New expressions of racism among young people in Spain: an adaptation of the Meertens and Pettigrew (1992) prejudice scale.

Rodriguez, Francisco J. ; Herrero, Juan ; Ovejero, Anastasio 等

Prejudice is a complex phenomenon and a relevant issue in today's society insofar as it is related to social exclusion and discrimination (Pettigrew, 2008). This social character of prejudice has been highlighted by many authors, who suggest that direct and continuing contact with different ethnic groups can give rise to a perception of being threatened, both real--in terms of the distribution of resources and welfare--and symbolic--in terms of values and traditions (Stephan & Stephan, 2000). Moreover, the influence of the different types of threats on prejudice varies as a function of the exogroup and the historical and social context. In this sense, many observers have reiterated how certain public expressions of discrimination and prejudice toward other ethnic groups have been the object of public sanctions (even at the legal level) in Western societies (Pettigrew, 1998). Certain types of behavior and opinions about the exogroup which indicate blatant forms of prejudice and discrimination in terms of race, sex, age or sexual condition are increasingly being considered as "politically incorrect." In these societies, the foreseeable disappearance from the public sphere of "traditional" prejudice has given rise to an important theoretical view. In this sense, McConahay (1983, 1986) has argued that a new form of prejudice exists: modern prejudice. According to this author, modern prejudice will be sustained by beliefs that have nothing to do with social matters, that is, the new racism arises from the ambivalence between the values of one ethnic group (equality and liberty) and the negative feelings aroused by members of another group. Meertens and Pettigrew (1992) and Pettigrew and Meertens (1995) provide a European perspective on this issue, where immigrants are faced with differences in their country of destination (language, religion, and culture) and have their status as citizens questioned, often as a function of the socioeconomic necessities of the receiving country.

Racist behavior appears when social norms are weak or ambiguous, so that prejudice is attributed to factors other than the ethnic group. For this reason, it is not surprising that prejudice is constantly evolving toward new forms which are more difficult to recognize and combat. A distinction is thus made between blatant and subtle prejudice. Blatant prejudice has two forms: a perception that the exogroup is threatening, with the result that this group is rejected; and opposition to close or intimate contact with members of the exogroup. Subtle prejudice, on the other hand, refers to and is identified with traditional values, the exaggeration of cultural differences, and the negation of positive emotions (Meertens & Pettigrew, 1997). Prejudice can therefore be thought of as a complex phenomenon which can be seen as a mode of resolving problems and tensions which have come about in different contexts of interracial contact: in the experience lived by members of the racist group who find in the group that is the object of their discrimination an escape valve for their social and psychological difficulties (Wievorka, 1992). Prejudice is also difficult to evaluate because in modern democratic societies there have been systematic campaigns against prejudice, racism, and xenophobia which have led people to seek to appear to be tolerant without abandoning their prejudiced attitudes (Gaertner & Dovidio, 1977; Meertens & Pettigrew, 1997; Saucier, Miller, & Doucet, 2005). Prejudice has traditionally been evaluated through questionnaires which attempt to measure the degree of agreement with certain statements related to racial issues (explicit measures of prejudice). In recent times it has been observed that this methodology does not capture social realities because people yearn for social desirability and hence offer politically correct responses.

Faced with this problem, in the final decades of the 20th century, less reactive questionnaires were designed. Of particular note is the work of Meertens and Pettigrew (1992) whose aim was to make better predictions of behavior in situations where social desirability is important; that is, when people have the opportunity (sufficient time) motivation--both external (they do not want to appear prejudiced) and internal (existence of values contrary to prejudice)--to evaluate the consequences of explicit attitudes, they are no longer the best predictors of behavior toward members of the other group. In other words, the automatic activation of negative attitudes toward different groups may be modified by the social context (Kawakami & Dovidio, 2001).

Interest in the implicit aspects of prejudice thus comes from the prevailing norm in our society that open expression of our prejudices should be avoided. This interest has in turn been reinforced by methodological advances which have made it possible to measure the implicit attitudes not captured by traditional methods (Meertens & Pettigrew, 1997). While it is possible therefore to continue using traditional self-report questionnaires, new items capable of identifying subtle prejudice should be included. In Europe, the increasing immigrant population has brought ethnic question to the forefront, with racist and xenophobic attitudes being detected in growing segments of the population. There is therefore a more pressing need to improve the detection of racist and xenophobic attitudes and trends in order to be able to predict and correct them. As Michel Wieviorka (1992) noted some years ago, in our view correctly, our duty at present is not so much to explain, directly or indirectly, the wave of racism which threatens Europe. What is important is to construct the instruments necessary for understanding these types of phenomena. This is the objective of the present study--to analyze and adapt a scale of subtle and blatant prejudice toward the exogroup based on adaption of Rueda and Navas (1996) using a new, broad sample of adolescents in the Autonomous Community of the Principality of Asturias (Spain). This region has a small inflow of immigrants (overwhelmingly Latin American) and a small settlement of gypsies.

