Data came from 132 first generation immigrant adolescents (45.2% female and 54.8% male) between 13 and 18 years and with a mean age of 15 (SD = 1.13) studying at public secondary schools in the Basque Country (Spain). 67% of the participants were Latin Americans, 24% came from Maghreb, and 9% had migrated from Eastern Europe. Adolescents from different regions of origins had a socio demographic homogeneous profile. Thus, there were not significant differences between groups neither on age (X² ₍₂₎ =3.371; p>0.05), nor on length of residence in Spain (X² ₍₂₎ =1.363; p>0.05). Also, corrected typified residuals analysis showed there were not relations between region of origin, sex (X² ₍₂₎ =2.67; p>0.05), and socio educational label of fathers (X² ₍₂₎ =9.86; p>0.05) and mothers (X² ₍₂₎ =5.16; p>0.05).

Student selection was conducted using a convenience sample, according to school availability and the voluntary participation of the students. Four public schools agreed to participate in the study. In order to ensure the participants had sufficient language proficiency to respond the surveys, teachers were asked to identify students whose linguistic skills were not appropriate for this task, and consequently, their participation on the study was discarded. Parent consent was requested by means of a personalized letter. All participants were informed about the goals and confidentiality conditions of the research in order to comply with ethical and legal frameworks. Participants completed the questionnaire protocol voluntarily in their educational centers and during regular class time. Questionnaire administration was done collectively and had an average length of 40 minutes. A member of the research team was available in order to help participants to understand some of the items, even if they were sufficiently language proficient to respond the questionnaires.

- Length of residence. An open-ended question was built to register the number of years and months of foreign students’ residence in Spain.

- School Achievement perception. School achievement was conceptualized as the perceptions of adolescents about their own academic performance, considering different school tasks as a whole. Spain’s data protection rules prevent access to student scores and previous studies confirm statistically significant positive relationships between students’ perceptions and their actual results (Hishinuma et al., 2001; Serna, Yubero, & Larrañaga, 2008; Stanton-Salazar & Dornbusch, 1995; Verkuyten & Thijs, 2002). An ad-hoc scale was designed that included students’ perceptions about their achievement and the feedback they perceived from teachers, peers, and families. This instrument consists of 4 likert-type items with 5 answer options (from 1 = unsatisfactory to 5 = excellent) (e.g., “In general, I think that my school performance is…”). Cronbach's alpha was 0.89 for the studied sample.

- Perception of discrimination. From the Immigrant Adolescent Questionnaire (Berry et al., 2006). Two items from this questionnaire were used in order to measure perceived discrimination at school. It has 5 Likert-type answer options (from 1 = never to 5 = very often) (e.g., “I have been teased or insulted because of my ethnic background,” “Teachers treated me unjustly because of my ethnic background”). Cronbach's alpha was 0.75 for the studied sample.

- School Adjustment Scale (Gilman & Huebner, 2006). Consisting of 6 Likert-type items, with 5 answer options (from 1 = totally agree to; 5 = totally disagree) (e.g., “I feel uneasy about going to school in the morning,” “I wish I could stop coming to school”). Cronbach's alpha was 0.63 for the studied sample.

- Cultural Dissonance between Home and School Scale. Derived from the Patterns of Adaptive Learning Scale (PALS) (Midgley, Maehr, & Hruda, 2000). It consists of 5 Likert-type items, with 5 answer options (from 1 = totally agree to 5 = totally disagree) (e.g., “I feel troubled because my home life and my school life are like two different worlds”) (α = 0.73).

- Immigrant Acculturation Scale (Bouhris & Barrette, 2006), to evaluate acculturation orientations. It consists of 5 Likert-type items with 7 answer options (from 1 = completely disagree to 7 = completely agree) (e.g., “I wish to maintain my own culture, instead of adopting the Basque Country’s culture” (Spanish). Cronbach's alpha was 0.62 for the studied sample. This scale has been used widely across others studies (Bourhis, Moise, Perreault, & Senecal, 1997; Montreuil & Bourhis, 2001; Bourhis et al., 2009a; Bourhis & Hill, 2010).

- Psychological well-being. A scale measuring psychological well-being for adolescents was used: Escala de Bienestar Psicológico para Adolescentes, (BIEPS-J; Casullo, 2002). This scale has been constructed based on the six dimensions of well-being proposed by Ryff (1989) and evaluates a global dimension, and four additional specific dimensions: control of situations, psychosocial links, projects, and self-acceptance. The scale is comprised of 13 items consisting of statements with three Likert-type options according to the degree of agreement with each of them (1 = disagree to 3 = agree) (e.g., “I have people who help me if I need it”). Its reliability for this sample was α = 0.74.

- Intergroup boundaries permeability. 1 Likert-type item (Navas et al., 2004) with 5 answer options was applied (1 = totally disagree; 5 = totally agree) (e.g., “Immigrants who live in the Basque Country (Spain), have the same opportunity to participate in social life under the same conditions as local people do”).

