• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Factors that affect biliteracy development and maintenance of Swahili in bilingual (Swahili-English) speaking children

Dissertation
Author: Josephine Yambi
Abstract:
This qualitative study investigates the bilingual and biliteracy (Swahili-English) development of two elementary and three middle school Kenyan children across home and school contexts in the United States. Guided by sociocognitive and sociocultural theories of language and literacy, the study explores the factors at home and school that supported the Swahili speaking children's bilingual and biliteracy development and how well the children comprehended and wrote narrative and expository texts in English and Swahili. The primary participants included the five children, all of whom received some instruction in all English classrooms, with two of them also receiving limited Swahili instruction, and three of them receiving part-time English-as-second-language (ESL) instruction. The data collected in the homes included observational field notes on language use, Swahili journals that students wrote in weekly, performance-based reading tasks in English and Swahili, and interview data from the parents and children on language use, and the children's literacy histories, literacy identities, language preferences and attitudes. The data collected at school included observational field notes on the language and literacy instruction that the children received, English writing samples, school assessment data, and interviews with the teachers and multilingual, multicultural district coordinator. Data were analyzed using the constant comparative method (Miles & Huberman, 1994). The findings indicated that the extent of Swahili used in the children's homes differed considerably even though all of the families strongly maintained other aspects of their Kenyan cultural identities. Although the parents said that they wanted their children to become bilingual and biliterate in English and Swahili, they primarily supported English literacy because their children's homework was in English, and English was the school language. Only one parent consistently supported her child's Swahili literacy development. Data from the classroom observations showed that the literacy instruction in ESL and mainstream classrooms effectively supported the children's English literacy development. When three of the Kenyan parents volunteered to teach three days a week 45-minutes Swahili class, then the two younger students participated, but the school instruction they received was not sufficient to help them read and write grade level Swahili. The data on student performance illustrated differences in the children's strengths and weaknesses in both English and Swahili literacy. Prior to entering school in the United States, all of the children had received some literacy instruction in Swahili and English in Kenyan schools. Although the two elementary students' age of arrival to the Unites States fell within the category of 5-7 year olds, immigrant children often take longer to attain academic English than children who arrive between ages 8-11; one of the children performed well and exited ESL within three years. Also, one of the older students, all of whom arrived in the Unites States when they were 8-12 years old, demonstrated much higher literacy performance in English than the other students. The students' varied performance in English indicated that other factors besides age of arrival in relation to English academic achievement are important to consider. The findings revealed that English was the stronger language for four of the five children, and that the child who was the strongest English reader and writer also was becoming a balanced biliterate in English and Swahili. Four of the five children were either losing or not increasing their oral and literate Swahili proficiency because the home and school contexts contributed to subtractive bilingualism and biliteracy. A major implication of the study is that immigrant parents need to be aware of the adverse effects of only emphasizing English literacy on their children's bilingual and biliterate development and identities. The language loss findings illustrate the strong hegemonic influence of English in both the United States and in Kenya.

Table of Contents List of Tables....................................................................................................................ix Chapter 1 Introduction.....................................................................................................1 Research Rationale............................................................................................... 1 Purpose of Study....................................................................................................4 Theoretical Perspectives........................................................................................5 Definition of Terms................................................................................................8

Chapter 2 Literature Review..........................................................................................11 Factors That Affect Bilingual Children’s Reading Comprehension...............11 Cognitive Factors in Bilingual Children’s Writing..........................................27 Language Loss......................................................................................................32 Supportive Bilingual and Biliteracy Development Contexts...........................36 Sociolinguistic Context of Swahili in Kenya......................................................44

Chapter 3 Methodology.................................................................................................. 51 Research Design...................................................................................................51 Research Site........................................................................................................52 Participants...........................................................................................................55 Pilot Study............................................................................................................59 Data Collection Procedures.................................................................................60 Data Analysis........................................................................................................71 Research Validity/Credibility.............................................................................74

Chapter 4 Findings..........................................................................................................76 Family Profiles.....................................................................................................76 Home and Family Influences on Bilingualism and Biliteracy Development.........................................................................................................81 School Literacy Instruction and Assessment...................................................104 Performance Assessment of Students’ Biliteracy Development....................140

Chapter 5 Summary, Discussion and Implications.....................................................204 Summary.............................................................................................................204 Discussion of the Findings.................................................................................205 Implications for Education and Research.......................................................228 Limitations of the Study....................................................................................231

References.......................................................................................................................234

