup. In particular, the potential influence of perceived status differences among younger Spaniards on contact with ethnic minority groups will be analyzed. Over the last twenty years immigration flows in Spain have led to a gradual increase in ethnic communities with higher or lower levels of integration into the host society and, as with other multicultural countries, various degrees of closeness or distance between them and the autochthonous population have emerged. Then, an exploration of these representations may offer useful information about the structure of beliefs that underlie and justify the type of approaches (contact) that ingroup members hold towards ethnic minorities.

The Schema Concept

Schemas are the cognitive structures in which the information a person has about him/herself and other people, events, objects or stimuli, is stored and organized. They comprise mental representations derived from personal experiences or from accessible, socially-shared knowledge that shape perceptions when new social information is received. In this way, they lead to a fast and effective processing of new input (Fiske & Taylor, 1991; Markus, 1977).

In an intergroup relation setting a group schema embodies the structure of beliefs a person has about his/her group of reference and about other social groups. When someone sees him/herself as a member of a specific group (e.g., young people, students, men, women, etcetera), the way in which social facts are addressed and interpreted will be closely linked to the norms or directives of this ingroup (Garcia-Leiva, 2005). A key example of this kind of beliefs are group stereotypes, which are formed by cognitive representations of people that are based on traits or attributes drawn from their group membership (Macrae, Stangor & Hewstone, 1996). These representations are also built from contextual factors (e.g., shared family beliefs), as well as from socializing with other people (Ashmore & Delboca, 1981). The configuration of this type of schema facilitates connections in the memory to process new data and avoids unnecessary cognitive efforts. However, a direct consequence of these previously-outlined routes is that people tend to seek confirmation or reinforcement of stereotypic information every time they think of outgroups (Stangor & Schaller, 1996).

Along with the aforementioned characteristics, schemas also include affective information and behavior protocols for dealing with particular people in specific situations. As a result of these protocols, all those details considered as irrelevant for the interaction are dismissed (Harwood et al., 2000; Stangor & Schaller, 1996). The role of schemas in hypothetical or real situations of intergroup contact has been registered in previous studies (Goodman & Gareis, 1993; Scherer & Petrick, 2010). Harwood et al., (2000), for example, analyzed the schematic representations of intergenerational communication in younger and older adults. The researchers found that these representations contained different positive and negative perceptions, as well as utility and rejection regarding intergenerational conversations. They also marked the potential influence of these schemas on expectations about future encounters between both groups too. Later, Lin, Harwood & Hummert (2008) replicated some of these results but reported an additional significant relationship between the outgroup stereotypes (older adults), the intergenerational communication schemas (e.g., positive, neutral, negative), and the satisfaction resulting from the conversation: that is to say, older adults' negative stereotypes (e.g., bitter) were positively related to negative communication schemas (e.g., hostility), and to lower levels of satisfaction with the meeting.

The Context of Ethnic Majorities and Minorities in Spain

In general, an ethnic minority can be described as a set of people that are perceived by others -and by themselves-, as distinct from the rest of the population in terms of identity and cultural attributes such as language, religion, dress code, and/or genetic peculiarities (Garcia, Adroher & Blanco, 1996). Ethnic minority groups have settled and expanded in a relatively short period of time, as Spain became an immigration-receiving country as of the mid-1990s.¹ In March 2013, the foreign population legally living in Spain was 5,467,955 inhabitants (11% of the overall population)² of which, 40.6% were EU citizens, 26.9% were from Latin American countries, 21.2% from different regions in Africa, and 6.9% from Asia. On the other hand, although the Gypsy population in Spain is not known accurately, it is estimated to constitute between 9.1 and 18.2% of the overall Spanish population.³

Although Spain has traditionally been depicted as tolerant of minorities in comparison with other European countries (e.g., Eurobarometer 296), attitudes toward immigration have experienced substantial changes since the 1990s. For instance, the proportion of people that considered the immigrant population in Spain as "excessive" was only 29% in 1996, whereas this percentage increased to 46% in 2011 (www.cis.es). In this regard, there is a strong tendency to link ethnic minority groups to the scarce economic resources they possess: this is, the distinctive characteristics of the ethnic minorities (e.g., language, culture, and phenotype) are often associated to socioeconomic conditions less favorable than those of the average autochthonous population. Furthermore, the presence of non-EU citizens in Spain, continually amplified by the media, nurtures a general perception of "illegality" regarding members of these minorities. As suggested in previous intergroup approaches such as the realistic threat (Stephan, Diaz-Loving & Duran, 2000), the competence for scarce resources has boosted more negative attitudes, reinforced stereotyped beliefs and justified rejection towards outgroups in the population; even among highly educated people, leftists, or people with certain experience living abroad (Cea D'Ancona & Valles, 2009).

