Browsing by Author "Pena, Sylvia C."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A study of factors which affect the decision-making process in bilingual/bicultural education: a proposed theoretical framework
    (1976) Pena, Sylvia C.; Verner, Zenobia B.; Sterrett, Marvin D.; Weinstein, Joshua; Sobrino, Josephine
    Purpose of the study. The purpose of this study was (1) to define those constructs from the body of literature on curriculum which guide the curriculum development process, (2) to define those factors in bilingual education which affect the decision-making process, (3) to define those factors with regard to the languages in the program which also affect the decision-making process, and (4) to suggest a theoretical framework for the development of bilingual programs at the local level which accommodate the above factors. Scope of the study. The major focus of this study was on the development of a theoretical framework for bilingual programs from a curriculum development perspective at the local school district level. Although the planning, management, and evaluation phases as indiscreet tasks are considered crucial to the eventual success of the curriculum development precess, these activities were not within the scope of this study. As such, they were touched upon in a peripheral manner and only within the context of the development task. The referents for this study were limited to the Mexican American learner participating in bilingual education programs and therefore to Spanish and English as the languages in the program. Proposed framework. The philosophical considerations in curriculum development within the context of bilingual education assume a primary role. The overriding factors which affect the nature and scope of bilingual programs can be summarized as being the learner, society, knowledge, and driving and restraining forces. A universe of values can be said to determine the relative weight of importance placed upon each of the foregoing factors. Consequently, a philosophy of education as dependent variable (P) will reflect the value placed on the independent variables, i.e., the learner (V), the society (W), knowledge (X), and driving (Y) and restraining (Z) forces. The coefficients 'a,' 'b,' 'c,' 'd,' and 'e' represent the relative value placed on each independent variable. [...]
  • No Thumbnail Available
    Item
    Predictors of second language learners' error judgments in written English
    (1983) Guerra, Veronica A.; Walker de Felix, Judith; Acton, William R.; Pena, Sylvia C.; Spuck, Dennis W.
    This study investigated the relations existing between five independent variables and two aspects of second language judgment: error correction and error recognition. The independent variables were: native language skills, years of schooling in the first language, length of residence, previous instruction in EFL and instruction in ESL. The six measures of the two dependent variables, error recognition and error correction, were: recognition of correct responses, recognition of local errors, recognition of global errors, production of correct responses, correction of local errors and correction of global errors. The six measures of error judgment were based on Burt and Kiparsky's classification of errors as conceptualized in their text. The Gooficon (1976). In order to achieve the primary objective of studying the afore mentioned relationships, it became necessary to construct and validate an instrument that would operationally define the six dependent measures of error judgment. Therefore, this study was conducted in two separate but related investigations: the measurement study and the predictive study. In the measurement phase of the study, several analyses were conducted to assess the validity and reliability of the instrument and to test the validity of the global-local distinction as a valid construct. Several measures of validity and reliability were used to test the two scales which comprised the GLD (global-local distinction) instrument. Several alpha factor analyses verified that the instrument validly represented measures of recognition and correction, as well as the complexity of the clausal embedding structure inherent in the six syntactical patterns chosen for inclusion in the two scales. In addition, six Kendall's Tau coefficients indicated that the two scales represented a significant interaction between the type of error (global or local) and the rank order of levels of syntactic organization, but not the one hypothesized by Burt and Kiparsky (1976) in The Gooficon. The results indicated that the deeper, semantic structure of the sentences as affected by the intrinsic SVO order of English displayed a significant relationship with error types regardless of surface clausal complexity. The two factor analytic solution, in addition, indicated that the global-local distinction was not as absolute a dichotomy as claimed by Burt and Kiparsky. The results showed that the errors were relative to the specific language functions: correction or recognition, and the level of discourse in which they were used. For the predictive study, six multiple regression equations, one canonical correlation and Pearsonian correlations were used to obtain a measure of the various hypothesized relationships between the five independent and the six dependent variables. For the test of the global hypothesis, the results showed that there was about 43[percent] variance shared between the two sets of variables. However, after the amount of shared variance had been accounted for, the redundancy index showed that about 26[percent] of the variance in the dependent variables could be explained by the linear combination of the two independent variables. For the test of the subhypotheses, the equations showed that the variable native language skills explained the largest proportion of variance for all dependent measures, except for correction of local errors. The variable years of schooling in L1 by itself did not explain any significant variance, but together with native language skills, it explained the next highest proportion of variance. The variable ESI instruction displayed a significant amount of additional variance in the correction measures especially in the correction of local errors. One possible interpretation to this complex pattern of relationships can be offered based on Cummins (1978, 1979a, 1979b, 1982) and on Krashen's (1978) monitor model of second language learning. It appears that ability in the first language together with years of schooling in the first language are significantly related to ability to comprehend and recognize errors in the second language, but when it comes to production or correction of errors in the second language, schooling in the second language was just as significantly related as first language proficiency and schooling in the first language. The results seem to support the hypothesis that there is an interdependence between L1 and L2 cognitive academic language proficiency. However, within this interdependence framework, the results'showed that the cognitive academic skills involved in comprehension and recognition of errors are a function of native language proficiency together with L1 schooling. but those involved in production of correct responses or correction of errors are more a function of monitoring resulting from schooling in addition to native language proficiency and L1 schooling together.