Development of a cultural self-efficacy scale for adolescents (CSES-A)

Share Embed


Descrição do Produto

International Journal of Intercultural Relations 33 (2009) 301–312

Contents lists available at ScienceDirect

International Journal of Intercultural Relations journal homepage: www.elsevier.com/locate/ijintrel

Development of a cultural self-efficacy scale for adolescents (CSES-A)§ Elena Briones a,*, Carmen Tabernero b, Carlo Tramontano c, Gian Vittorio Caprara c, Alicia Arenas d a

Department of Personality, University of Salamanca, Avenida La Merced, 109-131. 37005 Salamanca, Spain Department of Education, University of Cordoba, Cordoba, Spain Department of Psychology, University of Rome ‘‘La Sapienza’’, Rome, Italy d Department of Basic Psychology, University of Salamanca, Salamanca, Spain b c

A R T I C L E I N F O

A B S T R A C T

Article history: Accepted 20 March 2009

We developed a cultural self-efficacy scale for adolescents (CSES-A) and tested its psychometric properties using both exploratory and confirmatory factor analyses. Cultural self-efficacy (CSE) was defined as person’s perception of his/her own capability to function effectively in situations characterized by cultural diversity. On the basis of Bandura’s guideline for the development of a domain-specific self-efficacy measure, we tailored 50 items after reviewing literature about cultural competence, adolescents’ school-problems and social self-efficacy scales developed in previous studies in intercultural contexts. After pre-testing and analyzing psychometric properties of the scale, we selected 33 items. Eight hundred sixty-eight adolescents with five different cultural origins completed a set of questionnaires, including the CSES-A, internal control expectancies, general self-efficacy, academic expectancies, number of people from diverse cultures they keep in touch with, acculturation attitudes, perceived enrichment of other cultures, acculturation stress and demographic data. An EFA with MPLUS 2.14 highlighted a five-factor solution with 25 items that was supported by a subsequent CFA. The five factors were: self-efficacy in mixing satisfactorily with other cultures, in understanding different ways of life, in processing information from other cultures, in coping with loneliness and in learning and understanding other languages. The pattern of correlation with internal control expectancies, general self-efficacy and cultural variables supported the validity of the scale. CSES-A may be useful for future research on multicultural contexts, in which selfefficacy in cultural adaptation could be a fundamental variable. ß 2009 Elsevier Ltd. All rights reserved.

Keywords: Cultural self-efficacy Adolescents Cultural diversity Cultural settings

1. Introduction Immigration is a world-wide phenomenon. In the past few years, Spain has become the most multiethnic country in the European Union (OCDE, 2006), even though immigration is a relatively new phenomenon in this country. The Ministry of Education and Science showed that immigrant students represent 8.44% of the total of non-university students (MEC, 2007). Compulsory Secondary Education in Spain is characterized by the highest level of growth in the enrolment of foreign students in the last 10 years.

§

The research was supported by the Spanish Ministry of Science and Technology under Grant BSO 2003–09222/PSCE and SEJ 2006–07741/PSCE. * Corresponding author. Tel.: +34 626 576193; fax: +34 923 29 46 07. E-mail addresses: [email protected] (E. Briones), [email protected] (C. Tabernero), [email protected] (C. Tramontano), [email protected] (G.V. Caprara), [email protected] (A. Arenas). 0147-1767/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijintrel.2009.03.006

302

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

This growing cultural diversity in the school context makes often adolescents mix with students from other ethnic and cultural origins. This diversity can give rise to a lack of understanding and harmonic co-existence that should lead, in turn, to problems in social adaptation for students. Therefore, the growing ethnic diversity should imply an effort in educational institution aimed at promoting the functional value of the capacity to handle both one’s own and others’ culture (Bandura, 2006a). In this way, Bandura’s Cognitive Social Theory emphasizes the importance of self-efficacy, namely believes in one’s own ability to carry out actions in a social context with the intention of achieving the desired result (Bandura, 1986, 1992, 1997), as a pivotal predictor of people’s behaviors. In multicultural context, Fan and Mak (1998) found out that students with greater cross-cultural self-efficacy were more likely to feel at ease and to mix with students from host society, fostering their socio-cultural adaptation. There is a general accordance in considering self-efficacy believes to be especially important in cross-cultural interactions (Bandura, 2006a; Fan & Mak, 1998; Harrison, Chadwick, & Scales, 1996; Hechanova-Alampay, Beehr, Christiansen, & VanHorn, 2002; Li & Gasser, 2005; Tsang, 2001). In such interactions, sojourners and local residents face many barriers that hinder their effectiveness, like verbal and non-verbal communication, cultural ignorance and fear of being rejected, which make these cross-cultural interactions more stressful than communications that occur within the same culture (Church, 1982). Bandura (1986, 1992, 1997) stated that self-efficacy affects human behavior through four processes: cognitive, affective, motivational and selection. Hence, in a multicultural context, it is possible to hypothesize that people with a high level of cross-cultural self-efficacy may be more likely to succeed in their interactions and expect positive experiences than those with low self-efficacy (cognitive process). Likewise, people with high cross-cultural self-efficacy may be less likely to feel anxiety in cross-cultural interactions than those with low self-efficacy and feel more satisfaction when establishing new relationships with people from other cultures (affective process). Moreover, people with high cross-cultural self-efficacy may be more prone to have cross-cultural contacts and enjoy its benefits (motivational process). Finally, individuals with high cross-cultural self-efficacy may choose to have contact with their hosts even despite the fact that cross-cultural communication may entail to have a certain amount of cultural knowledge, language and tolerance for ambiguity (selection process). Hence, self-efficacy in cross-cultural interactions has been related to the socio-cultural adaptation of the sojourners. Tsang (2001) suggested that people who feel confident in their efficacy in managing their own lives may more actively seek new cultural experiences. Feedbacks related to their new behavior foster cultural knowledge and reduce uncertainty in future cross-cultural interactions. In the same line, Harrison et al. (1996) found that the greater the cross-cultural selfefficacy of American expatriates the more they report better adjustment in Europe. Likewise, in a longitudinal study Hechanova-Alampay et al. (2002) found that in international students general and social self-efficacy was significantly positive related to their adaptation to the new culture and negative related to their stress levels. Piontkowski, Florack, Hoelker, and Obdrza´lek (2000) studied different variables related to interaction among groups (e.g., contact, self-efficacy and perception of cultural enrichment), evaluating the extent to which they may predict acculturation attitudes. For this purpose, they selected as a theoretical framework Berry’s approach (1992), according to which, when two different cultural groups come into contact for a long time, they are involved in an acculturation process that entails change in each group. In this process it is assumed that members of both groups were characterized by specific attitudes concerning how this acculturation process will take place. Berry and colleagues’ (Berry & Kim, 1988; Berry, Kim, Power, Yong, & Bujaki, 1989) model of acculturation is based on the combination of two dimensions, namely intention to maintain cultural identity and intention to maintain relations with other groups, that allow to identify four acculturation attitudes: integration, assimilation, separation and marginalization. Who has an integration attitude prefers that each one maintained his/her own culture of origin, but at the same time promotes participation in the culture of the host society. Who has an assimilation attitude prefers to abandon the culture of origin, and supports inter-group contacts and relations. Who has a separation attitude prefers to maintain the culture of origin and reject contact between cultures. Finally, who has a marginalization attitude reject both intercultural contacts and maintaining the culture of origin. To sum up, researches carried out in this theoretical framework (Allard & Landry, 1992; Briones, Tabernero, & Arenas, 2005; Piontkowski et al., 2000) have shown that people with a strong belief in their general or social capability are more likely to feel prone and motivated to integrated themselves in another cultural group. On the other hand, the lower their level of self-efficacy in an inter-group situation, the more they avoid contacts with another cultural groups and prefer attitudes of separation or marginalization. Nevertheless, up to now literature concerning the measurement of self-efficacy in situations of interaction with people from different cultures or in diverse cultural contexts has involved studies characterized by measures of self-efficacy either general (Harrison et al., 1996; Hechanova-Alampay et al., 2002; Piontkowski et al., 2000; Tsang, 2001) or too specific (e.g., nursing in Vargas, Molino, Shellman, Cantero, & Bernal, 2006), not being applicable to other populations, such as adolescents. Hence, we detected the need to create a cultural self-efficacy (CSE) measure for adolescent with diverse cultural origin. 1.1. Conceptual delimitation of the construct to be evaluated Perceived self-efficacy is not a comprehensive trait, but rather a set of domain specific self-beliefs linked to differentiated spheres of operation. Hence, scales created under the approach ‘‘the same measurement for everything’’ have usually low