In this scenario, the contribution of the news media to the reinforcement of this type of discourse has been pointed out in previous studies. Igartua et al. (2008), for example, reported on the sociocognitive effects of consuming news items that link negative events to immigrants: people exposed to such news in experiments showed a greater tendency to evaluate immigration and ethnic minorities as social problems, and to express negative opinions and beliefs about the consequences of immigration for the country than those in the control condition. These opinions and beliefs also contribute to increase expressions of prejudice, mostly in subtle ways, among the Spanish society as has been previously reported by scholars (Gomez-Berrocal & Moya, 1999; Navas, 1998; Rueda & Navas, 1996). Subtle prejudice emerges when people feel their cultural homogeneity is threatened by the habits, values and traditions of the foreign population (Pettigrew y Meertens, 1995). Thus, the cultural differences between the autochthonous and the foreign populations are magnified and there are constant calls to stick to core Spanish traditions and values. Because these reasons are not directly related to the genetic or physical aspects of outgroupers, there is a high degree of consensus about them among ingroupers. In an everyday situation, this is can be seen in members of the largest majority avoiding direct contact with ethnic minorities; making comments under their breath or whispering, revealing their rejection by staring or displaying patronizing behavior (Cea D'Ancona & Valles, 2011).

As is the case in other multiethnic societies, comparable studies noted that certain ethnic minorities are preferred over others. Those perceived by members in the largest ingroup as more similar to them in terms of culture, language, religion, phenotype, and/or socioeconomic status, are often appreciated more. According to survey reports presented by the Instituto de Estudios Sociales Avanzados de Andalucía (IESA) and the Observatorio Español del Racismo y la Xenofobia (Oberaxe), the communities from Latin America, Asia, and Southern Africa are tolerated more than groups commonly associated to the Islamic religion (Maghrebis, Egyptians), people from Eastern Europe (Romanians, Bulgarians), and those belonging to the Gypsy community (Cea D'Ancona & Valles, 2009; Perez-Yruela & Desrues, 2007). Although Gypsies are not considered "immigrants" because they constitute one of the oldest ethnic minorities in Europe, they have been systematically relegated to the lowest positions in the Spanish social hierarchy due to a long and complex history of mutual animosity (Gamella, 1996).

Survey respondents in these studies labeled people coming from developing countries or from disadvantaged regions "immigrants", whereas the term "foreigners" was most used for people coming from developed countries without a permanent residence in Spain, but with money enough to spend in the country. In this sense, foreigners are better regarded than immigrants.

Method

Participants

In order to draw the latent representations of ethnic minorities from ingroupers, four sessions of focus groups were carried out with a total of 26 participants (69% female, average age = 21.38 years [SD = 1.92]). The sample used in this study involved undergraduate students at a large university in central Spain. Each session comprised six participants, except for one session that comprised eight. All the participants received a small monetary incentive. University students seem a good option for keeping a certain balance between homogeneity (e.g., age, education, familiarity with the social environment) and heterogeneity, by including students from different backgrounds (e.g., Psychology, Chemistry, Sociology, and Communication), and from different regions in Spain such as Galicia, Catalonia, Andalusia and the Basque Country (see Morgan, 1997).

Procedure

On arrival, participants were asked to fill out an intergroup thermometer measure (Esses, Haddock, & Zanna, 1993), reporting on a scale from 0 to 100 the extent to which they felt social and emotionally closer to ten ethnic groups with higher presence in Spain. Regions were emphasized rather than individual countries because this would stimulate the potential connections present in participants when speaking about the nationalities of each ethnic group (e.g., the nationalities that come into mind when speaking about Eastern Europeans). The groups included were: Northern Europeans, Southern Europeans, Eastern Europeans, Central Europeans, Gypsies, Southern Saharans, Northern Americans, Maghrebis, Latin Americans, and Asians. Additionally, the scale included an explanation about the conceptual difference between immigrants and ethnic minorities: the former being newcomers with a low level of integration into the host society and the latter being communities of second-generation immigrants (born and raised in the country), or people who have lived in the country for many years and display a higher level of social integration.

Each session was moderated by two pre-trained assistants and digitally recorded with an average time of 90 minutes per session.

Interview Script

For this study, a semi-structured interview script that included aspects related to perceptions, beliefs, experiences and expectations for contact with ethnic minorities was introduced. In this view, the script was divided into four thematic domains: 1) Knowledge about ethnic minorities; 2) Experiences of contact with ethnic minorities; 3) Perceptions of distance, and 4) Typical attributes of the group category.

Data Analysis

In order to analyze the underlying dimensions in the participants' responses, an adaptation of the protocol in Paez, Valdosedal, Igartua, Basabel and Iraurguij (1992) was applied. This consisted of transforming the qualitative data into quantitative units for further multivariate statistical analyses. This cross-technique has been used in previous studies in order to obtain the central ideas within the information gathered and the relationships existing between these ideas. The protocol includes the following steps: 1) transcribing all the group interviews, 2) listing the ideas expressed during the sessions, 3) creating a system of categories, 4) recoding of all the ideas using this system, and 5) analyzing and interpreting the information with descriptive and multivariate tests.