Appendix A Sample Teacher Interview Questions.....................................................244

Appendix B Sample Student Interview Questions......................................................245 Appendix C Sample Parent Interview Questions.......................................................246

viii

Appendix D Sample Reading Texts, Retelling Templates and Student Retellings.........................................................................................................................249

Appendix E World-Class Instructional Design and Assessment Consortium (WIDA) Performance Definitions...........................................................261

Appendix F Writing Scoring Rubric............................................................................262

ix

List of Tables Table Page 1 Language Use in Kenyan Schools 46 2 Students’ Background Information 56 3 Total Classroom Observations per Student 62 4 Summary of Kenyan and US Schooling and Family Information 77 5 Kevin, Robert and Diana’s English Language Proficiency Test Scores (Spring 2006) 132 6 Kevin, Robert and Diana’s Illinois Measure of Annual Growth in English (IMAGE) Reading Scores (Spring 2006) 135 7 Summary of Students’ English Reading Comprehension Performance 142 8 Ratings of English Writing 161 9 Summary of Students’ Swahili Reading Comprehension Performance 171 10 Ratings of Swahili Writing 188

1

Chapter 1 Introduction

Research Rationale Many children all over the world are becoming bilingual and bi-literate due to education language policies implemented in their respective countries. In the US, for example, there are 14 million students who are non-native speakers of English in grades K-12 (August, 2008; US Census Bureau, October 2006). One quarter of these students are foreign born while three quarters are US born. Many of these students are becoming bilingual, that is, they are learning to speak more than one language. They are speaking their home language and English, because English is the language of the school. Contemporary research on upper-elementary and middle school bilingual children’s literacy development has mostly investigated second language (L2) reading, with a focus on cognitive factors that affect academic development. Major findings from this research indicate that successful bilingual readers use many of the same metacognitive and cognitive strategies as successful monolingual native English speakers to read English except that the bilingual readers tend to know less of the English vocabulary and topics emphasized in English reading texts and assessment (García, 1991; García, 2000; Hardin, 2001; Jiménez, García & Pearson, 1995; 1996). Jiménez, et al (1996) also found advantages that a proficient bilingual reader had when she used Spanish and English cognates as a strategy when monitoring comprehension of text in either language, an advantage that the monolingual reader was lacking. Cognates are

2

words found in two languages with same the ancestral roots, and such words are similar in form and meaning (García, 2003). In addition, researchers interested in bilingual children’s reading have mainly studied Spanish-English children. One reason for this focus is that there are large numbers of Spanish speaking children in US schools (August, 2008). However, to further understand bilingual children’s reading, studies on students of different language backgrounds also are needed. Swahili-English speaking children, for example, are among the 3% of children labeled “Other” on data reflecting the origins and percentages of immigrant children in US schools (Fix and Passel, 2003). Moreover, Swahili and English are two linguistically unrelated languages, while Spanish and English are closely related languages. English is a Germanic language from the Indo-European group of languages while Swahili is a Bantu language in the Niger-Congo group of languages (Heine & Nurse, 2000). It is highly likely that the reading performance of bilingual students speaking unrelated languages will differ from that of bilingual students speaking related languages, especially in terms of bilingual issues, such as cross-linguistic transfer. Cross linguistic transfer occurs when bilingual children use reading skills and knowledge gained in one language to read in another. Researchers have also observed that when immigrant children are immersed in the majority language, English, it is not uncommon for them to experience language shift from the home or primary language (L1) to the L2. They gradually lose their L1 and literacy abilities, particularly if they moved to a new country when their L1 was not yet fully developed. As a result of language shift and/or loss, often the children and extended family members may become disconnected socially and culturally (Wong-Fillmore,