METHOD

Participants

The participants in the study are 1,782 high school adolescents in the region of Asturias, Spain of both sexes (48.7% male and 51.3% female), aged between 12 and 18 (mean = 15.16; SD = 1.39), and the majority of whom have the perception of coming from a medium-level socioeconomic background (89.8% compared to 7.6% who feel they come from a high-level background, and 2.6% from a low level). With regard to level of education, 68 (3.8%) were in the first year of Obligatory Secondary Education (E.S.O.), 237 in the second year (13.3%), 296 in the third year (16.6%), 859 in the fourth year (48.2%), and 322 in the last two years of high school (18.1%). This range permits the detection of social and racial prejudice during one of the most challenging periods of development from a social and educational point of view.

Variables

Blatant and subtle prejudice (Items 1-20). The Meertens and Pettigrew (1992) scale of blatant and subtle prejudice, translated into Spanish by Rueda, Navas, and Gomez-Berrocal (1995), and adapted to juvenile realities in the context of the research was used. This scale consists of 20 items, 10 of which evaluate subtle prejudice (1, 3, 5, 6, 11, 12, 14, 16, 18, 19) and the other 10 evaluate blatant prejudice (2, 4, 7, 8, 9, 10, 13, 15, 17, 20), following the response format used by Meertens and Pettigrew (5-point Likert scale where 1 = complete disagreement and 5 = complete agreement).

Feeling toward the exogroup (Item 21). Respondents have to indicate their attitude toward the group on a scale of 10 to 100, where 10 indicates a "highly unfavorable attitude" and 100 a "highly favorable attitude." The use of a single item to obtain an index of the attitude toward an exogroup has been used in the research on attitudes toward African Americans (Stangor, Sullivan, & Ford, 1991), as well as in the measurement of attitudes toward gypsies in Spain (Gomez-Berrocal & Navas, 2000).

Political ideology. This is an evaluation of the political views of adolescents according to where they place themselves on a 10-point scale, ranging from the extreme right (1) to the extreme left (10).

Socio-demographic characteristics. Information was obtained on age, gender (1 = male; 2 = female), socioeconomic level (1 = high, 2 = medium; 3 = low), and the residence (1 = rural; 2 = urban).

Procedure

The questionnaire was distributed as part of the campaign of sensitization against racism and xenophobia of the Movement for Peace, Disarmament, and Freedom-Asturias (MPDL-A) during the academic years 2004-05 and 2005-06 in several high schools in the region of Asturias. Participants were from 20 secondary-level schools in Asturias, 14 in urban areas, and 6 in rural areas. Participants were selected using stratified sampling in two stages, with the schools being selected in the first stage and the students in the second. The schools selected are representative of the region of Asturias, with greater representation of urban-based schools.

Following a series of contacts with principals, a common agenda for applying the instruments was established. During these contacts, the researchers explained the objectives of the study and emphasis was placed on the voluntary nature of participation. A total of 1,782 adolescents took part in the study.

Analysis

The analysis was carried out in two stages. In the first stage, the factor structure ofthe blatant and subtle prejudice scales was analyzed and checked whether the structure observed for the adult population was maintained in the adolescent population. In this stage the psychometric properties of the scales were explored. In the second stage, the main covariates of subtle and blatant prejudice were analyzed. In this stage multilevel regression techniques were used due to the hierarchical structure of the data (students nested in schools). This technique permits standard errors to be more reliably estimated because the clustering of the data in level 1 (schools) is taken into account (Raudenbusch & Byrk, 2002).

