Puerto Rican And U.S. Most instruments designed to measure acculturation have relied on specific cultural behaviors and preferences as primary indicators of acculturation. In contrast, feelings of belonging and emotional attachment to cultural communities have not been widely used. The Psychological Acculturation Scale (PAS) was developed to assess acculturation from a phenomenological perspective, with items pertaining to the individual’s sense of psychological attachment to and belonging within the Anglo-American and Latino/Hispanic cultures. Responses from samples of bilingual individuals and Puerto Rican adolescents and adults are used to establish a high degree of measurement equivalence across the Spanish and English versions of the scale along with high levels of internal consistency and construct validity. The usefulness of the PAS and the importance of studying acculturation from a phenomenological perspective are discussed.
Psychological acculturation refers to changes in individuals’ psychocultural orientations that develop through involvement and interaction within new cultural systems. Rather than conceptualizing acculturation as a process in which people lose connection to their original culture (Gordon, 1978), new research has emphasized the individual’s negotiation of two cultural entities (Berry, Poortinga, Segall, & Dasen, 1992; Buriel, 1993). Responding to distinct sets of norms from the culture of origin and the host culture, acculturating individuals emerge with their own interpretation of appropriate values, customs, and practices as they negotiate between cultural contexts (Berry, 1980). People vary greatly in their abilities to function within new cultural environments (LaFromboise, Coleman, & Gerton, 1993) and may seek different levels of attachment to and involvement in a host culture or their culture(s) of origin (Padilla, 1980). To study individuals’ cultural orientations, measures of acculturation traditionally have focused on individuals’ behaviors and behavioral preferences and have relied heavily on language use and other behaviors as indicators of acculturation (Marin, Sabogal, VanOss Matin, Otero-Sabogal, & Perez-Stable, 1987; Szapocznik, Kurtines, & Fernandez, 1980). For example, Szapocznik et al. (1980) described acculturation as based in two primary dimensions: cultural behaviors and values. Paralleling their conceptualization of acculturation, the Behavioral Acculturation Scale (Szapocznik, Scopetta, Kurtines, & Aranalde, 1978) includes items most closely related to cultural behaviors and preferences (e.g., What language do you speak at home? and What language do you prefer to speak?).
Similarly, Cuellar, Harris, and Jasso (1980) measured acculturation with items pertaining primarily to cultural behaviors and values (e.g., What language do you prefer?). This measure also included several items concerning migration history (e.g., Where were you raised?) and one item concerning ethnic self-identification (i.e., How do you identify yourself?). These factors can be important in interpreting individuals’ acculturation experiences; however, rather than assessing personal acculturation factors and sociodemographic factors as separate concepts, Cuellar et al. (1980) combined these items within the same measure. We feel that this approach may be problematic in two primary ways. First, such modes of measurement blur distinctions between factual histories of individuals (e.g., age of arrival on the U.S.
mainland) and the assessment of individuals’ acculturative change. Second, measures heavily based on cultural behaviors may not assess adequately individuals’ acceptance and understanding of the values from each culture (Betancourt & Lopez, 1993; Rogler, 1994) or grant sufficient attention to individuals’ emotional attachments to each culture (Estrada, 1993). Alternatively, new instruments can be designed to measure acculturation as it is psychologically experienced by the individual. Reviews of the acculturation literature have identified cultural loyalty, solidarity, identification, and comprehension as overlapping elements of psychological responses to cultural exposure (Berry, 1980; Betancourt & Lopez, 1993; Szapocznik & Kurtines, 1980). To assess these psychological components of acculturation, the 10-item Psychological Acculturation Scale (PAS) was developed. Unlike traditional measures, the PAS targets individuals’ psychological negotiation of two cultural entities (in this case, Anglo-American culture and Latino/Hispanic culture), with particular attention to their sense of emotional attachment to and understanding of each culture. This set of studies was designed to assess the psychometric properties of the PAS.
In particular, cross-language equivalence, internal consistency, and convergent and discriminant validity were examined. CROSS-LANGUAGE EQUIVALENCE Back translation and decentering are widely used methods for determining cross-language equivalence of a scale (Brislin, 1986). For example, to create a Spanish version of an English-language measure, one person translates from English to Spanish, and a different person translates the Spanish version back into English. Discrepancies in the translated versions are resolved through decentering, a process of several iterations whereby the measure is pulled away from the idiosyncrasies of the source language (i.e., the original English-language version). We share the concerns of Bontempo (1993) and Olmedo (1981) about the validity of this accepted procedure.
Even when original and back-translated versions are quite similar, measurement equivalence can still not be assumed or guaranteed for the two language versions because concepts and wordings for scale items originally were produced in only the source language (Bontempo, 1993; Olmedo, 1981). As an alternative, we have developed a dual-focus approach to creating bilingual measures, whereby the conceptual content of each item is developed and then words are generated to express that concept in each language (see Erkut, Alarc6n, Garcia Coil, Tropp, & Vazquez, in press, for details of this procedure). In developing the PAS, our goal has been to compose item wordings that express the relevant concepts with equal clarity, affect, and level of usage in both languages. CONVERGENT AND DISCRIMINANT VALIDITY Convergent and discriminant validity were assessed by examining the predicted relationships between respondents’ psychological acculturation scores and traditional validation measures of acculturation (e.g., place of birth, percentage of lifetime living on the U.S. mainland) as well as culture-specific behaviors and preferences that have been employed in other acculturation scales. Paralleling the results from previous studies of acculturation (e.g., Matin et al., 1987; Szapocznik et al., 1978; Triandis, Kashima, Hui, Lisansky, & Matin, 1982), we expected psychological acculturation scores to be higher among respondents with greater exposure to the new culture (i.e., Anglo-American culture) and greater exposure to English during childhood. Similarly, we predicted that respondents’ language preferences for completing the questionnaires would be associated with their psychological acculturation scores, such that those who chose the Spanish version would tend to have lower psychological acculturation scores than those who chose the English version.