Results

In total, the participants articulated 738 ideas in all the group interviews. Two researchers who were unaware of the aims of this study randomly coded 15% of the ideas to carry out a test for reliability. The agreement reached in the four domains was 81.75% (Scott's Pi = .79), therefore, the intercoder reliability in this study was acceptable (see Table 1).

The codes that were created for each of the four domains formed the category system for this study. These codes were designed according to semantic similarities in the ideas that were gathered by using a "bunch" technique (see Igartua, 2006). The first domain, for instance, comprised six codes related to diverse types of knowledge that participants held about ethnic minorities. A preliminary test for descriptive statistics was then performed introducing each group as the unit of analysis and the different codes as dependent variables. An overall description of the number of ideas gathered in each domain can be seen in Table 1.

All codes were submitted afterwards to cluster analysis in order to see what codes were more frequent among the discussion groups, and thereby, to uncover underlying cognitive dimensions. Ward's method was employed here as the algorithm for clustering the codes because it provides more interpretable and well-defined solutions than other methods. The following descriptions summarize clusters at two levels in the hierarchy (6 to 7 clusters and 4 clusters). Since clusters in the first level of the hierarchy represent the most cognitively accessible and agreed ideas among participants, second-level solutions with more than seven clusters were discarded because they were inconsistent and did not seem to offer any additional insights. First-level clusters are identified by uppercase letters and second-level clusters with lowercase letters.

The first cluster, A) Identification of ethnic minorities in the local context (n = 34) is separated from the rest and comprises ideas related to everyday life experiences of contact without assessments on the quality of this contact. The second cluster, B) Ethnic minorities as a social problem in the national context agglutinates various reasons that explain why talking about minorities is considered highly relevant. From this larger cluster, two subclusters emerged at the following level: on one hand expressions about the increasing number of immigrants⁴ in the country and the policies required for their social integration formed the first subcluster (a) (n = 46), and several reasons explaining preference for people coming from developed countries rather than from non-EU or developing countries (e.g., high differences in culture, history, phenotypes, socioeconomic level), made up the second subcluster (b) (n = 19). The third cluster, C) Knowledge about ethnic minorities at national level agglutinates perceptions regarding ethnic groups throughout the national territory. This larger cluster is subdivided into demoscopic perceptions of the ethnic communities in Spain (c) (n = 29), and expressions that criticize the biased treatment of news about ethnic minorities by the media when linking these communities to violence, crime, and social problems (d) (n = 27). The last cluster in this domain, D) Apathy toward the issue of minorities (n = 9) concentrated all comments that pointed out a lack of interest about this issue, and/or, the fact that is not a frequently-discussed topic among family or friends. Table 2 shows some extracts from the ideas contained in each cluster.

In order to draw a spatial representation of the distribution of ideas, all codes in this domain were submitted to a multidimensional scaling⁵ test (MDS-ALSCAL). A two dimensional solution seemed adequate because it explained most of the variance in the proximity-distance among ideas (R² = .99; Stress = .037). The perceptual map created depicts a confluence of ideas among the four groups interviewed: the closer the dots the more similar the participants' responses are. As shown in Figure 1, the vertical dimension concentrates actual notions on ethnic minority groups (local and national contexts). Contrary to the cluster analysis test, the criticisms related to the way the media treats news about these social groups appear at the upper side of the quadrant, just at the opposite end. The distance between these codes is clearer difference between the knowledge that participants have regarding ethnic minorities and the information displayed in the news about them.

Conversely, the horizontal dimension opposes two levels of perceived importance on talking about minorities. The ideas associated to this perception gather together on the left side -closer to the preferences for certain minorities-, whereas on the other hand, articulations about a lack of interest on the issue lay on the right side of the quadrant. Thus, in accordance with the cluster analysis test, the perceived importance of this issue seems to be linked to the preferences for certain groups, which ultimately implies competition for scarce resources. The scores in the intergroup thermometer reflect this preference more clearly: the participants feel closer to members of groups coming from Western Europe (Southern 82%, Central 80%, Northern 78.1%), whilst they feel the least close to Maghrebis (54%), Gypsies (48.9%), and people from Eastern Europe (47.6%). The perceived closeness to Latin Americans, Asians, and Southern Saharans seem to be moderate (76%, 63.2% y 60.1% respectively).