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

303

explicative and predictive value, since most of the items in this case may have little or no relevance for the domain of operation under study (Bandura, 2006b). Developing self-efficacy scales related to specific domains of functioning requires a conceptual analysis of the selected domain that could make possible to specify which aspects of personal efficacy should be measured. To our purpose, we referred to the definition of cultural competence provided by Ang, Van Dyne and Koh (2006), and Earley and Ang (2003). Earley and Ang (2003) defined cultural competence as a set of behaviors and congruent attitudes that allow people to function effectively in intercultural situations. They identified strategy or metacognition, knowledge, motivation and behavior as factors that make up cultural competence and they built a scale to evaluate them. Strategy factor referred to the extent to which people are aware of their intercultural experiences, reflecting processes they use to acquire and understand knowledge and cultural information. Knowledge factor referred to the extent to which people understand cultures are at the same time similar and different. Motivational factor referred to people’s interest in experiencing other cultures and in interacting with people from different cultures. Behavior factor referred to people’s capability of adapting their verbal and non-verbal behavior to different cultural contexts. This requires a flexible repertory of behavioral responses suited to a variety of situations, as well as the ability to modify one’s behavior according to the characteristics of a specific interaction or a particular context. Taking as a reference this concept of cultural competence and Bandura’s (1997) definition of self-efficacy, we defined CSE as the perception of one’s own capability to mobilize motivation, cognitive resources and courses of action necessary in situations characterized by cultural diversity. 1.2. Aims and hypotheses The aim of this study was to construct and assess the psychometric properties of a CSE measurement for adolescents with different cultural origins. In the first stage of this study we proposed the following: a) A review of CSE and social measurements applied in cultural exchange contexts, and the identification of stressful situations that students from a different cultural origin have to deal with in the academic context. Consequently, we generated a broad set of items following Bandura’s recommendations (Bandura, 2006b). b) An analysis of the psychometric properties of the scale, selecting the most suitable items to apply to the population. For this purpose, we administered the scale to a pilot sample – college students – and examined its factorial structure, internal consistency and validity. We went on to the second stage in which we proposed the following: c) An application of the CSE scale to the target sample – adolescents – (cultural self-efficacy scale for adolescents (CSES-A)), in educational environments characterized by cultural diversity, with the aim of analyzing psychometric properties of the scale: internal structure, reliability and validity. We also included different measures that made possible to perform a concurrent criterion validity analysis. A priori we did not propose any hypothesis concerning factorial structure and internal consistency of the scale. Nevertheless, the validity analysis of the instrument was guided by some general hypotheses. Since the scale has been developed after a thorough review of the existing literature and went subsequently under a process of selection of items better tapping the theoretical construct in the first stage of the study, we considered that the scale would have shown a good validity of content. As pertain to the validity of construct, we implemented both an exploratory (EFA) and a confirmatory factor analysis (CFA) to prove the existence of a single underlying psychological construct which gives meaning to the scores on the scale. As regards the concurrent validity, it was analyzed through CSE intercorrelation with indexes assessing closely related construct of adaptation process. CSE was expected to be positively related to the number of people from other cultures with whom contact is maintained (Tsang, 2001), perception of cultural enrichment (Piontkowski et al., 2000) and academic expectations (Hechanova-Alampay et al., 2002), and negatively related to acculturation stress (Hechanova-Alampay et al.). We also expected that the proneness to prefer the ‘‘integration’’ acculturation strategy would be more frequent in those with a high level of CSE (Allard & Landry, 1992; Briones et al., 2005; Piontkowski et al.). Moreover, we included the measurements of expectancies of internal control and general self-efficacy because, in accordance with Fan and Mak (1998), we expected them to be moderately and positively related to CSE, but not with cultural type variables. 2. First stage: construction of the initial scale (50 items) The development of this measurement was based on three main aspects: first, Bandura’s (2006b) guidelines for constructing self-efficacy scales and a deep review of the domain under evaluation, namely cultural competence (Earley and Ang, 2005); second, a review of self-efficacy scales used in cultural exchange contexts (Fan & Mak, 1998); and third, identification of secondary school students’ cultural stressors made by themselves and their teachers (Berry, Kim, Minde, & Mok, 1987; Martines, 2005).

