The present study, like that of Bruck and Genesee (1995), compared bilinguals (Spanish and English) to monolingual speakers to determine if bilinguals show an advantage. Like Durgunoglu et al. (1993), this study further asked if beginning readers' performance on phonological awareness tasks in one language would be indicative of the level of performance on these same tasks in the present language. The innovation in this study was that, unlike Durgunoglu et al. (1993) who only tested for transfer effects in one direction, we tested for transfer effects in both directions between English and Spanish. The English and Spanish tests were given not only to bilinguals, but also to monolinguals who were either taught in English or Spanish. Both language versions were administered to all the children, regardless of language dominance or knowledge of the "foreign" language. Overall, the purpose then was to investigate general phonological awareness and its implications for reading ability and its possible transfer between languages within monolingual and bilingual subjects.
Different predictions followed from each of the studies discussed above. First, we predicted that results could be replicated such that reading ability and performance on the phoneme awareness tests were related. Second, taking into account Durgunoglu et al.'s (1993) assertions of language transfer, we predicted a general transfer of phoneme awareness between languages for bilinguals. Consistent with predictions by Walley (Bruck and Genesee 1995) of a bilingual advantage, we expected bilinguals to outperform monolinguals. In addition, based upon Bruck and Genesee's (1995) findings of instructional influences in grapheme-phoneme associations (and acknowledging that students had been formally introduced to these concepts by the time that testing began), we expected that language of instruction would facilitate better performance in that language when compared to the other.
In order to investigate the relationship between phoneme awareness and reading, to examine how bilingual versus monolingual language background may affect its development, and to demonstrate transfer of this ability between Spanish and English, we employed the Yopp-Singer PST, the Mann PST and their Spanish-language adaptations in this study. The Peabody Picture Vocabulary Test (Spanish and English) and the Woodcock Reading Mastery Tests (English) measured vocabulary and reading development. Teacher judgments of the reading ability and language background of the students were considered as well.
Subjects were 52 first grade students from a public elementary school in the Garden Grove (CA) Unified School District. The children came from predominantly low-income backgrounds, and a majority were of Mexican heritage. The children ranged in age from six years, three months to seven years, five months. Only students who returned signed parental permission slips participated in this study.
Subjects were classified into groups according to: a) teacher ratings of language competency and b) language spoken at home. The teachers ranked the children according to the following rating scale: 1) English only (E.O.); 2) bilingual with English dominance (E.D.); 3) bilingual with equal dominance (Eq.D.); 4) bilingual with Spanish dominance (S.D.) and; 5) Spanish-only (S.O.). Nine students were E.O.; nine were E.D.; 12 were Eq.D.; 20 subjects were S.D.; three were S.O. Teachers also rated the children's English reading ability on a five point scale, low (1) to high (5). These children came from bilingual English-instruction and Spanish-instruction classrooms.
The children were assessed for language competency and reading ability using the following standard instruments and their Spanish-language, examiner-generated adaptations: a) the Peabody Picture Vocabulary Test-Revised (PPVT-R), a measure of receptive (heard) vocabulary (Dunn and Dunn 1981); b) a Spanish-language adaptation of the PPVT-R prepared by the author; c) the Word Identification subtest of the Woodcock Reading Mastery Tests, which measures skill in word naming (Woodcock 1973), and; d) the Word Attack subtest of the Woodcock Reading Mastery Tests, which measures subjects' ability to decode nonsense words (Woodcock 1973).
The tests of phoneme awareness that were administered to the children are as follows: a) the Mann PST, which involves the presentation of four illustrations of four words, three of which start with the same sound: the child is to identify the picture that represents the word that starts with a different sound; b) a Spanish adaptation of the Mann PST prepared by the author; c) the Yopp-Singer PST in which the children are presented with a one-syllable word and are asked to say the individual sounds and letters that comprise the word, and; d) a Spanish adaptation of the Yopp-Singer PST prepared by the author.