For the second domain, 17 codes were created from descriptions of the type of contact that participants usually maintain with members of ethnic minorities. The participants' assessments of contact included variations on the valence (positive, neutral, negative) and levels of involvement (personal, impersonal) during interactions. The cluster analysis revealed four larger clusters: the first one titled E) Frequent contact (n = 14), summarizing expressions of regular contact with people from ethnic minorities in general. The second cluster, F) Gypsy-Latino, presented a subdivision with descriptions about negative and impersonal contact with Gypsies, including situations in which their actions or behavior brought about negative social consequences (e) (n = 6); whereas the other subdivision consisted of statements of positive and personal contact with Latin Americans including friendship or romantic relationships (f) (n = 5). The third cluster, G) Neutral contact, showed some dispersion by incorporating on one hand ideas about lack of contact with minorities, assertions of having pleasant or cordial relationships with Maghrebis, and comments about minimum or impersonal interactions with Latin Americans (g) (n = 17); on the other hand, the second subcluster was formed of statements of few and impersonal interactions with Asians and Gypsies, and cordial relationships with Southern Saharans (h) (n = 9). The third cluster, H) Neutral-positive contact, shows even more dispersion with a first subcluster that incorporated ideas including neutral contact with Eastern Europeans, negative and impersonal contact with Maghrebis, and positive and impersonal contact with Gypsies and Northern Americans (i) (n = 8). On the contrary, the other subcluster grouped together positive yet impersonal contact with Southern Europeans, Eastern Europeans, and Asians; including negative contact experiences with Northern Europeans (j) (n = 8). Selected excerpts of the ideas contained in these clusters can be seen in Table 3.

The third domain of analysis integrated 23 codes corresponding to perceptions of closeness, distance or indifference towards ethnic minorities. The cluster test uncovered a first group of similarities titled I) Distance-Gypsies (n = 50), that enclosed a series of beliefs and detailed reasons of why participants feel rejection towards Gypsies (e.g., their lack of respect for social harmony, resistance to integrate, looking for trouble). The second cluster in this domain, J) Distance-closeness-Central (n = 86), integrated most comments about perceptions of distance from Maghrebis (e.g., victimization of women, religious fanaticism) and Eastern Europeans (e.g., criminal activities); but also included several reasons why participants feel closer to Central Europeans (e.g, geographical proximity, cultural, economic or physical similarities). The cluster K) Closeness (n = 83) gathered the largest number of ideas on perceived closeness toward Latin Americans (e.g., same language, religion), Southern and Northern Europeans (e.g., same history, European identity), Southern Saharans and Asians (e.g, "they do not cause troubles"). However, some ideas regarding experienced distance from Gypsies were also included.

Finally, the cluster L) Coexistence showed the greatest dispersion with two subgroups that included descriptions of coexistence (without any perceptions of closeness or distance) with Southern Saharans, Latin Americans, Northern Europeans and Central Europeans. Only a couple of ideas on perceived closeness to Maghrebis and Americans were included in this subcluster (k) (n = 12). The second subgroup grouped together perceptions of distance toward Asians and Latin Americans, but also certain disregard for Maghrebis (l) (n = 22); whereas a third subdivision included perceptions of closeness toward Gypsies, distance from Southern Saharans, and indifference to Eastern Europeans (m) (n = 24). Extracts from the ideas in these clusters are shown in Table 4.

In accordance with the objective of exploring what kind of relationship exists between the representations of minorities, the type of contact held with their members, and the assimilation of structural hierarchies in terms of closeness and distance, an additional MDS test was performed on the codes in the third and fourth domains taken together. Two dimensions explained most of the variance in the proximity among ideas (R² = .99; Stress = .064). As can be seen in Figure 2 the perceptual map is arranged according to a horizontal dimension that separates, on the left side, distance perceptions toward Gypsies, Maghrebis, and Eastern Europeans together with a generalized perception of closeness towards Central Europeans, whereas descriptive notions of contact, as well as scattered perceptions of distance from and closeness to the rest of the ethnic minorities, are gathered on the right side. Conversely, the second dimension contrasts ideas and beliefs about familiar minorities (upper quadrant), and expectations regarding those groups that are less familiar because of a lack of opportunities for contact or a lack of information about them (lower quadrant).

The codes considered for the fourth domain offer some support for this last interpretation. In this section, the positive and negative traits that participants evaluated as typical in each group were analyzed. Although an index for each ethnic group could be created out of most of the stereotypic traits that emerged in the focus groups, it goes beyond the purpose and pages of this study. Instead, a descriptive table indicating the number of interventions addressing the attributes in each group was created. According to Table 5, the assessments of the traits seen to be typical in Gypsies - especially the negative ones-, were the most prevalent; followed by the characteristics of Latin Americans, Asians, Maghrebis, and Southern Saharans.⁶ The traits of Eastern Europeans are the fewest in proportional terms, which might denote a lack of consensus or knowledge about the typical attributes of this specific minority. It is important to mention that participants provided well-defined descriptions about the minorities shown in Table 5, but this was not the case for other groups such as Northern, Central, or Southern Europeans. Moreover, because there were very few ideas describing or assessing traits in these groups (n = 31 overall), and the score obtained in the reliability test for them was unacceptable, they were dropped from the analysis.