304

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

As previously stated, Bandura (2006b) highlighted as an essential aspect in the construction of self-efficacy scales the identification of specific domains of functioning to be studied. We focused on cultural domain, identified as a framework the concept of cultural competence posed by Earley and Ang (2003) and, consequently, developed items related to people’s perception of their own competence in the four key determinants of this domain: strategy, knowledge, motivation and behavior. The literature review carried out on paper of CSE in the integration of different cultural groups in the school (Daly & Brown, 2004a; Daly & Brown, 2004b; Gong & Fan, 2006; Li & Gasser, 2005) revealed a general accordance in attesting the importance of Fan and Mak’s (1998) social self-efficacy scale. This scale has been constructed selecting four items from the Social Self-Efficacy Scale (Sherer et al., 1982) and eight items adapted from the College Self-Efficacy Instrument (Solberg, O’Brien, Villareal, Kennel, & Davis, 1993). The foster subset reflected aspects related to initiative, capability and persistence in getting to know people and making friends, and the general capability to get on well at social meetings; the latter measure reflected youths’ confidence in eight school situations: participation in class discussions, asking teachers questions in class, talking to professors and staff, getting a date, making new friends and joining a student organization. We recognized the relevance of this scale since it represented the first attempt of incorporating elements related to social interactions between adolescents from different cultures. Fan and Mak (1998), on the basis of previous focus groups, developed eight items specifically related to needs of students who find themselves in a new culture. They included cultural aspects such as: having interests in common with the people in the country of residence, being confident of one’s own linguistic skills, expressing difficulty in holding a conversation with most people, feeling relaxed in social situations, difficulty in expressing a different opinion or asking for information, whether one has topics of conversation in common with the local inhabitants and whether ones enjoy activities of people in the host country. Nevertheless, this scale did not take in account elements identified by students from the majority culture, hence it could only be addressed to groups of immigrant students. Moreover, in accordance with Bandura’s (2006b) guidelines, we identified situations that students from different cultures perceived as problematic in the school context. We, thus, worded items related to different challenges or impediments that students face in cultural interchange, and selected an adequate gradation of difficulties in situations proposed in order to avoid ceiling effects. Taking heed of the strictly cultural elements that give rise to acculturation stress, we considered relevant the classification made by Berry et al. (1987) in five categories: physical stressors (e.g., climate, non-family environment, residence and safety), biological stressors (e.g., food and malaise), social stressors (e.g., homesickness, feeling separated or different from the rest, difficulty in making new friends and loneliness), cultural stressors (e.g., the difference between the cultural values of the society of origin and the one of residence, contact with racial discrimination, etc.), and functional stressors (e.g., language, economic difficulties, transport, etc.). Moreover, Martines (2005) gathered the opinions of teachers about the conflictive situations that students from difference cultures face. These difficulties resulted mainly linked to professional/vocational needs, professional/vocational stressors, aspects concerning ethnic identity, inadequate criteria for assessing the skills of immigrant adolescents’ in their native language and the differences in the language skills of bilingual/bicultural students in the dominant culture and of immigrant students. Considering all the aspects previously presented, 50 items were developed and worded in accordance with the reading level of the participants’ educational stage (Bandura, 2006b). Ambiguity and technical words that does not form part of the adolescents’ daily language were avoided and the items tended to be brief and explicit. As regards the answer format, following Bandura’s (2006b) indications, we present a five-point answer structure, keeping the same structure of other scales and descriptors that we include in our study. The scale was introduced by short and plain instructions that set the appropriate norms for judging efficacy. Thus, students were asked about their operational capabilities at that moment (Bandura, 2006b) using the following indications: ‘‘Speaking to people from a different culture, I can. . .’’, ‘‘If I lived in a different culture, I would be able to. . .’’and ‘‘Approaching a different culture, I can. . .’’. Finally, we carried out a pre-test to verify that items were clear and easy to understand. 3. Analyses of the psychometric properties of the initial scale (50 items) Initially, we administered the scale to a sample with characteristics similar to those of the target population. According to Osterlind (1989), in implementing a pre-test study the sample must include between 50 and 100 participants in order to allow detecting the most problematic items, difficulties in understanding instructions, errors in the format of the instrument, errata, etc. Below we give the characteristics of the method and the results of these psychometric analyses. 3.1. Method 3.1.1. Pre-test sample – college students In this stage, the scale was administered to first-year psychology students who freely decided to collaborate in the study. One hundred sixty-four (87.2% female, 12.8% male) students, ranging in age between 17 and 35 years (mean age = 18.57, SD = 2.16), participated. As regards their cultural origin, 3% were European, 1.8% South American, 93.3% Spanish and 1.8% did not inform about their origin. Concerning people from other cultures with whom they had contact, except for 9.8% that

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

305

affirmed they had no relationship of this type, 52.4% reported having some kind of contact with less than three people, 25.6% with between four and six people, 3% with between seven and nine people and 9.1% with more than 10. 3.1.2. Measurements Socio-demographic information. Students provided information relating to their gender, age and origin. CSES-A. Participants graded their answers to 50 items on a scale from 1, which implies that they do not feel capable of coping with the situation presented, to 5, which shows total confidence in their ability to cope with it. The score was calculated using the mean of the scores. Cultural contact. We adapted and selected the question ‘‘how many people from a different culture do you mix with frequently?’’ from Individual Network of Ethnic Contact (INEC) with friends scale (Landry & Bourhis, 1997; Montreuil & Bourhis, 2004). The students answered using a five-point scale, from 1 (no one), to 5 (more than 10 people). 3.1.3. Procedure It was guaranteed to participants that their answers would be confidential and processed anonymously. There was no time limit for answering the questions; nevertheless, the questionnaire was completed in approximately 15 min. 3.2. Results The metric analysis sought a set of items that would maximize the variance of the test. For this purpose, we selected items with a high power of discrimination, high standard deviation and mean scores situated around the middle of the scale (Nunnally & Bernstein, 1995), and we also considered the conceptual aspects that motivated its creation. We found that the mean scores for the 50 items ranged between 2.28 and 4.43, and the standard deviation ranged between the value of 0.72 and 1.19. 3.2.1. Analysis of the dimensionality of the instrument (internal structure) The dimensionality of the scale was investigated using exploratory factor analysis. In order to facilitate the interpretation of the factor analysis, we followed the recommendations of Fabrigar, Wegener, MacCallum, and Strahan (1999) and performed a principal-axis factor analysis with promax rotation, given the high relation between factors (Nunnally & Bernstein, 1995). Bartlett’s sphericity test, which was equal to x2(1176, N = 164) = 3832.77, p < 0.001, and the Kaiser–Meyer–Olkin index (KMO) with a result of 0.82, guaranteed that the correlation matrix was suitable for exploratory factor analysis. Fourteen factors with eigenvalues greater than one emerged. Together, they explained 58.07% of the variance. Hence, taking as a basis this last analysis, we considered appropriate not to take into account 12 of the items that showed loading lower than 0.30 (Stevens, 1992). 3.2.2. Estimation of reliability In order to calculate reliability, we used the analysis of internal consistency through covariance between items using Cronbach’s a, an index recommended by Bandura (2006b). The reliability coefficient for the 50 items that initially formed the questionnaire was high (a = 0.921). Nevertheless, taking as a basis the analysis of the items, we considered appropriate to carry out the study of reliability with the 38 items selected in the previous section, given that the factors formed an adequate grouping of items. The a for the 38 items was 0.909. Five items were characterized by low loadings in the factor analysis, because their reading was difficult, or because their contribution to the reliability of the measurement was low, hence we decided to eliminate them. Finally, we had 33 items, which obtained an a reliability equal to 0.903. Nevertheless, we decided to perform the same type of factor analysis with these 33 definitive items in order to observe the possible variations in the grouping. Bartlett’s sphericity test was equal to x2(528, N = 164) = 2 677.188 (p < 0.001) and KMO was equal to 0.84. The 14 factors explained 58.13% of the variance. 3.2.3. External validity Results showed that CSES-A (constructed with the 33 items selected) had a strong correlation (0.27, p < 0.01) with the number of people from other cultures with whom there was a relationship. 4. Second stage: analysis of the psychometric properties of the 33 items 4.1. Method 4.1.1. Target Sample – adolescent students Sample selection criteria are as follows. We made reference to statistics published by the Spanish Ministry of Education related to the countries of origin of immigrant students in secondary schools and their distribution in Spain by Autonomous Regions and provinces for academic year 2001–2002, in order to guarantee an adequate cultural differentiation in the sample.

