Hispanic Journal of Behavioral Sciences

Children who interpret for their immigrant parents are referred to as language brokers. The present study examines the relationship of language brokering to academic performance, biculturalism, academic self-efficacy, and social self-efficacy. The many adultlike experiences of children who broker on a regular basis suggest that their cognitive and socioemotional development may be accelerated relative to children of immigrant families who broker infrequently or not at all. Latino adolescents (n = 122) from immigrant families were participants in the study. Results showed that, as expected, language brokering was positively related to biculturalism, and in turn, both of these variables were positively related to academic performance. In addition, the strongest predictor of academic performance was academic self-efficacy. Results also indicated that, to some degree, language brokering is a gendered activity, with females reporting more brokering than males.

Research that examines the effect of Latino ethnicity on misdemeanor sentencing decisions is virtually nonexistent. In this study, the authors analyze 1992 misdemeanor sentences in three nonurban Nebraska counties with relatively large Latino populations. Support for "gringo justice" emanates as Latinos overall had significantly higher proportions of individuals charged with misdemeanor offenses other than simple traffic violations (i.e., misdemeanor alcohol and drug offenses). They also were recipients of a higher mean number of charges. Subsequently, Latinos received significantly higher meanfines and mean days probation but not mean days jail in comparison to their White counterparts. Findings suggest that biased discretion in the enforcement, processing, and sentencing of misdemeanor offenses is but one part of the cumulative disadvantage that Latinos face in the overall criminal justice system.

This article explores whether a data set on demographic and economic characteristics of Hispanic-origin persons in the United States from the 1990 U.S. Population Census can inform the discussion on the detenninants of remittances from this population to selected Latin American and Caribbean countries. Parts I and 2 of the article discuss available data on remittances flows to selected Latin American countries and on the characteristics of persons of Hispanic onigin in the United States, respectively. Part 3 reviews the literature on the determinants of remittances. Part 4 discusses demographic and economic characteristics of migrants from Costa Rica, Guatemala, Honduras, El Salvador, the Dominican Republic, and Mexico residing in the United States in 1990. Part 5 uses graphic and simple statistical techniques to test some hypotheses regarding remittances-sending behavior The explorations confirmed several propositionsfrom the literature regarding the determinants of remittances, particularly the importance of household, family, or personal income.

Based on Trope 's (1986) model of perception, the effects of physical context on stereo-typing and willingness to affiliate with Mexican American males were examined. In this two-part study, the effect of context alone was explored by showing a stilled videoframe of a library room and a crime scene without people in them. Results indicated that the context affected expectations of the type of people who might be seen in those contexts. In the second study, the interaction of context and stereotyping was examined by placing two Mexican American males (one of whom was "stereotypical") in scholar, crime, and neutral contexts. Results reveal an interaction between context and target stereotyping whereby physical context did not affect stereotyping except when the target could not be easily categorized by ethnicity alone. Discussion centers on contributions of these studies to the literature on context, stereotyping of Mexican Americans, and future research.

This study examines the effect of employment instability on annual eamings of Mexican-origin men. Data are taken from the 1987 Current Population Survey annual demographic file. The author exploits the laborforce participation information providedfor the year prior to the survey. Respondents who experienced spells of unemployment or worked part-time involuntarily during the previous calendar year are categorized as unstably employed The author estimated linear regression models separatelyfor Mexican-origin and White workers. Results show that, net of human capital endowments and economic sector location, employment instability exerts a heavier penalty on Mexican-origin workers than on their White counterparts.

The goal of this study was to examine the relationship between acculturation and familism in a samnple of 182 Hispanics of Puerto Rican descent. Exploratory factor analysis was used to establish the multidimensionality of each construct. Consistent with earlier studies, three dimensions of acculturation and familism were identified. Each dimension offamilism was then regressed on a set of explanatory variables that included three dimensions of acculturation. For two dimensions offamilism (familial obligations and support from relatives), acculturation was positively related to familism; for the other dimension (family as referents), the relationship was nonsignificant. In explaining variance in attitudinal familism, the results suggest that acculturation may be better represented as a single, second-orderfactor In interpreting the findings in the context of previous research, however; it is useful to treat acculturation as a multidimensional construct.

The present study, using bilingual (English-Spanish) college freshmen as participants, evaluated the psychometric characteristics of the Personality Inventoryfor Youth (PIY) that was translated into Spanish. The scales and subscales of the PIY-Spanish Version demonstrated internal consistency and temporal stability estimates comparable to their English counterparts established in both the present sample of college freshmen and in the PIY normative sample of students attending regular education classes ranging from 9 to 18 years of age. Multivariate analysis did not identify a language effect within the four study samples (English-English, Spanish-Spanish, English-Spanish, Spanish-English) for either profile scales or subscales, and the substantial correlations between English and Spanish versions obtainedfrom the same bilingual students approached and occasionally exceeded the limits established for temporal reliability in both English and Spanish translations.