Discussion

The present study explored how the representations of ethnic minorities are conformed in the cognitive schemas of the Spanish majority. Accordingly, the potential influence of perceived status differences among young Spaniards on contact with members of ethnic minorities was analyzed. The identification of different ethnic groups in the social setting denotes a high presence of these communities in the daily life of participants, whereas the awareness of such groups at the broader national setting is -to a certain extent-, linked to the information spread by the media even if participants are conscious of the negative treatment of immigration or ethnic minorities in the news. Because of the media, people know about the number of boats illegally entering the Spanish coasts, their access to social benefits, percentages of unemployed foreigners, or the "excessive" number of overall immigrants living in the country. However, the representations at local and national settings suggests that the perceptions derived from the information in the news media (e.g., "the Chinese are nearly the majority now"), differ from the participants' daily life experiences (e.g., "Well, I don't see any Gypsies when I go into town"). In this sense, the treatment of news on ethnic minorities seems to have a stronger effect at the abstract level, which is the part of the social reality that has not been experienced personally (Cea D'Ancona & Valles, 2011; Igartua et al., 2008).

The two additional representations were configured as variations on the perceived importance of the ethnic minority issue. On one hand, there is a homogenized view of ethnic minorities as a social problem, given that people used both "ethnic minorities" and "immigrants" as interchangeable concepts with the implied negative meaning of the latter. In fact, the participants emphasized the classic difference between "us" and "them" when talking about minorities in this section (see Tajfel & Turner, 1986). On the other hand, a lack of interest and/or knowledge on the issue emerged, something that has also been observed in other surveys on attitudes toward immigration (www.cis.es). Participants in this study, all undergraduate students, frequently expressed more concern about their professional careers than about social problems such as immigration or minorities, which they said, had little impact on them.

With regard to the representations of contact (domain 3) obtained from the set of tests, a pattern of "impersonalization" seemed to be dominant in the ideas clustered in this section. Although many of the positive personal contact experiences were associated with Latin Americans and Southern Europeans, most of the expressions basically described a lack of personal involvement in contact regardless of their negative or neutral valence. The following extract depicting a negative-impersonal contact with Gypsies exemplifies this type of pattern:

Me too! For example in my hometown, in the building where I live, a group of Gypsies live in the flat at the top floor. They are involved in drug trafficking... everybody knows it, everyone in my town knows it. But the police have not shown up yet.. .well, they came once: they [Gypsies] were sent to prison for six months but they are back again. (Focus group 2)

Another large proportion of ideas unveiled a tendency to interact in neutral and impersonal ways with members of ethnic communities, that is to say, despite describing a peaceful coexistence with members of minorities, there is little willingness to form affective or long-term relationships:

Hence, I have always had contact with people [from ethnic minorities] as my parents used to have [...]: renting out an apartment [to them] and so on; or people that used to come home and stay for a while. Argentines, Brazilians, Uruguayans, and Cubans used to live in my house back then. Also now, in the classroom, with some group tasks and so on. (Focus group 4)

Conversely, the ranking of ethnic groups drawn from the thermometer measure supported some findings in other European studies (Cea D'Ancona & Valles, 2009, 2011; Hagendoorn, Drogendijk, Tumanov & Hraba, 1998; Perez-Yruela & Desrues, 2007; Snellman & Ekehammar, 2005). Among the reasons justifying the positions in the rank, differences in culture, economy and phenotype were the most-often cited (see Gomez-Berrocal & Navas, 2000; Pettigrew & Meertens, 1995). Many perceptions of closeness to Western Europeans were based on cultural similarities among countries. This is: they assumed a common European identity, but not a Spanish one, when such differences were addressed:

Towards Europeans [feels closer to] because of the geographic proximity and culture: they are more similar, especially those in the South [...] the Portuguese are next door. The Italians are basically like us, very similar but with a different language; and, I don't know... the Greeks are not so similar, though. I think that the issue with Europeans is that, because of the European Union and all that stuff, they have made us feel like we share some common characteristics among all Europeans. Hence, whether you like it or not, it gets integrated into our lives. (Focus group 2)

Similarly, this European identity is used as a basis to explain perceptions of distance in relation to the other ethnic minority groups:

Traditions, for instance. If you deal with Latin Americans or South Saharans they see women in a lower position, but Europeans don't: they treat everyone equally and show more respect for civil rights and so on. Perhaps it is because of the economic resources they have [perceives distance]. Normally, Africans and 'those' alike [sic] usually have less money than Europeans. (Focus group 2)

It is more common seeing a blond person in Europe than seeing, for example, a black person. Probably because different things are associated: different conditions and social status are associated to blond and black people...but I don't see why, because blacks are Europeans too and they possibly live in marginalized conditions. I think this is why [feels more distance]. (Focus group 4)

In addition, the MDS of the codes about contact and distance showed that the ideas of distance from Gypsies, Maghrebis, and Eastern Europeans -those minority groups at the lowest positions on the scale-, appear very distant from the contact experiences. In other words, the perceptions of distance and rejection for these groups precede the experiences and shape expectations for any kind of interaction with them. While some researchers suggest that more familiarity and knowledge about outgroups can reduce homogeneous perspectives (e.g., Lee, 2001), the long record of conflict between Gypsies and non-Gypsies in Spain seems to be a determinant of the representations associated to this group (Gomez-Berrocal & Moya, 1999; Paez, 2004). This "familiarity" would explain all the ideas expressed about their attributes and about other familiar groups such as the Latin Americans.