306

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

As pertain to origin of immigrant students, the majority came from South America (44.8%) and Africa (19.1%). In terms of geographical distribution, the greatest number of immigrant students was concentrated in Madrid (20313) and Andalusia (12008). Within Andalusia, the province of Almeria had the highest number of immigrants and the greatest balance in terms of origin of the students (35.81% Africans and 25.98% South Americans). Among all the schools with a percentage of immigrants greater than 50%, we randomly selected three secondary schools in Madrid and three in Almeria. Description of adolescent student sample is as follows. We included 926 (48.1% females, 51.9% males) students from Compulsory Secondary Education, ranging in age between 12 and 18 years (mean age = 14.43; SD = 1.17). Their origin was grouped into five categories: European (5.6%), South American (18.4%), Asian (2.1%), African (18.2%) and Spanish (55.8%). As regards to knowledge of Spanish, 2.2% reported a very low level, 2.5% low, 11.4% intermediate, 21.6% high and 62.2% very high. Given that the questionnaire was administered in Spanish, the subsequent analyses were performed taking into account only the adolescents with a level equal to or above intermediate, which implies a subsample of 868 students. Participants indicated that they had been living in Spain for 10.04 years on average (range: 0.10–18 years; SD = 5.31). As regards people from other cultures with whom they maintain contact, 4.8% affirmed that they had no relationship with people from other cultures, 25.5% reported some type of contact with fewer than three people, 22.4% with between four and six people and 47.2% with more than 7 people. 4.1.2. Measurements Socio-demographic data. Students provided information about their age, gender, country of birth, amount of time spent living in Spain and their level of knowledge of the Spanish language. Participant answered using a five-point Likert scale (1 = very low level of knowledge; 5 = very high level of knowledge). CSES-A. The 33 items selected in the previous stage were administered to participants, who graded their answers using the same five-point Likert scale described above. The score was calculated using the mean of the answers. Cultural contact. Participant answered the question ‘‘how many people from a different culture do you mix with frequently?’’ (INEC, Landry & Bourhis, 1997; Montreuil & Bourhis, 2004) by selecting one of the four options of answer offered: no one, between 1 and 3, from 4 to 6, or more than 7. Acculturation attitudes. They were evaluated by means of two items based on Berry’s acculturation model, used previously in Spain (Rojas, Garcı´a, & Navas, 2003) and in Germany and Switzerland (Piontkowski et al., 2000). One is related to maintain one’s own culture (‘‘Any group of immigrants in Spain should live in this country according to their customs’’) and the other is related to the interest in mixing with the majority group (‘‘Any group of immigrants in Spain should participate fully in the life of this society’’). The students answered these items using a five-point Likert scale (1 = disagreement; 5 = agreement). Combining the answers to these two items, we computed for each participant a variable with four levels corresponding to the prevalent acculturation attitude, namely integration, assimilation, separation and marginalization, as previously defined. Perceived enrichment of other cultures. To evaluate perceived enrichment in every culture included in the study, we considered students’ opinion about the influence of the exogroups on the endogroup (Rojas et al., 2003), through these two items: ‘‘how do you think other cultural groups affect the life of the people originally from this country?’’ and ‘‘how do you think the lives of the immigrants arriving in this country are affected by the Spanish people?’’ Participants rated their answers using a five-point Likert scale (1 = very negatively; 5 = very positively). We obtained two measures, one related to perceived enrichment through the influence of the no-dominant groups on the dominant culture, and the other to perceived enrichment in other cultural groups through contact with the dominant culture. Acculturation stress. We used 23 items selected from the SAFE scale (Mena, Padilla, & Maldonado, 1987). This scale measures acculturation stress in social, attitudinal, family and environmental contexts, as well as perceived discrimination towards the immigrant population. The participants rated the intensity of the stress experienced in each proposed situation on a six-point Likert scale, from 1 (not at all stressful) to 5 (extremely stressful), with 0 indicating that the specific situation has never happened to them. An example of item is: ‘‘It annoys me that some members of my family do not understand my new values’’. The SAFE scale has proved an adequate internal consistency in previous studies (Hovey, 2000). In our study, the reliability coefficient for the scale was a = 0.93. Following the indications of authors, a variable was created summing the scores across items. Academic expectations. Participants answered five items that reflected the subsequent educational levels which they were supposed to reach in their future. They rated the perceived probability of passing these levels using a five-point Likert scale (1 = definitely no; 5 = definitely yes). Reliability coefficient for the scale was a = 0.79. The variable was computed as mean of items score. Internal control expectancies. This measurement comprises four items of the Battery of measures of generalized control expectancy (BEEGC; Palenzuela, Prieto, Almeida, & Barros, 1997) corresponding to the measurement of the expectancies of internal control. Students rated their answers using a five-point Likert scale (1 = no; 5 = definitely yes). Reliability coefficient for the scale was a = 0.64. The variable was computed as mean of items score. General self-efficacy. It was measured by three items selected by BEEGC (Palenzuela et al., 1997). Students answered using the same 5-point scale previously presented. Cronbach’s a was equal to 0.66. The variable was computed as mean of items score.