3

1991). Family members who do not speak English, the children’s newly acquired language, can no longer communicate with the children. Research is lacking on the factors that affect children’s L1 literacy development while learning their L2. Available research on younger children (e.g., Wong-Fillmore, 1991) suggests that school programs for preschool non-native English speakers together with parental attitudes toward English contribute to children’s loss of their L1. Therefore, researchers (e.g., Collier, 1995; Goldenberg, Rueda & August, 2006; Hornberger, 1992) have raised the need for additional research that will explore children’s biliteracy development in relation to various socio cultural influences that contribute to the underdevelopment of L1. In their recent analysis of studies related to socio cultural influences on the literacy attainment of language-minority students and youth, Goldenberg, Rueda & August (2006; 2008) note that most of the socio cultural studies reviewed do not include school achievement data. Therefore, the combination of home data and achievement data is a contribution that this study is making to the field. However, it is important to note that there is also a lot of individual variation among L2 learners. L2 learners vary in terms of age, attitudes, and home literacy experiences and expectations for literacy (Gregory, 1996; Schwarzer, 2001). For example, while some immigrant families are planning to live in the US indefinitely, others are planning to return to their home countries (Weisberg & Ortiz, 2000). A number of researchers have studied bilingual children’s reading (e.g., García, 1998, 2000; Droop & Verhoeven, 1998) or writing (e.g., Moll, Sáez & Dworin, 2001; McCarthey, García, Lopez-Velásquez, Lin & Guo, 2004; McCarthey & García, 2005) in school contexts. However, fewer researchers have investigated both writing and reading

4

in bilingual children. More research is needed to examine the bi-literate experiences of different language groups, such as Swahili and English speakers. Researchers interested in the biliteracy development and maintenance of L1 of language minority students also need to investigate students who do not receive L1 instruction in the school context but who had literacy in L1 before immigrating to the US. To further our understanding of biliteracy development in children acquiring bilingualism successively, that is, learning L2 after they have acquired the basic linguistic foundations in L1, more researchers need to explore bilingual children’s literacy development in their L1 and L2 (García, 2000). For example, “researchers need to study how bilingual children’s knowledge and use of L1 interact with their knowledge and use of L2 at school and in other environments; and to what extent children’s bilingualism and biliteracy abilities play a role in their academic development” (García, 2000, p. 830).

Purpose of Study Moll, Saez and Dworin (2001) describe the intellectual advantages of being bi- literate, which range from “gaining access to valued cultural resources, to developing metalinguistic awareness, to deliberately exploiting literacy as a tool for thinking” (p.436). This study investigated Swahili-English bilingualism and biliteracy in children as part of the broader social contexts, including home and school. It investigated the influences in the home and school contexts in relation to facilitating or delaying the development of bilingualism and biliteracy in the children studied.

5

The study was guided by the following questions: 1. What are the factors at home and school that support Swahili speaking children’s bilingual and biliteracy development or contribute to their Swahili literacy attrition or loss?

2. How well do Swahili speaking students comprehend and write narrative and expository texts in English and Swahili?

Theoretical Perspectives To investigate students’ bilingualism and biliteracy development I employed theoretical perspectives that combine cognitive and sociocultural aspects of literacy. The cognitive perspective guided the analysis of student successes and challenges in reading and comprehending texts, as well as writing, in English and Swahili. The sociocultural perspectives were important for understanding sociocultural contexts and influences that helped shape student’s bilingualism and biliteracy development. Such contexts included the school (e.g., teachers, instruction, materials), home (e.g., language use and language preferences, literacy practices, and children’s attitude toward the languages), and the larger community. The cognitive perspective of this study is informed by theories of literacy development that acknowledge the importance of native language in the development of literacy in the L2 (Collier, 1995; Cummins, 1979; 1981). Cummins’ (1979) interdependence and threshold hypotheses suggest that there is an underlying cognitive or academic proficiency which is common across languages that makes it possible for the transfer of cognitive/academic or literacy related skills across languages. As a result, knowledge and skills acquired in L1 can be used to accomplish tasks in L2. Therefore, Cummins (1979) interdependence hypothesis suggests a reciprocal relationship between

6

L1 development and L2 development, that is, the languages of a bilingual are interdependent. If the L1 is well developed, it would be easier to develop the L2. By drawing from this theory we can understand how children in this study engaged in English and Swahili reading and writing. The sociocultural perspectives on literacy are informed by theories that acknowledge the importance of socio cultural contexts and influences that affect literacy learning. Sociocultural theory underscores the importance and influence of the context and interaction on language development. Vygotsky’s socio cultural theory on human development suggests that human development cannot be viewed outside of its social context, and that it occurs as a result of meaningful interaction and relationships between novices (e.g. child) and experts (e.g. adult) (Vygotsky, 1978). Through culturally constructed tools, most importantly language, and active participation in daily activities, elementary processes, such as oral language, reading and writing, can be transformed into higher order thinking. Vygotsky (1978) described this process in his concept of zone of proximal development, which refers to what the learner or child can accomplish with the assistance of an expert or adult. In the process of learning, the child internalizes what she is learning with the assistance of an adult and makes that knowledge her own. In other words, what a child can accomplish with assistance today, she can do by herself later (Cole, John-Steiner, Scribner & Souberman, 1978). Therefore, according to this view, literacy learning is socially based, interactive and influenced by context. Regarding bilingualism and biliteracy development of linguistically diverse students, Hornberger (2005) maintains that “biliteracy addresses the conjunction between bilingualism and literacy” (p.319). In her work, Hornberger (1989) proposed a