Taken together, the findings in this study lead to the following theoretical implications: first, it offers a new frame of reference on how the schematic representations of ethnic minority groups are configured among young Spanish university students. In line with the concept of schema used in Harwood et al., (2000), such representations included not only latent information with respect to ethnic minorities, but also anecdotal knowledge, affective trends regarding interactions, and underlying beliefs about the willingness to advance or avoid contact with certain groups. Moreover, this research offered an insight on the way that social status differences are introjected and shape the relationships between the majority group and the ethnic minority groups. The hierarchy built on perceptions of closeness and distance towards these communities reflects such differences and configures an "expectation" that underlies group interaction processes. In this sense, the mental structures uncovered in this study constitute the basis from which contact and communication with outgroups will be performed.

The practical implications of this study offer a wider perspective for improving future contact interventions: particularly, it is already known that the closeness perceived to Western Europeans is associated to cultural, economic, historic, and general identity-based aspects (i.e., similarities). In the case of Latin Americans, closeness perceptions derive from sharing language and religion. Therefore, a focus on making salient similarities in contact interventions involving socially disadvantaged ethnic minorities (Gypsies, Maghrebis, Eastern Europeans) might help to reduce stigma and perceptions of distance from these outgroups at least at the "expectation" level. In addition, introducing information that disconfirms negative beliefs about each ethnic community may contribute to gradually modify anchored representations in these mental representations.

Limitations of this study include the relatively small sample of participants in the group interviews. Although groups were balanced in terms of age, education and socioeconomic levels, most of the participants in the discussion teams were mostly women. Gender-balanced groups and people of different ages and sociodemographic profiles would enrich the representations of minorities in the Spanish context. Likewise, only representations among members of the Spanish majority group were obtained, but not perceptions from members of ethnic minorities. Further replies must consider representations in both groups, as well as in different contact settings (e.g., academic, work context, international cooperation).

To conclude, this study aimed to broaden our understanding of how the representations and expectations of contact with ethnic minorities are organized among people that form a part of the largest majority group in Spain. Although it was not a primary objective, the methodology used in this study could be applied to investigate and interpret the association between the cognitive representations and behavior to be adopted in other intergroup relations such as between men and women. For the purpose of enhancing our current knowledge on intergroup social interactions, a closer look at the relationship between cognitive process and social behaviors in majority and minority groups is expected to contribute new and relevant pathways to improve intergroup relations.

Acknowledgments

We thank Valeriano Pineiro Naval and Maria Marcos Ramos from Salamanca University, and Claire Wright and Nilsa Marlen Tellez from the Autonomous University of Nuevo Leon for their support with coding and data analysis. This work is also part of the first author¹ s PhD thesis.

¹ With the sole exception of Gypsies, who arrived on Spanish soil in the fifteenth century.

² Numbers are reported in March 2013 (http://extranjeros.meyss.es).

³ Fundación Secretariado Gitano (www.gitanos.org).

⁴ Despite the fact that the conceptual difference between immigrants and ethnic minority groups was explained to participants in each group interview, they used both categories as interchangeable for addressing ethnic minorities.

⁵ Multidimensional scaling is a multivariate technique that is often applied in social sciences to uncover underlying dimensions in a series of similarity or distance assessments made by subjects (Arce, De Francisco & Arce, 2010). Such attributions are represented in a series of scatterplots that, taken together, configure a "perceptual map" in which the axes characterize the latent dimensions and the points are the subjects' beliefs, evaluations or opinions: the distances between points in the map mirror the subjective distances between the respondents.

⁶ Participants called all the people coming from Asian countries "Chinese", whereas they labeled all people coming from Eastern European countries "Romanians". Cea D'Ancona & Valles (2011) explain that this type of generalization is caused by proportional flows in these groups (e.g., Chinese residents in Spain are proportionally greater in number than Taiwanese or Koreans, while there are more Romanians than Bulgarians or Hungarians). The link between Maghrebis and South Saharans is more a product of considering the Islamic religion as a common denominator in both groups.

References

1. Allport, G. W. (1979). The nature of prejudice (3rd ed). Cambridge, MA: Addison-Wesley.         [ Links ]

2. Arce, C., De Francisco, C. & Arce, I. (2010). Escalamiento multidimensional: Concepto y aplicaciones. Papeles del Psicólogo, 37(1), 46-56. Retrieved from: http://www.papelesdelpsicologo.es/pdf/1795.pdf.         [ Links ]

3. Ashmore, R. D., & DelBoca, F. (1981). Conceptual approaches to stereotypes. In D. Hamilton (Ed.), Cognition processes in stereotyping and intergroup behavior (1st ed, pp. 1-35). New Jersey: Erlbaum.         [ Links ]