RESULTS

Factor Structure

We first tried to replicate the factor structure of the blatant and subtle prejudice scales. Following Meertens and Pettigrew (1997), both scales contain various sub-scales in the adult population. Thus, the blatant prejudice scale comprises a sub-scale of Rejection and Threat (6 items) and another of Intimacy (4 items). The subtle prejudice scale, on the other hand, is comprised of a sub-scale of Traditional Prejudice (4 items), another on Cultural Prejudice (4 items), and a sub-scale of Affective Prejudice (2 items). To check the consistency of this factor structure in the data, a Principal Component Analysis with orthogonal rotation (Varimax) was carried out separately for the 10 items of blatant prejudice and the 10 items of subtle prejudice, with the criterion of extraction being the number of original factors postulated by the authors. The results are presented in Table 1.

Focusing initially on blatant prejudice, it can be observed that items 8, 17, and 15 have very similar loadings for both factors, leading to interpretation problems. The internal consistency (Cronbach alpha) for the original scales in the sample is adequate (0.78). Regarding the subtle prejudice scale, it was found that items 05 and 03 were not clearly assigned to the original factors, which again caused problems of interpretation. Moreover, internal consistency is low for the Traditional and Affective factors, and moderate for the Cultural factor. These preliminary results raise doubts as to whether the structure of the sub-scales observed for the adult population is maintained in the adolescent population.

While these factors were extracted to reflect the structure proposed by Meertens and Pettigrew (1997), once it was confirmed that this structure did not adequately mirror the sample data, a principal component analysis with Promax rotation for each scale with extraction for values greater than or equal to 1 was carried out. Following this, items with a low communality (less than 0.30) were eliminated. This led to the elimination of items 19, 18, 06, and 09 from the scales, which provided a satisfactory factor solution, with loadings greater than 0.40 and cross-loadings less than 0.30 in other factors. The results are presented in Table 2.

As can be seen in Table 2, the factor structure is now reduced to one factor per scale, with no differentiation between the original subscales. Moreover, the loadings of the items in the factor are satisfactory, which suggests that for the sample analyzed, the single factors of blatant and subtle prejudice adequately describe the structure of the data.

Multi Level Analysis

The next step was to analyze the covariates of blatant and subtle prejudice with statistical techniques which permit the hierarchical structure of the data to be taken into account. We estimated two multilevel regression models with blatant and subtle prejudice as dependent variables and gender, age, socioeconomic status, affection toward the exogroup, political ideology, and belonging to a rural or urban area as covariates. The results for each of these models are presented in Table 3.

Regarding blatant prejudice, it can be seen that the fixed effects analyzed are significant, except for socioeconomic status. Adolescents who are male, who lean ideologicaily toward the right, and who live in a rural environment tend to be more in agreement with expressions which reflect blatant prejudice. Younger adolescents and those who have more negative feelings toward the exogroup also show higher levels of blatant prejudice.

With regard to subtle prejudice, we observed that the set of explanatory variables present a different pattern. Age, socioeconomic status, and living in an urban or rural environment are not found to have a statistically significant relationship with subtle prejudice. As with blatant prejudice, male adolescents, those with a right-leaning ideology, and those who have less positive feelings toward the exogroup are most in agreement with statements which reflect subtle prejudice.

Finally, after accounting for the clustering of the participants in schools, both regression models show that the means of blatant and subtle prejudice vary significantly. This can be interpreted as indicating that there is room for new variables which could explain this variation in racism across schools.

DISCUSSION

This research focused on the analysis of two scales of blatant and subtle prejudice in a sample of 1,782 adolescents in the region of Asturias, Spain. The main objective was to analyze whether the psychometric characteristics and factor structure of the original scale obtained in other studies with adult populations can be generalized to the school-based adolescent population. Another aspect of the study was the analysis of different covariates of blatant and subtle prejudice.

Regarding the first objective, it was observed that the data do not allow us to confirm that the structure proposed for the adult population is comparable to that of the school-based adolescent population. In contrast to the two general scales of blatant and subtle prejudice and their corresponding sub-scales which have been studied in the literature, we observed a uni-dimensional structure for each scale. These results, obtained through a process of depuration of the scales and progressive elimination of items with low commonality, suggest that the items corresponding to blatant prejudice and those corresponding to subtle prejudice evaluate a single dimension--prejudice--subtle or blatant. With regard to blatant prejudice, no differences were found between the items referred to as blatant hostility (threat and rejection) and those which are related to intimacy; nor are there differences between items which refer in a "subtle" way to both traditional and cultural prejudices or to affective prejudice. In fact, these items appear not to belong to the sub-scale of subtle prejudice itself, at least with respect to the school-based adolescents analyzed in this study.