7

framework of continua of biliteracy that details the existence of multiple and complex relationships between bilingualism and literacy and the importance of the contexts, media and content through which biliteracy develops. Hornberger (2005) maintains that the continua of biliteracy model hypothesizes that contexts influence biliteracy development and use at every level from two-person interaction (micro) to societal and global relations of power (macro) and that they comprise a mix of oral to literate, monolingual to multilingual varieties of language and literacy (p. 329). In addition, the development of biliteracy occurs along intersecting L1 to L2, oral to written, receptive (listening and reading) to productive (speaking and writing) language and literacy skills, and literacy practices. Literacy practices include uses of and attitudes toward language and literacy. She maintains that biliteracy learning may proceed in any direction along those intersecting continua (p. 331). According to the continua of biliteracy, the content that bi-literate learners and users read and write is as important as how (development), where (context) or when and by what means (media) they do so. Hornberger argues that whereas schooling traditionally privileges majority, literary, and decontextualized content, the continua model argues for greater curricular attention to minority, vernacular, and contextualized whole language texts. Minority texts include those by minority authors, written from minority perspectives; vernacular ways of reading and writing include notes, poems, plays, and stories written at home or in other everyday non school contexts; contextualized whole language texts are those read and written in the context of biliteracy events, interactions, practices, and activities of bi-literate learners’ everyday lives (p.334).

8

The media in the continua of biliteracy model refers to language varieties and scripts through which multilingual literacies are expressed, and the sequences or configurations in which they are acquired and used. The model defines these in terms of the linguistic structures of the languages involved (on a continuum of similar- dissimilar), their orthographic scripts (from convergent to divergent), and the sequence of exposure to or acquisition of the language/literacies (ranging from simultaneous to successive) (p.337). Hornberger (1992) argues that “the implications of the model of biliteracy outlined by the continua framework are that the more the contexts of their learning allows bi-literate learners to draw on all points of all nine continua, the greater are the chances for their full bi-literate development” (p. 199). The continua of biliteracy framework is a relevant research tool for this study, which sought to investigate how Swahili- English speaking children’s bilingualism and biliteracy development are influenced and affected by the home and school contexts.

Definition of Terms Bilingualism is defined as possessing basic or minimal communicative skills in a second or foreign language (Hakuta, 1990; Hornberger, 1989). At the individual level, a distinction is made between simultaneous and sequential/successive bilingualism. Simultaneous bilingualism begins from the onset of language acquisition; in other words, a child is exposed to two languages and acquires speaking skills in the languages at the same time. Sequential or successive bilingualism, on the other hand, begins after the basic components of first language knowledge have been established (McLaughlin, 1984), that is, some knowledge of L1 is acquired before L2 is introduced. In this study,

9

the children are referred to as bilingual because they spoke Swahili and English at varying levels of proficiency. All children experienced successive bilingualism. In this study, the term biliteracy is used to describe children’s literate competences in the two languages, developed to varying degrees (Dworin, 2003; García, 2000). Similar to the concept of bilingualism, simultaneous biliteracy development means learning to read and write in both languages at the same time while successive biliteracy means acquiring the ability to read and write in the L2 after acquiring literacy in the L1. The terms receptive and productive have been used in the study regarding students’ language development and proficiency. Receptive proficiency refers to listening and reading competence while productive proficiency refers to speaking and writing competence (Hornberger, 1989). Given that the children in this study are sequential bilinguals, the term home language is used to refer to the language that the child spoke before learning English. It is sometimes referred to as the first or primary language (L1) versus the foreign or the second language (L2). In East Africa, due to inter- ethnic marriages, some children are exposed to more than one home language, that is, the language of each of the parents. In this situation, the language used between parents, and between parents and their children, is the language of wider communication. In this study four children spoke Swahili as their home language before learning English, while the fifth child spoke both Swahili and an ethnic language but was literate only in Swahili. The term language loss refers to lack of first language development, delayed first language, or progressive loss of previously acquired language ability (Verhoeven & Boeschoten (1986). In some occasions, the term language attrition is also used for the

10

same situation. In this study the terms language loss and language attrition are used to refer to progressive loss of the home language. Sometimes the term subtractive bilingualism is used when individuals lose their first language as they acquire the second language. Additive bilingualism, on the other hand, occurs when students continue to develop their first language as they acquire the second language (Lambert, 1975). Therefore, in an additive bilingualism situation, the first language is maintained. It has been observed that L1 maintenance in bilingual education contexts contributes to students’ self esteem (Lambert, 1975). Although the term language maintenance refers to the actual use of the L1, it relates also to a broader concept that relates to positive attitudes toward the language (Fishman, 1991).