4. Cameron, L., Rutland, A., Turner, R., Holman-Nicolas, R., & Powell, C. (2011). Changing attitudes with a little imagination: Imagined contact effects on young children's intergroup bias. Anales de Psicología, 27(3), 708-717. doi:10.6018/135311.         [ Links ]

5. Cea D'Ancona, M. A., & Valles, M. S. (2009). Evolución del racismo y la xenofobia en España (Informe 2009). Madrid: OBERAXE. Ministerio de Trabajo e Inmigración.         [ Links ]

6. Cea D'Ancona, M. A., & Valles, M. S. (2011). Evolución del racismo y la xenofobia en España (Informe 2011). Madrid: OBERAXE. Ministerio de Trabajo e Inmigración.         [ Links ]

7. Emessik, D., & Mackie, D. M. (1989). Intergroup relations (1st ed.). Annual Review of Psychology, 40, 45-81. doi: 10.1146/annurev.ps.40.020189.000401.         [ Links ]

8. Esses, V. M., Haddock, G., & Zanna, M. P. (1993). Values, stereotypes, and emotions as determinants of intergroup attitudes. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition and stereotyping: Interactive processes in group perception (1st ed., pp. 137-166). Orlando, FL: Academic Press.         [ Links ]

9. Fiske, S. T., & Taylor, S. E. (1991). Social cognition (1st ed.). New York: McGraw Hill.         [ Links ]

10. Gamella, B. (1996). La población Gitana en Andalucía. Un estudio exploratorio de sus condiciones de vida (1st ed.). Sevilla: Departamento de Antropología y Trabajo Social de la Universidad de Granada.         [ Links ]

11. García, H., Adroher, S., & Blanco, M. (1996). Minorías étnicas: gitanos e inmigrantes (1st ed.). Madrid: Editorial CCS.         [ Links ]

12. García-Leiva, P. (2005). Identidad de género: modelos explicativos. Escritos de Psicología, 7, 71-81. Retrieved from: http://www.escritosdepsicologia.es/descargas/revistas/num7/escritospsicologia7_revision4.pdf.         [ Links ]

13. Gómez-Berrocal, C., & Moya, M. (1999). El prejuicio hacia los gitanos: características diferenciales. Revista de Psicología Social, 14(1), 15-40. doi: 10.1174/021347499760260055.         [ Links ]

14. Gómez-Berrocal, & Navas, M. (2000). Predictores del prejuicio manifiesto y sutil hacia los gitanos. Revista de Psicología Social, 15(1), 3-30. doi: 10.1174/021347400760259848.         [ Links ]

15. Goodman, M. D., & Gareis, K. C. (1993). The influence of status on decisions to help. Journal of Social Psychology, 133(1), 23-31. doi:10.1080/00224545.1993.9712115.         [ Links ]

16. Hagendoorn, L., Drogendijk, R., Tumanov, S., & Hraba, J. (1998). Interethnic preferences and ethnic hierarchies in the former Soviet Union. International Journal of Intercultural Relations, 22, 483-503. doi:10.1016/S0147-1767(98)00020-0.         [ Links ]

17. Harwood, J., Mckee, J., & Lin, M.C. (2000). Younger and older adults' schematic representations of intergenerational communication. Communication Monographs, 67(1), 20-41. doi:10.1080/03637750009376493.         [ Links ]

18. Hraba, J., Hagendoorn, L., & Hagendoorn, R. (1989). The ethnic hierarchy in The Netherlands: social distance and social representation. British Journal of Social Psychology, 28, 57-69. doi: 10.1111/j.2044-8309.1989.tb00846.x.         [ Links ]

19. Igartua, J. (2006). Métodos cuantitativos de investigación en comunicación (1st ed.). Barcelona: Bosch.         [ Links ]

20. Igartua, J. J., Muñiz, C., Otero, J. A., Cheng, L., & Gómez-Isla, J. (2008). Recepción e impacto socio-cognitivo de las noticias sobre inmigración. Revista de Psicología Social, 23(1), 3-16. doi: 10.1174/021347408783399552.         [ Links ]

21. Lee, A. Y. (2001). The mere exposure effect: An uncertainty reduction explanation revisited. Personality and Social Psychology Bulletin, 27, 1255-1266. doi: 10.1177/01461672012710002.         [ Links ]

22. Lin, M. C., Harwood, J., & Hummert, M. L. (2008). Young adults' intergenerational communication schemas in Taiwan and the USA. Journal of Language and Social Psychology, 27(1), 28-50. doi:10.1177/0261927X07309510.         [ Links ]

23. Lin, M. C., Zhang, Y. B., & Harwood, J. (2004). Taiwanese young adults' intergenerational communication schemas. Journal of Cross-Cultural Gerontology, 19(4), 321-342. doi:10.1023/B:JCCG.0000044687.83806.3e.         [ Links ]

24. Lorenzi-Cioldi, F., & Clémence, A. (2001). Group processes and the construction of social representations. In M. A. Hogg & S. Tindale (Eds.), Blackwell handbook of socialpsychology: Group processes (1st ed., pp. 311-333). Oxford: Blackwell.         [ Links ]