Mean length of residence in Spain was 4.6 years (SD = 3.3). Adolescents’ immigrants perceived their academic performance as medium-low, with 52% perceiving that their performance at school was low, while 40% considered it to be average, and only 8% rated it as high. They also showed medium-low perceptions of cultural dissonance, medium levels of school adjustment levels, and low levels of perceived discrimination at school. However, and regarding to the presence of discrimination from teachers and classmates, the most part of them (62.1%) considered it as moderate, whereas a low percentage of foreign students felt discriminated frequently (7.6%).

As expected (Berry et al., 2006) students’ orientation to integration or biculturalism was over the middle point on the scale between 1 and 7 points (M = 4.37; SD = 2.04). Additionally, immigrant students showed a high score in separation, that is, their tendency to prefer maintaining their cultural norms of origin at the expense of adopting elements of local culture, whereas the obtained mean in cultural assimilation, which represents the desire to leave their own cultural heritage to adopt the customs and values of the host society, was low. In the same way, their score in orientation to marginalization reflects a low willingness of immigrant adolescents to abandon their culture of origin and to adhere to the local culture, due to the fact that they feel uncomfortable in both of them. The tendency to individualism, that is, to distance from both the cultural background and the local in order to focus on personal social mobility, own needs and projections, was also low. Overall, it is important to notice that the highest scores were obtained in the acculturation orientations towards separation from the local culture (“I want to keep the culture of my country, instead of adopting the culture of the Spanish, Basque Country”), and integration (“I want to maintain the culture of my country, and also adopt some aspects of the culture of the Spanish (Basque Country”). Both orientations share an interest in maintaining one’s own cultural values background but, at the same time, adopting the host community cultural elements. In line with this finding, the lowest scores correspond to those options involving disengagement from them. This result indicates that immigrant students would be willing to adopt local elements, to the extent that it is also possible to maintain their cultural background (Table 1).

TABLE 1 TABLE 1 Descriptive statistics and Pearson’s correlations between immigrant students’ perceptions of academic performance and Psychosocial variables Source: own work. ^*

p<0.05

As for the distribution of immigrant students, 47% showed a high preference for cultural separation, one-third (33.3%) showed a moderate preference, and 19.7% reported a low orientation to separate from the culture of the host country. With regards to integration, 37.9% showed a moderate tendency, 33.3% a high tendency to biculturalism, and 28.8% a low tendency. The vast majority of immigrant adolescent (82.6%) reported a low orientation to assimilation, while 9.1% showed a high disposition towards this option, and only 8.3% scored average. As for marginalization, 88.6% showed low scores, and l6.1% showed a high tendency to reject adhering themselves to their original culture, as well as to the culture of the host society. Finally, 77.3% of foreign students showed a low disposition to individualism as an acculturation orientation, while 11.4% showed average levels and, 11.4% of them high levels.

Correlation analysis presented in Table 1 showed that there was no statistically significant relationship between perceived academic performance of immigrant adolescents and their length of residence in Spain. Instead, there was a positive and a statistically significant relationship between academic performance and a perception of school adjustment, and a negative and significant relationship with home/school cultural dissonance.

Academic performance of immigrant adolescents showed a positive relationship with permeability of inter-group boundaries. Conversely, academic achievement was negatively related both to discrimination at school and cultural separation. This finding shows the negative influence that the perception of being unfairly treated by peers and teachers has on the level of academic achievement among immigrants. Contrary to what was expected, there were no statistically significant relationships with integration, assimilation, marginalization, and individualism (see Table 1).

With the aim of identifying the predictors of immigrant adolescents’ academic performance, a linear stepwise regression analysis was conducted that included all the variables with statistically significant correlations with the perception of achievement from these students. Variables were entered in the following order: sex, ethnic group, school adjustment, psychological well-being, cultural dissonance, perceived discrimination at school, inter-group boundaries permeability, and separation. As shown in Table 2, the last regression model included 4 steps (F(4.113) = 23.24; p =0.001). The model explains 44% of foreign students’ perceptions of academic achievement (R² = 0.46 y R² adjusted = 0.44). The variables retained in the model were: cultural dissonance between home and school, ethnic discrimination perception at school, school adjustment, and psychological well-being (see Table 2).

TABLE 2 TABLE 2 Multiple linear stepwise regression analysis predicting immigrant students’ academic performance Source: own work. ^*

p<0.05

Given that cultural dissonance, psychological well-being, school adjustment, and ethnic discrimination perception at school explained academic performance perceptions’ variability, a multiple mediation model was explored. Thus, in order to know the possible dampening effects of cultural dissonance on academic performance, we present the analysis of mediation system-based macro SPSS from Sobel Test, which determines whether the indirect mediators effect is significantly different from zero. If the confidence interval does not include 0, then the effect is significant and mediation exists (Preacher & Hayes, 2004) (Figure 1).

Figure 1.

Note: **p<0.01; ***p<0.001