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

307

4.1.3. Procedure Two trained psychologists blind to the aims of this study administered the questionnaires, after receiving parents’ authorization. Participants completed questionnaires in classrooms during school time. In order to encourage sincerity in answering (Bandura, 2006b), students were make aware about the importance of their contribution to increase the comprehension of and guide the development of programs designed to help people to cope with situations they have to face in daily life. There was not time limit to answer the questionnaires and the confidentiality of the information provided was guaranteed. 4.2. Results Descriptive analysis shown a good approximation of the data to the normal curve, for CSES-A (mean skewness and mean kurtosis in absolute value were respectively 0.30 and 0.90). The means of the items ranged from 2.54 (for item 16) to 3.93 (item 31). The standard deviation was higher than 1 for all items, ranging between 1.06 (item 21) and 1.42 (item 31). 4.2.1. Analysis of the dimensionality of the instrument (internal structure) In order to analyze internal consistency we calculated the corrected correlation between the score of the item and the total CSES-A scale. Coefficients ranged between 0.417 (item 32) and 0.737 (item 24) and were considered adequate since they were greater than 0.30 (Nunnally & Bernstein, 1995). Once again we followed the recommendations of Fabrigar et al. (1999) in the exploratory factor analysis and performed a principal-axis factor analysis, using the promax oblique method, given the high relationship among the factors (Nunnally & Bernstein, 1995). Bartlett’s sphericity test was equal to x2(528, N = 868) = 14 243.577 (p < 0.001) and KMO index was 0.955, indicating that correlation matrix was adequate. Table 1 Loading of the items, corrected correlations and reliability of the factors. Corrected item-total correlation

a if item deleted

1. CSE in mixing satisfactorily with other cultures (a = 0.914) 34 0.779 35 0.745 26 0.725 24 0.724 21 0.700 25 0.693 16 0.677 48 0.540 47 0.529 20 0.499 10 0.316

0.706 0.668 0.723 0.759 0.696 0.751 0.583 0.654 0.695 0.583 0.548

0.904 0.906 0.904 0.902 0.905 0.902 0.911 0.907 0.905 0.911 0.913

2. CSE in understanding other ways of life (a = 0.863) 39 0.810 38 0.757 37 0.738 40 0.575 46 0.530 49 0.483 41 0.472 42 0.386

0.655 0.674 0.528 0.651 0.639 0.589 0.608 0.565

0.839 0.857 0.843 0.844 0.849 0.847 0.852 0.839

3. CSE in coping with homesickness and separation (a = 0.854) 30 0.905 31 0.895 29 0.806 32 0.649 14 0.395 23 0.354

0.747 0.735 0.763 0.600 0.513 0.488

0.809 0.811 0.805 0.837 0.852 0.855

4. CSE in processing information about other cultures (a = 0.837) 2 0.694 5 0.683 6 0.624 7 0.578 1 0.547

0.663 0.664 0.638 0.644 0.589

0.798 0.798 0.805 0.804 0.818

5. CSE in learning and understanding another language (a = 0.909) 45 0.908 44 0.903 43 0.857

0.829 0.830 0.797

0.861 0.860 0.889

Items

Loadings

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

308

Considering eigenvalues greater than 1, we suggested a five-factor solution, that seemed adequate even considering the analysis of the reproduced correlation matrix, in which only few residuals (1%) were greater than 0.10 (McDonald, 1985). The five factors explain 54.82% of the variance of the scale. In Table 1 the principal loadings of each item on the factors are included. Besides items 47 and 48 which had strong loading also on a second factor (0.365 and 0.397, respectively), the rest of the items did not show loading greater than 0.30 on more than one factor. Moreover, in the same table, we included the corrected correlation for each item with its factor and Cronbach’s a if each of these elements is eliminated in the factor. These five factors were defined as follows: (a) self-efficacy in mixing satisfactorily with other cultures (eigenvalue = 13.339), (b) self-efficacy in understanding other ways of life (eigenvalue = 2.523), (c) self-efficacy in coping with homesickness and separation (eigenvalue = 1.551), (d) self-efficacy in processing information about other cultures (eigenvalue = 1.528) and (e) self-efficacy in learning and understanding a foreign language (eigenvalue = 1.189). 4.2.2. Estimation of reliability Cronbach‘s a for the entire scale was 0.951. Furthermore, we performed a reliability analysis for each of the five factors obtained in the factor analysis. As can be seen in Table 1, reliability coefficients ranged from 0.837 to 0.914. Considering factor analysis results, corrected correlations of the items with the entire scale and the reliability analysis, we decided to drop items 10, 14, 23, 42, 46, 47, 48 and 49. The reduced 25-item scale maintained adequate reliability both for the entire scale (a = 0.935, for the 25 items) and for each factor. In particular, we replicated the reliability analysis for the first three factors, highlighting adequate values (respectively for factor 1, eight items, a = 0.896; for factor 2, five items, a = 0.808; for factor 3, four items, a = 0.873). 4.2.3. Exploratory and confirmatory factor analysis with 25 items The construct validity of the reduced 25-item scale was analyzed by EFA and CFA. The full sample of participants was randomly split into two halves to increase stability of our results. An EFA was implemented on the first subsample in order to verify the adequacy of the five-factor solution. In Table 2 we reported the descriptive statistics for the items. As can be seen, a great number of items showed a value for skewness and kurtosis above the absolute value of 1, attesting a non-normal distribution. In these cases, as suggested by West, Finch, and Curran (1995), it is not recommended to use a normal distribution-based estimator. Hence we implemented an EFA, using as a method of esteem Unweighted Least Squares (ULS) which is robust for no-normal distribution of data. Table 2 Means, standard deviations, skewness, kurtosis, corrected item-total correlation, a if item deleted, and a by factors in first (N = 433) and second subsample (N = 434), and factor loadings of EFA, performed on the first subsample. Items

Mean (1/2)

SD (1/2)

Loading (1)

Skewness (1/2)

Kurtosis (1/2)

Corrected item-total correlation (1/2)

a if item deleted (1/2)

with other cultures (a1 = 0.960, a2 = 0.959) 1.67/1.76 0.726 1.33/1.34 1.68/1.89 0.686 1.18/1.29 1.63/1.74 0.638 1.24/1.14 1.74/1.83 0.620 1.42/1.38 1.61/1.74 0.607 1.08/0.97 1.56/1.63 0.590 1.20/1.28 1.59/1.65 0.560 0.81/1.06 1.69/1.81 0.368 1.32/1.35

2.99/2.34 2.43/1.77 2.75/2.00 2.61/2.12 2.52/1.81 2.85/2.54 2.21/2.17 2.78/2.33

0.876/0.875 0.833/0.821 0.846/0.883 0.809/0.852 0.822/0.852 0.827/0.861 0.779/0.787 0.791/0.819

0.954/0.951 0.955/0.955 0.955/0.951 0.956/0.953 0.955/0.953 0.955/0.952 0.957/0.957 0.957/0.955

2. CSE in coping with homesickness and separation (a1 = 0.932, a2 = 0.940) 30 3.11/3.18 2.05/2.08 0.919 1.43/1.45 31 2.98/3.01 2.10/2.17 0.909 1.35/1.46 29 3.31/3.48 2.03/2.10 0.804 1.18/1.27 32 3.22/3.26 2.00/2.07 0.729 1.30/1.29