25. Macrae, C. N., Stangor, C., & Hewstone, M. (1996). Stereotypes and Stereotyping (1st ed.). New York: Guilford Press.         [ Links ]

26. Markus, H. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63-78. doi: 10.1037/0022-3514.35.2.63.         [ Links ]

27. Martínez, C. (2000). Intención de contacto, ansiedad y actitud en la relación con exogrupos. Revista de Psicología Social, 15(1), 31-44. doi:10.1174/021347400760259857.         [ Links ]

28. Morgan, D. L. (1997). Focus groups as qualitative research (2nd ed.). Beverly Hills: Sage.         [ Links ]

29. Navas, M. (1998). Nuevos instrumentos de medida para el nuevo racismo. Revista de Psicología Social, 13(2), 233-239. doi: 10.1174/021347498760350731.         [ Links ]

30. Osbeck, L. M., Moghaddam, F. M., & Perreault, S. (1997). Similarity and attraction among majority and minority groups in a multicultural context. International Journal of Intercultural Relations, 21(1), 113-123. doi: 10.1016/S0147-1767(96)00016-8.         [ Links ]

31. Páez, D. (2004). Relaciones intergrupales. In D. Páez, I. Fernández, S. Ubillos & E. Zubieta (Coords.). Psicología social, cultura y educación (1st ed., pp. 752-768). Madrid: Pearson educación.         [ Links ]

32. Páez, D., Valdoseda, M., Igartua, J.J., Basabel, N., & Irauguij, J. (1992). Las representaciones sociales del alcohol. Revista de Psicología Social Aplicada, 2 (2-3), 33-54. Retrieved from: http://diarium.usal.es/jigartua/files/2012/07/Paez-et-al-RPSA-1992.pdf.         [ Links ]

33. Pérez-Yruela, M., & Desrues, T. (2007). Opinión de los españoles en materia de racismo y xenofobia 2007. Madrid: OBERAXE. Ministerio de Trabajo y Asuntos Sociales.         [ Links ]

34. Pettigrew, T.F., & Meertens, R.W. (1995). Subtle and blatant prejudice in Western Europe. European Journal of Social Psychology, 25, 57-75. doi: 10.1002/ejsp.2420250106.         [ Links ]

35. Rueda, J., & Navas, M. (1996). Hacia una evaluación de las nuevas formas del prejuicio racial: las actitudes sutiles del racismo. Revista de Psicología Social, 11(2), 131-149. doi: 10.1174/02134749660569314.         [ Links ]

36. Scherer, R. F., & Petrick, J. A. (2010). The effects of gender role orientation on team schema: A multivariate analysis of indicators in a U.S. federal health care organization. Journal of Social Psychology, 141(1), 7-22. doi: 10.1080/00224540109600519.         [ Links ]

37. Snellman, A., & Ekehammar, B. (2005). Ethnic hierarchies, ethnic prejudice and social dominance orientation. Journal of Community & Applied Social Psychology, 15, 83-94. doi: 10.1002/casp.812.         [ Links ]

38. Staerklé, C., Clémence, A., & Spini, D. (2011). Social Representations: A Normative and Dynamic Intergroup Approach. Political Psychology, 32(5), 759-768. doi:10.1111/j.1467-9221.2011.00839.x.         [ Links ]

39. Stangor, C., & Schaller, M. (1996). Stereotypes as Individual and Collective Representations. In C.N. Macrae, C. Stangor & M. Hewstone (Eds.). Stereotypes and Stereotyping (1st ed., pp. 3-39). New York: Guilford Press.         [ Links ]

40. Stephan, W., Diaz-Loving, R., & Duran, A. (2000). Integrated threat theory and intercultural attitudes: Mexico and the United States. Journal of Cross-Cultural Psychology, 31(2), 240-249. doi: 10.1177/0022022100031002006.         [ Links ]

41. Stephan, W. G., & Stephan, C. (1985). Intergroup anxiety. Journal of Social Issues, 41, 157-176. doi: 10.1111/j.1540-4560.1985.tb01134.x.         [ Links ]

42. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & L. W. Austin (Eds.), Psychology of intergroup relations. (1st ed., pp. 7-24). Chicago: Nelson-Hall.         [ Links ]

43. Tropp, L., & Pettigrew, T. (2005). Relationships between intergroup contact and prejudice among minority and majority status groups. Psychological Science, 16(12), 951-957. doi: 10.1111/j.1467-9280.2005.01643.x.         [ Links ]

Correspondence:
Salvador Alvídrez
Facultad de Ciencias Políticas
y Administración Pública
Campus Mederos
Ave. Praga y Trieste s/n
Edificio de Posgrado
Col. Residencial Las Torres
64930, Monterrey (México).
E-mail: salvidrez@hushmail.com

Article received: 07-10-2013
Revised: 04-02-2014
Accepted: 20-08-2014