11

Chapter 2 Literature Review In this chapter I review four main areas that are important for informing the present study that explores factors affecting biliteracy development and language maintenance in Swahili-English bilingual speaking children. First, I review research related to L2 reading comprehension and L2 writing. Next, I discuss research on language loss which is relevant considering the participating children’s Swahili and English development. Also, I discuss research about supportive bilingual and biliteracy contexts in school and other contexts. To conclude, I discuss the sociolinguistic context of Swahili in Kenya to provide background information on language use in the home country of the case studies of the present study. The focus of the research reviewed in this chapter is on upper elementary and middle school children because of the age group of the participating students in the study. However, research on younger children, high school students and adults is also reviewed when appropriate.

Factors That Affect Bilingual Children’s Reading Comprehension Researchers have observed that both non native speakers of English and native speakers of English use similar reading processes to comprehend English texts (Bernhardt, 2000; Gregory, 1996). For example, both types of speakers make use of graphophonic knowledge, that is, they use clues concerning patterns of letters and the sounds they make. Secondly, they both make use of lexical knowledge, which is responsible for sending out clues about the word and the company it keeps. Third, they

12

make use of syntactic information, that is, the clues about the language structure; and they also make use of semantic knowledge, that is, the meaning of words within the culture or within the text they are reading. In addition both groups make use of their background knowledge when reading. However, there are also some differences between L1 and L2 readers. Bernhardt (1991; 2000; 2003) emphasizes that since there is a first and a second language, the second language reading process is different from the first language reading process due to the nature of information stored in the reader’s memory. L2 readers also encounter more unfamiliar L2 words, and in relation to semantics, they have fewer associative links between L2 words. They also lack background knowledge for some of the English topics and text structures. Vocabulary knowledge. For second- language English learners, vocabulary knowledge is a highly significant variable for effective text comprehension (Bernhardt, 2000; Fitzgerald 1995). Compared to monolingual English speakers, L2 speakers often encounter more unfamiliar words in English reading texts. García (1991; 2000) showed that unknown vocabulary items in test items negatively affected L2 students’ English reading test performance. Savile-Troike (1984) studied 19 students from 7 different language groups who had very little prior knowledge of English to identify variables impacting the academic achievement of L2 children. The students were in grades 2 nd through 6 th and received English as second language instruction. Their family backgrounds were similar, and all of them had some literacy knowledge in their L1. The students were tested on aspects of competence in morphology, syntax and vocabulary, verbosity patterns of social interaction, first language performance and personality factors. Saville-Troike reported

13

that reading achievement, as measured by a Comprehensive Test of Basic Skills, correlated significantly with the number of words that each child used orally over a period of one year. She noted that students who were strong in oral English had corresponding high English reading achievement. She found vocabulary knowledge in the L2 to be the most important aspect of oral proficiency for school achievement. García (1991) studied factors that influenced the English reading test performance of 51 Hispanic students as compared with the performance of 53 Anglo students enrolled in the same 5 th and 6 th grade classrooms. Factors that were studied included differential effects of time constraints during the reading test, students’ prior knowledge of the reading texts and information regarding test questions. García also asked students about vocabulary used in the test. García found that the Hispanic students’ unfamiliarity with vocabulary terms used in the test questions and answer choices was one of the factors that adversely affected their test performance. García also reported that some Spanish speaking children showed enhanced understanding of the texts and test questions when they heard the questions in Spanish. Background knowledge. Researchers have found that often L2 readers do not have sufficient background knowledge to fully comprehend the topics they read in the L2 (Abu-Rabia 1996; Droop and Verhoeven, 1998; García 1991; Jiménez, García & Pearson, 1995; Jiménez, García & Pearson, 1996). García’s (1991) analysis of the standardized English reading test performance of Hispanic and monolingual Anglo 5 th and 6 th graders also found that the test performance of the Hispanic students was negatively affected by their limited prior knowledge of test topics and their poor performance on questions that required use of background knowledge. But when differences in prior knowledge were