1.96/1.91 1.66/1.79 1.36/1.43 1.79/1.57

0.852/0.896 0.854/0.890 0.844/0.854 0.820/0.789

0.912/0.909 0.912/0.911 0.913/0.922 0.916/0.942

3. CSE in learning and understanding another language (a1 = 0.925, a2 = 0.952) 44 3.72/3.75 1.75/1.93 0.875 1.15/1.24 43 3.40/3.40 1.82/2.05 0.862 1.21/1.34 45 3.56/3.57 1.94/2.10 0.820 1.29/1.36

2.03/1.65 2.03/1.65 1.78/1.47

0.875/0.913 0.891/0.902 0.801/0.883

0.887/0.920 0.880/0.926 0.913/0.942

4. CSE in processing information about other cultures (a1 = .920, a2 = .943) 2 3.74/3.81 1.59/1.72 0.710 1.16/1.42 6 3.84/3.88 1.54/1.72 0.658 1.21/1.24 1 3.42/3.52 1.55/1.77 0.634 1.54/1.68 7 3.78/3.89 1.59/1.78 0.630 1.00/1.07 5 3.76/3.81 1.55/1.74 0.599 1.10/1.25

2.83/2.71 3.11/2.43 3.79/3.10 2.52/1.79 2.72/2.38

0.786/0.872 0.801/0.840 0.815/0.827 0.800/0.839 0.766/0.850

0.904/0.925 0.901/0.931 0.898/0.933 0.901/0.931 0.908/0.929

5. CSE in understanding other ways of life (a1 = 0.948, a2 = 0.942) 40 3.82/3.86 1.67/1.86 0.670 39 3.69/3.59 1.66/1.92 0.668 38 3.47/3.44 1.74/1.99 0.583 41 3.76/3.82 1.71/1.94 0.515 37 3.03/3.07 1.89/2.11 0.511

2.66/1.94 2.93/2.26 2.59/1.97 2.27/1.56 2.63/1.96

0.831/0.845 0.857/0.867 0.861/0.876 0.806/0.803 0.778/0.830

0.939/0.929 0.937/0.924 0.936/0.923 0.941/0.936 0.944/0.936

1. CSE in mixing satisfactorily 24 3.78/3.84 26 3.91/3.93 34 3.92/4.07 25 3.83/3.87 35 4.13/4.17 21 4.18/4.18 16 4.18/4.27 20 3.73/3.69

1.27/1.30 1.37/1.48 1.31/1.39 1.12/1.16 1.49/1.49

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

309

The five-factor solution explained 58.96% of the total variance. The factor loadings shown in Table 2 attested the identification of the following factors: a) self-efficacy in mixing satisfactorily with other cultures (eigenvalue = 10.437), (b) self-efficacy in coping with homesickness and separation (eigenvalue = 2.298), (c) self-efficacy in learning and understanding a foreign language (eigenvalue = 1.547), (d) self-efficacy in processing information about other cultures (eigenvalue = 1.297) and (e) self-efficacy in understanding other ways of life (eigenvalue = 1.055). The second subsample was used to verify the former solution, implementing a CFA model in which five first-order factors and one second-order factor, defined CSE, were posited. Each first-order factor was measured by items identified in the EFA and no cross-loadings were allowed. In order to evaluate the solution we take into account different goodness of fit indexes: x2, root mean square error of approximation (RMSEA), root mean square residual (RMSR), comparative fit index (CFI) and the Tucker–Lewis index (TLI). Since CFI and TLI values above 0.90 (Bentler, 1995; Byrne, 1994), RMSEA values below or equal to 0.06 and SRMR values equal to or below 0.09 (Hu & Bentler, 1999) are considered adequate, fit indexes obtained (x2(269, N = 234) = 617.673, RMSEA = 0.056 (0.050–0.062), SRMR = 0.050, CFI = 0.94 and TLI = 93) showed that the five-factor model fit well the data. The significant value identified for the x2 index could depend on the sample size and could be discarded. A schematic representation of the five-factor solution is presented in Fig. 1. 4.2.4. Obtaining external evidence of validity To study the concurrent validity, we performed a correlation analysis between CSE and its five factors and cultural contact, perception of cultural enrichment, acculturation stress and academic expectations. We also carried out a x2 analysis with preference for acculturation strategies. As can be seen in Table 3, the positive and significant pattern of correlation of CSE and its five factors with cultural contact, perception of cultural enrichment and academic expectations was coherent with our expectation and attested for concurrent criterion validity. The relation that CSE and its factors showed with acculturation stress was not in line with our hypothesis. Nevertheless, performing this analysis splitted by cultural groups, we found out that even though the relation resulted no-significant, it tends to be negative for Latin American, Asian and African groups of immigrant students. Furthermore, cultural variables – cultural contact and perception of cultural enrichment – had a no significant relation with expectancy of internal control and general self-efficacy, despite the fact that they showed a positive and significant relation with CSE and its five factors. We consider this result as a further indication of concurrent criterion validity of the CSEA scale, since the domain-specificity (cultural competence) of the scale guaranteed the association with variables that it is intended to explain besides the expectancies of internal control and general self-efficacy. Finally, we carried out a x2 test with the preference for acculturation strategies, after dichotomizing the CSE variable. Given that self-efficacy threshold must be determined empirically (Bandura, 2006b), we considered the level in CSE to be low when the mean score on the scale was below 20% and high when it was above 80%. The result of the analysis was significant

Fig. 1. Preference for acculturation strategies according to CSE level (25 items).

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

310

Table 3 Correlations, means and standard deviations of stage 2 sample.

1. CSE (25 items) 2. Mix 3. Understand 4. Cope 5. Process 6. Language 7. Cultural contact 8. Perceived enrichment 1 9. Perceived enrichment 2 10. Acculturation stress 11. Academic expectations 12. Control expectancies 13. General self-efficacy Mean Standard deviation

1

2

3

4

5

6

7

8

9

10

11

12

13

0.889 0.800 0.603 0.802 0.708 0.234 0.349 0.169 0.163 0.317 0.172 0.298 3.327 0.761

0.662 0.395 0.663 0.532 0.241 0.325 0.163 – 0.278 0.209 0.310 3.650 0.865

0.311 0.565 0.509 0.153 0.288 0.131 0.070* 0.290 0.182 0.219 3.187 0.930

0.312 0.296 – 0.161 0.088* 0.289 0.101 – 0.118 2.760 1.181

0.510 0.270 0.315 0.147 0.112 0.278 0.206 0.263 3.468 0.903

0.147 0.196 0.077* 0.165 0.260 0.071* 0.196 3.162 1.187

0.089* – – – – – 3.121 0.952

0.366 0.070* 0.159 – – 4.178 1.184

–0.097 0.137 0.092* – 4.052 1.263

–0.199 –0.241 0.121 23.105 21.398

0.260 0.351 3.729 0.855

0.477 4.185 0.629

4.189 0.636

Note: These correlations were significant at p < 0.01 or p < 0.05 (*).