Finally, we also predicted that psychological acculturation scores would be better predictors of individuals’ cultural behaviors and preferences than would their degree of exposure to the new culture. Three studies were conducted to document the psychometric properties of the PAS. Study 1 The first study was designed to examine internal consistency and cross-language equivalence with respect to respondents’ scores on the PAS. Method SAMPLE AND PROCEDURES Respondents were recruited through community centers and neighborhood contacts in several districts within the greater Boston area. Respondents received $10 for their participation, which consisted of completing a questionnaire.
Participants in this study were 36 self-identified bilingual Latinos (10 men and 26 women). Respondents’ ages ranged from 13 to 58 years (M = 28.6 years). Of the respondents, 13 were born on the mainland of the United States and all others were born in Puerto Rico, Mexico, or other Spanish-speaking countries in Central and South America. Percentage of lifetime in the United States was calculated by dividing the number of years living in the U.S. mainland by the age of the respondent (an index previously used in research by Marin et al.
 and Triandis et al. ). Respondents’ percentage of lifetime in the United States ranged from 4% to 100% (M = 75.2%). All respondents responded to both Spanish and English versions of the questionnaire. Spanish and English versions were presented to each respondent in a random order.
MEASURE Psychological Acculturation Scale. The PAS consists of 10 items concerning individuals’ psychological responses to differing cultural contexts (see Table 1). Item wordings for the PAS were generated simultaneously in Spanish and English by a team of bilingual, bicultural, and monocultural researchers. No items were included in the scale which could not be directly and easily expressed with parallel wording in both languages. Subsequently, all potential items were discussed in focus groups of Spanish/ English bilingual adolescents and adults drawn in the greater Boston area.
Items were continuously reworded, as suggested by feedback from successive focus groups and discussions among members of the research team. Altogether, six focus groups were conducted, at which time both focus group participants and research team members were satisfied with item wordings and felt no further revisions were necessary. A readability analysis was conducted for items on the English version of the PAS, using the Microsoft Word 5.0 grammar program (no Spanish grammar program was available). The Flesch estimate of reading ease (74.7%) indicated that the English version of the PAS is fairly readable, corresponding with a Grade 6 to 7 reading level. Item responses for the PAS were scored on a 9-point Likert-type scale, ranging from 1 (only Hispanic/Latino) to 9 (only Anglo/American), with a bicultural orientation defining its midpoint.
Thus, a bicultural orientation (equally Hispanic/Latino and Anglo/American) could be defined as a parallel sense of connection to both cultures (Cuellar et al., 1980). In addition, items regarding migration history, language use, and other demographic variables were included in the questionnaires distributed to each respondent. Results CROSS-LANGUAGE EQUIVALENCE On a 9-point scale, mean PAS scores were 4.37 (SD = .86) and 4.42 (SD -1.06) for the Spanish and English versions, respectively. Means and standard deviations for the Spanish and English versions of scale items are provided in Table 1. Mean item scores were nearly identical for each language version, showing a high degree of consistency in respondents’ scores across the Spanish and English versions.
The correlation between individuals’ total PAS scores from the Spanish and English versions was also extremely high, r(35) = .94, suggesting a high degree of cross-language measurement equivalence. Individual Spanish/ English version item-to-item correlations ranged from .70 to .92, with the exception of two: (a) In what culture(s) do you feel confident that you know how to act? r(36) = .37; and (b) In what culture(s) do you know what is expected of a person in various situations? r(36) = .64. INTERNAL CONSISTENCY Alpha coefficients of reliability for scores on the Spanish and English versions of the PAS were .83 and .85, respectively. Item total correlations ranged from .22 and .68 for scores on the Spanish version and from .27 and .71 for the English version, indicating highly similar patterns of item total correlations across individuals’ responses to the two versions. Study 2 The results from the first study indicated that scores on each language version of the PAS were internally consistent and that individuals’ responses to the PAS were highly comparable across the two language versions. Still, much research on Latinos has been criticized for treating members of different Latino subgroups as part of one homogeneous population (Marin & VanOss Marin, 1991). Therefore, a second study was designed to examine psychometric properties of the PAS within a more specific subgroup of Latino respondents. To date, most acculturation measures have been validated using Mexican American respondents.
In this study, Puerto Rican respondents were used for two reasons: (a) Puerto Ricans tend to be underrepresented in validation studies of acculturation measures, and (b) Puerto Ricans are the largest Latino subgroup in the northeast region of the United States. Method SAMPLE AND PROCEDURES Respondents were recruited for participation in the same manner as in Study 1. A total of 107 Puerto Ricans participated in this study, including 39 males and 64 females (4 respondents did not state their gender). Respondents’ ages ranged from 12 to 58 years (M = 27.9 years). Of the respondents, 85 were born in Puerto Rico and 21 were born on the U.S. mainland. Respondents’ percentage of lifetime spent in the United States ranged from 77% to 100% (M = 92%). MEASURES The measures used in Study 2 were equivalent to those employed in the first study. However, in this study, respondents were asked to respond only to one questionnaire in the language of their choice (i.e., either the Spanish version or the English version).
Cultural behaviors and preferences. Items pertaining to cultural behaviors and preferences were adapted from tradit …