14

statistically controlled, there was no difference between the monolingual and bilingual children on passage performance. This finding suggests that there were other factors than the prior knowledge that affected Hispanic students’ performance on the test. Jiménez, García and Pearson (1995) used think alouds to compare the reading processes of three 6 th grade students: a proficient bilingual (Spanish-English) Latina reader, a marginally proficient bilingual (Spanish-English) Latina reader and a proficient monolingual Anglo reader. A prior knowledge measure was also used to collect data. Based on this measure, the researchers found that, compared to the two bilingual readers, the familiarity of reading topics on the English passages was an advantage that benefited the monolingual English reader. In the same line of research, Jiménez, et al (1996) explored the reading performance of 8 successful bilingual Latina/o and 3 monolingual Anglo 6 th and 7 th grade students and 3 less successful bilingual Latina/o students. Among other findings, the results of this study showed that the successful bilingual and monolingual readers were able to discuss the reading texts during the think aloud procedures using prior knowledge of relevant topics. In a formative experiment on reading strategy instruction with five 7 th grade Latina/o students who were low –level readers, Jiménez (1997) reported that when culturally familiar topics were used, students produced extended discourse about the texts. Abu-Rabia (1996) also reported findings of a cultural familiarity effect from a sub-sample in a study that investigated the relationship between students’ attitudes toward second language and cultural background to reading comprehension. The participants were 8 th grade students in three different social contexts. The first group was 74 Arab students in Israel who were learning Hebrew as their second language; the

Full document contains 276 pages
Abstract: This qualitative study investigates the bilingual and biliteracy (Swahili-English) development of two elementary and three middle school Kenyan children across home and school contexts in the United States. Guided by sociocognitive and sociocultural theories of language and literacy, the study explores the factors at home and school that supported the Swahili speaking children's bilingual and biliteracy development and how well the children comprehended and wrote narrative and expository texts in English and Swahili. The primary participants included the five children, all of whom received some instruction in all English classrooms, with two of them also receiving limited Swahili instruction, and three of them receiving part-time English-as-second-language (ESL) instruction. The data collected in the homes included observational field notes on language use, Swahili journals that students wrote in weekly, performance-based reading tasks in English and Swahili, and interview data from the parents and children on language use, and the children's literacy histories, literacy identities, language preferences and attitudes. The data collected at school included observational field notes on the language and literacy instruction that the children received, English writing samples, school assessment data, and interviews with the teachers and multilingual, multicultural district coordinator. Data were analyzed using the constant comparative method (Miles & Huberman, 1994). The findings indicated that the extent of Swahili used in the children's homes differed considerably even though all of the families strongly maintained other aspects of their Kenyan cultural identities. Although the parents said that they wanted their children to become bilingual and biliterate in English and Swahili, they primarily supported English literacy because their children's homework was in English, and English was the school language. Only one parent consistently supported her child's Swahili literacy development. Data from the classroom observations showed that the literacy instruction in ESL and mainstream classrooms effectively supported the children's English literacy development. When three of the Kenyan parents volunteered to teach three days a week 45-minutes Swahili class, then the two younger students participated, but the school instruction they received was not sufficient to help them read and write grade level Swahili. The data on student performance illustrated differences in the children's strengths and weaknesses in both English and Swahili literacy. Prior to entering school in the United States, all of the children had received some literacy instruction in Swahili and English in Kenyan schools. Although the two elementary students' age of arrival to the Unites States fell within the category of 5-7 year olds, immigrant children often take longer to attain academic English than children who arrive between ages 8-11; one of the children performed well and exited ESL within three years. Also, one of the older students, all of whom arrived in the Unites States when they were 8-12 years old, demonstrated much higher literacy performance in English than the other students. The students' varied performance in English indicated that other factors besides age of arrival in relation to English academic achievement are important to consider. The findings revealed that English was the stronger language for four of the five children, and that the child who was the strongest English reader and writer also was becoming a balanced biliterate in English and Swahili. Four of the five children were either losing or not increasing their oral and literate Swahili proficiency because the home and school contexts contributed to subtractive bilingualism and biliteracy. A major implication of the study is that immigrant parents need to be aware of the adverse effects of only emphasizing English literacy on their children's bilingual and biliterate development and identities. The language loss findings illustrate the strong hegemonic influence of English in both the United States and in Kenya.