(x2(3, N = 342) = 39.228, p < 0.001) and the significant differences (p < 0.01) were located in the ‘‘integration’’, ‘‘assimilation’’ and ‘‘marginalization’’ strategies. Students with a high level of CSE showed greater preference for the ‘‘integration’’ strategy (76% of students with high CSE and 44% of students with low CSE), whereas those with a low level showed a greater preference for the ‘‘assimilation’’ (11 and 29%, respectively) and ‘‘marginalization’’ strategies (5 and 18% respectively). 5. Discussion and conclusions The main aim of the study was to construct and validate a CSES for adolescents with diverse cultural origins. Previously, we extensively reviewed the literature, analyzing the existing measures of social self-efficacy, the concept of cultural competency and the stressors for groups involved in an acculturation process. Then, following Bandura’s (2006b) recommendations, we developed an initial set of 50 items to be included in a CSE measure. After analyzing results derived by both a pilot study and a larger study including a target population of students with different cultural origins, we were able to select 25 items forming the final version of our scale. A CFA highlighted five first-order factors, namely self-efficacy in mixing satisfactorily with other cultures, in coping with homesickness and separation, in learning and understanding a foreign language, self-efficacy in processing information about other cultures, and in understanding other ways of life, tapping a second-order factor, defined as CSE. The scale showed high internal consistency and content, construct and criterion validity. The pattern of correlations of CSE and of the five first-order factors with cultural contact, perception of cultural enrichment and academic expectations, as well as the result deriving from the x2 analysis with the acculturation strategies, supported our hypotheses on criterion validity. Hence, as other authors have sustained (Allard & Landry, 1992; Black & Gregersen, 1991; Fan & Mak, 1998; Harrison et al., 1996; Piontkowski et al., 2000; Tsang, 2001), the higher the level of CSE, the greater the cultural contact and the perception of cultural enrichment, the higher the academic expectations and the greater the preference they showed for an integration strategy; in short, the better adolescents’ adaptation to their academic and social context. On the other hand, the expectancies of internal control and general self-efficacy were positively related to CSE, as we expected on the basis of previous studies (Fan & Mak, 1998), but, differently from CSE, they had no relation with cultural type variables. This result in particular pointed to the relevant domain specificity of the proposed scale. However, the expected negative relation of CSE with acculturation stress was not supported by results. According to Hechanova-Alampay et al. (2002), there may be different reactions to adaptation and these may take different route in time. That is, consistently with previous findings that showed how the students’ psychological emotions rose and fell according to the academic calendar (Golden, 1973), we could hypothesize that stress levels shown by students in this study could be due to the specific academic period in which our research was conducted, rather than to the experience of being a foreign student or to cultural interaction with other students. Notwithstanding, this could be one of the limitations of this study and we suggest to further investigate the relation between stress and CSE in a longitudinal perspective. As well, we recommend caution in considering scale validity since concurrent criterion validity was only tested through the analysis of correlations with self-report measures. In particular, could be interesting to carry out studies to further investigate validity of the CSE scale and its predictive power with respect to academic and social adaptation of students. The CSES-A was created in order to guarantee multiple applications. On the one hand, the process of construction of the scale and, in particular, the great attention paid to the wording of the items may guarantee its usefulness and availability with adolescents of diverse cultural origin, nationality and residence. Moreover, the possibility of

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

311

applying the entire scale or only subsets of items tapping specific factors separately can be of great interest in some studies. On the other hand, as Fan and Mak (1998) stated, students’ CSE can contribute to the prediction of their academic results and overall of their social adaptation. Hence, the scale could be useful to identify adolescents’ level of CSE in each of its facets, allowing the definition and the design of specific intervention programs (Mak & Tran, 2001), following Bandura’s (1986, 1997) indication about strategies for increasing self-efficacy. Following Li and Gasser (2005) CSE of all students, both those from minority cultures and those native, can be increased through their active participation in monitored cross-cultural social interactions, by observing peer behavior in social contexts, requesting feedback and strengthening their own performance, and by focusing their attention on their own behavior, more than on their emotional arousal during social interactions. We consider that these interactions would favor not only students’ CSE but also their social adaptation. Appendix A. Cultural self-efficacy scale for adolescents (25 items)1 (item numbers refer to the original 50-item scale) In the situations posed below, mark to what extent you feel capable of carrying them out using the options given. 1

2

Can not do at all

3

4

Moderately certain can do

5 Certain can do

Speaking to people from a different culture, I can. . . 1.2.5.6.7.-

Realize what I know about that culture. Use the information I have on that culture to understand people from that culture. Understand what I am being told. Make myself understood when speaking to people from a different culture. Maintain a conversation when most of the people are from a different culture.

If I lived in a different culture, I would be able to. . . 16.20.21.24.25.26.29.30.31.32.34.35.-

Make new friends. Ask information on terms related to that culture. Mix with classmates from a different culture form mine. Take part in social activities of the people of that culture. Create topics of conversation with people from that culture. Enjoy the activities of the people of that culture. Overcome homesickness. Overcome nostalgia for my friends. Overcome nostalgia for my family. Overcome loneliness. Work in groups of boys from different countries. Work in groups of girls from different countries.

1

2

Can not do at all

3

4

Moderately certain can do

Approaching a different culture, I can. . . 37.38.39.40.41.43.44.45.-

1

Understand other religious beliefs. Understand another type of family different from mine. Understand how couples relate in a different culture. Understand the art of a different culture. Understand the music of a different culture. Speak a language different from mine. Learn a language different from mine. Understand a language different from mine.

The scale was developed and administrated in Spanish. This English version represents a translation of the original scale.