Regarding the second objective, namely the analysis of the covariates of blatant and subtle prejudice, different patterns were observed between the scales as well as coincidences. Among the coincidences, we note that both types of prejudice are more characteristic of males, those who have a right-leaning ideology and those who have negative affection toward the exogroup. The latter two variables (ideology and affection toward the exogroup) have been consistently highlighted in the literature as important explanatory variables of prejudice, both subtle and blatant.

The key difference found is that blatant prejudice seems to be more prevalent among younger adolescents and those who live in rural environments, whereas these relationships are not observed for subtle prejudice. If, in accordance with the conceptualization of both types of prejudice, blatant prejudice is a form of prejudice toward the exogroup that societies appear to censure and is thus less likely to be manifested in public (Meertens & Pettigrew, 1992), the results indicate that this type of prejudice would be more typical of rural environments and of younger adolescents. This sociological distribution was not observed for subtle prejudice, with the levels being similar in rural and urban environments and among adolescents of different ages. Although results in this study suggest that the two scale scores of both blatant and subtle prejudice might be useful in differentiating racist and xenophobic attitudes among school-based adolescents in Spain, we should be cautious about the generalizability of results. However, some limitations of this study preclude going beyond what our data tell us. First, participants in the study do not represent the general population of school-based adolescents in Spain since our sample is not probabilistic and therefore not representative of the population. Second, it is unclear if the exogroups used (gypsies and immigrants) to which the items of the self-report questionnaire refer in this study would have influenced the adolescents' responses. In this sense, in other cultural contexts with ethnic minority groups that are neither immigrants nor gypsies (African American and native American in the USA or Indian natives in some parts of South America, to mention a few) results might be different.

In conclusion, although public expressions of racism are less frequent in Western societies--that is, in societies which are destinations for persons of different ethnic groups--it has yet to be confirmed whether this old style of racism has evolved into new forms. There is a general tendency in these receiving societies to make an effort to erradicate racist attitudes and behaviors, which has resulted in a drastic reduction in the percentage of people who, in surveys, say they are in agreement with racist views. However, more than just a reduction in prejudice, what seems to be happening is a profound transformation which is giving rise to new forms of racism and prejudice (Pettigrew, 1987, 2008) while at the same time racism and prejudice are becoming more subtle (Gaertner & Dovidio, 1977; Meertens & Pettigrew, 1997; Pettigrew & Meertens, 1995). The reliable measurement of this new form of racism is, in our opinion, necessary for identifying racist attitudes in the social context. Once this is achieved, we might be able to design and implement intervention programs which promote multiculturalism and which are based on reliable data on the prevalence of racist attitudes in society.

REFERENCES

Gaertner, S. L., & Dovidio, J. F. (1977). The subtlety of white racism, arousal and helping behavior. Journal of Personality and Social Psychology, 20, 218-222.

Gomez-Berrocal, C., & Navas, M. (2000). Predictores del prejuicio manifesto y sutil hacia los gitanos. Revista de Psicologia Social, 15(1), 3-30.

Kawakami, K., & Dovidio, J. F. (2001). The reliability of implicit stereotyping. Personality and Social Psychology Bulletin, 2(2) 212-225.

McConahay, J. B. (1983). Modern racism and modern discrimination. Personality and Social Psychology Bulletin, 9, 551-558.

McConahay, J. B. (1986). Modern racism, ambivalence, and the modern racism scale. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination and racism. New York: Academic Press.

Meertens, R. W., & Pettigrew, T. F. (1992). Le racism voile: Dimensions et mesure. In M. Wieviorka (Ed.), Racism et modernite (pp. 109-126). Paris: La Decouverte.

Meertens, R. W., & Pettigrew, T. F. (1997). Is subtle prejudice really prejudice? Public Opinion Quarterly, 61, 54-71.

Pettigrew, T. F. (1987). New patterns of racism: The different worlds of 1984 and 1964. Rutgers Law Review, 1, 673-706.

Pettigrew, T. F. (1998). Reactions towards the new minorities in Western Europe. Annual Review of Sociology, 24, 77-103.

Pettigrew, T. F. (2008). The social science study of American race relations in the twentieth century. Social and Personality Psychology Compass, 2, 318-345.

Pettigrew, T. F., & Meertens, R. W. (1995). Subtle and blatant prejudice in Western Europe. European Journal of Social Psychology, 25, 57-75.