5 Certain can do

312

E. Briones et al. / International Journal of Intercultural Relations 33 (2009) 301–312

References Allard, R., & Landry, R. (1992). Ehtnolinguistic vitality beliefs and language maintenance and loss. In W. Fase, K. Jaspert, & S. Kroon (Eds.), Maintenance and loss of minority languages (pp. 171–195). Amsterdam: John Benjamins. Ang, S., Van Dyne, L., & Koh, C. K. S. (2006). Personality correlates of the four factor model of cultural intelligence. Group and Organization Management, 31, 100–123. Bandura, A. (1986). Social foundations of thought and action: A social cognitive. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1992). Exercise of personal agency through the self-efficacy mechanism. In R. Schwarzer (Ed.), Self-efficacy: Thought control of action (pp. 3–38). Washington, DC: Hemisphere. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman. Bandura, A. (2006a). Adolescent development from an agentic perspective. In Pajares, F., & Urdan, T. (Eds.), Self-efficacy beliefs of adolescents. Vol. 5 (pp.1–43). Greenwich, CT: Information Age Publishing Bandura, A. (2006b). Guide to the construction of self-efficacy scales. In Pajares, F., & Urdan, T. (Eds.), Self-efficacy beliefs of adolescents. Vol. 5 (pp.307–337). Greenwich, CT: Information Age Publishing Bentler, P. M. (1995). Structural equations program manual. Encino, California: Multivariate Software Inc. Berry, J. W. (1992). Acculturation and adaptation in a new society. International Migration, 30, 69–84. Berry, J. W., & Kim, U. (1988). Acculturation and mental health. In P. Dasen, J. W. Berry, & N. Sartorius (Eds.), Health and cross-cultural psychology (pp. 207–236). London: Sage. Berry, J. W., Kim, U., Minde, T., & Mok, D. (1987). Comparative studies of acculturative stress. International Migration Review, 21, 491–511. Berry, J. W., Kim, U., Power, S., Yong, M., & Bujaki, M. (1989). Acculturation attitudes in plural societies. Applied Psychology: An International Review, 38, 185–206. Black, J. S., & Gregersen, H. B. (1991). Antecedents to cross-cultural adjustment for expatriates in Pacific Rim assignment. Human Relations, 44, 497–515. Briones, E., Tabernero, C., & Arenas, A. (2005). Variables psicosociales relacionadas con el proceso de integracio´n social de los estudiantes inmigrantes. Cultura y Educacio´n, 17(4), 337–347. Byrne, B. M. (1994). Structural equation modeling with EQS and EQS/Windows: basic concepts, applications and programming. Newbury Park: Sage. Church, A. (1982). Sojourner adjustment. Psychological Bulletin, 91, 540–572. Daly, A. J., & Brown, J. C. (2004a). International and domestic students’ international competencies and co- and cross-ethnic interactions. Paper presented at the 4th Annual Hawaii International Conference on Business. Daly, A., & Brown, J. (2004). Enhancing classroom success for international students through the ExcelL Intercultural Skills Program. School of Business, Christchurch College of Education. Earley, P. C., & Ang, S. (2003). Cultural intelligence: Individual interactions across cultures. Palo Alto: Stanford University Press. Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299. Fan, C., & Mak, A. S. (1998). Measuring social self-efficacy in a culturally diverse student population. Social Behavior and Personality, 26(2), 131–144. Golden, J. S. (1973). Student adjustment abroad. A psychiatrist’s view. International Educational and Cultural Exchange, 8, 28–36. Gong, Y., & Fan, J. (2006). Longitudinal examination of the role of goal orientation in cross-cultural adjustment. Journal of Applied Psychology, 91(1), 176–184. Harrison, J. K., Chadwick, M., & Scales, M. (1996). The relationship between cross-cultural adjustment and the personality variables of self-efficacy and selfmonitoring. International Journal of Intercultural Relations, 20, 167–188. Hechanova-Alampay, R., Beehr, T. A., Christiansen, N. D., & Van-Horn, R. K. (2002). Adjustment and strain among domestic and international student sojourners: A longitudinal study. School Psychology International, 23, 458–474. Hovey, J. D. (2000). Psychosocial predictors of acculturative stress in Mexican immigrant. The Journal of Psychology, 134(5), 490–502. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1–55. Landry, R., & Bourhis, R. Y. (1997). Linguistic landscape and ethnolinguistic vitality: An empirical study. Journal of Language and Social Psychology, 16, 23–49. Li, A., & Gasser, M. B. (2005). Predicting Asian international students’ sociocultural adjustment: A test of two mediation models. International Journal of Intercultural Relations, 29, 561–576. Mak, A. S., & Tran, C. (2001). Big five personality and cultural relocation factors in Vietnamese Australian students’ intercultural social self-efficacy. International Journal of Intercultural Relations, 25, 181–201. Martines, D. (2005). Teacher perceptions of multicultural issues in school settings. The Qualitative Report, 10(1), 1–20. McDonald, R. P. (1985). Comments on D. J. Bartholomew, Foundations of factor analysis: Some practical implications. British Journal of Mathematical and Statistical Psychology, 38, 134–137. MEC, Oficina estadı´stica (2007). Datos y cifras. Curso escolar 2007/2008. Secretarı´a General Te´cnica. Mena, F. J., Padilla, A. M., & Maldonado, M. (1987). Acculturative stress and specific coping strategies among immigrant and later generation collage students. Hispanic Journal of Behavioral Sciences, 9, 207–225. Montreuil, A., & Bourhis, R. Y. (2004). Acculturation orientations of competing host communities toward valued and devalued immigrants International. International Journal of Intercultural Relations, 28, 507–532. Nunnally, J. C., & Bernstein, I. J. (1995). Teorı´a psicome´trica. Madrid: McGraw-Hill. Osterlind, S. J. (1989). Constructing test items. Londres: Kluwer Academic Publishers. OCDE (2006). Donne´es sur les migrations internationales 2006. http://www.oecd.org/dataoecd/20/33/38064695.xls].(3th Dec 2007). Palenzuela, D. L., Prieto, G., Almeida, L. S., & Barros, A. M. (1997). Una versio´n espan˜ola de una baterı´a de escalas de expectativas generalizadas de control (BEEGC). Revista Portuguesa de Educac¸a˜o, 10, 75–96. Piontkowski, U., Florack, A., Hoelker, P., & Obdrza´lek, P. (2000). Predicting acculturation attitudes of dominant and non-dominant groups. International Journal of Intercultural Relations, 24, 1–26. Rojas, A. J., Garcı´a, M. C., & Navas, M. (2003). Test de sesgo endogrupal intere´tnico: Estudios de fiabilidad y evidencias de validez. Psicothema, 15(1), 101–108. Sherer, M., Madoux, J. E., Mercandante, B., Prentice-Dunn, S., Jacobs, B., & Rogers, R. W. (1982). The self-efficacy scale: Construction and validation. Psychological Reports, 51, 663–671. Solberg, V. C., O’Brien, K., Villareal, P., Kennel, R., & Davis, B. (1993). Self-efficacy and Hispanic college students: Validation of the College Self-Efficacy Instrument. Hispanic Journal of Behavioral Sciences, 15(1), 80–95. Stevens, J. (1992). Applied multivariate statistics for the social sciences. Hillsdale, NJ: Lawrence Erlbaum. Tsang, E. W. K. (2001). Adjustment of mainland Chinese academics and students to Singapore. International Journal of Intercultural Relations, 25, 347–372. Vargas, J. A., Molino, J. L., Shellman, J., Cantero, M. L., & Bernal, H. (2006). The level of cultural self-efficacy among a sample of Spanish nurses in southeastern Spain. Journal of Transcultural Nursing, 17(2), 164–170. West, S. G., Finch, J. F., & Curran, P. J. (1995). Structural equation models with non-normal variables. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues and applications. (pp. 56–75). Thousand Oaks, CA, USA: Sage.

. .

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.