Raudenbusch, S., & Byrk, A. (2002). Hierarchical linear and nonlinear models: Applications and data analysis methods. London: Sage.

Rueda, J. F., & Navas, M. (1996). Hacia una evaluacion de las nuevas formas del prejuicio racial: Las actitudes sutiles del racismo. Revista de Psicologia Social, 11, 131-149.

Rueda, J. F., Navas, M. S., & Gomez-Berrocal, C. (1995). Las nuevas expresiones del racismo: Adaptacion de una escala de prejuicio sutil. In J. C.

Sanchez & A. M. Ullan (Eds.), Procesos psicosociales basicos y grupales (pp. 357-370). Salamanca: Eudema.

Saucier, D. A., Miller, C. T., & Doucet, N. (2005). Differences in helping whites and blacks: A meta-analysis. Personality and Social Psychology Review, 9(1), 2-16.

Stangor, C., Sullivan, L. A., & Ford, T. E. (1991). Affective and cognitive determinants of prejudice. Social Cognition, 9(4), 359-380.

Stephan, W. G., & Stephan, C. W. (2000). An integrated threat theory of prejudice. In S. Oskamp (Ed.), Reducing prejudice and discrimination (pp. 23-45). Mahwah, NJ: Erlbaum.

Wieviorka, M. (1992). El espacio del racismo. Barcelona: Paidaos.

Juan Herrero, Department of Psychology. University of Oviedo, Oviedo, Spain

Anastasio Ovejero, Department of Psychology, University of Valladolid, Palencia Spain

Andrea Torres, Department of Psychology, University of Oviedo, Oviedo Spain

Requests for reprints should be sent to Franscisco J. Rodriguez, Departamento de Psicologia, Facultad de Psicologia, despacho 215, Plaza Feijoo, s/n, 33003-Oviedo, Spain, E-mail: gallego@uniovi.es

Table 1. Original (1) and obtained factor structure of the blatant and
subtle prejudice scales (N = 1,782).

 Blatant Prejudice

 Threat and
 Rejection Intimacy

Item 10 .67# .42
Item 09 .69# .18
Item 08 .44# .49
Item 02 .75# .06
Item 17 .39# .43
Item 15 .54# .58
Item 20 .27 .69#
Item 07 .05 .78#
Item 04 .31 .64#
Item 13 .17 .79#
[alpha] .78 .78

 Subtle Prejudice

 Traditional Cultural Affective

Item 05 .30# .49 .39
Item 01 .76# -.02 .09
Item 14 .76# .16 .01
Item 03 .44# .48 .33
Item 11 .18 .73# .20
Item 06 .24 .61# .30
Item 12 .05 .72# .06
Item 16 -.09 .69# .04
Item 19 .07 .17 .82#
Item 18 .05 .13 .81#
 .61 .70 .62

(1) Bolded loadings correspond to the original items for each
subscale.

Note: Bolded loadings correspond to the original items for each
subscale is indicated with #.

Table 2. Principal component analyses with Promax rotation for blatant
and subtle

 prejudice scales (1) (N = 1,782)

 Blatant Prejudice Subtle Prejudice

Item 10 .56 Item 05 .50
Item 08 .60 Item 01 .72
Item 17 .61 Item 03 .69
Item 15 .65 Item 11 .71
Item 20 .81 Item 12 .60
Item 07 .65
Item 04 .87
Item 13 .73
alpha .85 .70

(1) Loadings lower than .40 not shown

Table 3. Results of multilevel regression analyses of covariates of
blatant and subtle prejudice (N = 1,782)

 Blatant Prejudice

 B S.E. P
 Fixed Effects

Intercept 33.64 2.06 <.001
Male (a) 1.09 0.29 <.001
Age -0.30 0.12 .011
Socioeconomic status -0.67 0.44 .131
Affection towards Exogroup -0.21 .01 <.001
Ideology 0.24 0.07 <.001
Rural (a) 2.01 0.49 <.001
 Random Effects
Schools 31.00 1.08 <.001

 Subtle Prejudice

 B S.E. p
 Fixed Effects

Intercept 20.99 1.51 <.001
Male (a) 0.42 0.18 .020
Age -0.09 0.08 .244
Socioeconomic status -0.26 0.28 .368
Affection towards Exogroup -0.08 0.01 <.001
Ideology 0.14 0.04 .001
Rural (a) 0.80 0.73 .299
 Random Effects
Schools 12.69 0.45 <.001

(a) Category references are: female and urban