Transfer in the Academic Language Development of Post-secondary ESL Students

Binbin Jiang University of the Incarnate Word

Phyllis Kuehn California State University, Fresno

Abstract

This study examined the issue of transfer for low-intermediate ESL students enrolled in an academic English development course at the community college level. Late immigrant students (adult immigrants) had higher L1 academic language proficiency and generally made better progress in L2 academic language development as a result of instruction than did the early immigrant group (those who immigrated at a younger age). The findings support Cummins' Model of Common Underlying Proficiency (CUP) and the Interdependence Hypothesis between L1 and L2, complement previous studies on literacy development in first and second languages, and provide both quantitative and qualitative evidence on positive transfer of prior linguistic and cognitive skills from L1 to L2.

Sections of the Article

Introduction

The number of English as a Second Language (ESL) students has been increasing dramatically in schools in California and the rest of the United States in recent years (Crandall, 1993; U.S. Census Bureau, 1995 U.S. Census Bureau Projections on 2010). Over 22 million U.S. residents are foreign-born and approximately one-third of these are living in California. ESL students at the post-secondary level often have to take ESL classes to meet institutional standards for English language proficiency before they can become fully matriculated into the academic mainstream (Kasper, 1998).

Many ESL professionals have realized that general English proficiency is not all that ESL students need in order to succeed in mainstream courses. They also need English academic language development. Language proficiency is "only a means to an end: the critical outcome . . . is how well (students) succeed in school" (Saville-Troike, 1984, p. 217). According to Flowerdew (1994) and Wang (1996), educators and researchers generally agree that low proficiency in academic language and the distinctive type of English used in classrooms and in textbooks are contributing factors to academic failure among language minority and at-risk minority students (Wright & Kuehn, 1998).

A distinction between the development of academic language proficiency and the development of spoken or interpersonal language proficiency has been made by researchers within second language, bilingual, and first language learning contexts (Stotsky, 1979; Cummins, 1979a, 1981, 1992, 1994; Chamot & O'Malley, 1986, 1987, 1994; Valdez, Pierce, & O'Malley, 1992). Cummins (1981) suggests that cognitive academic language proficiency is necessary in order for students to use language for thinking and reasoning, rather than exclusively for basic interpersonal communication or conversation.

Theoretical Framework of Academic Language

Chamot and O'Malley (1994) posit that any pedagogical approach should have a sound theoretical basis to meet at least three criteria: (a) the approach should be grounded in research; (b) it should explain what and how something is learned; and (c) it should provide guidance for instruction (p. 12). This section reviews the theoretical framework approaches related to academic language. The construct of academic language and its terminology was initially advanced by Cummins (1981), and other research into academic language ensued. There has been general agreement among researchers that the understanding of the academic language used in the classroom may contribute to the academic achievements of students (Chamot & O'Malley, 1994; Collier, 1989; Cummins, 1981; Spolsky, 1989; Saville-Troike, 1991).

The conversational English used in informal interpersonal communications (also advanced by Cummins, 1981) is not usually thought to require the higher level thinking skills associated with academic language, which is a specific and specialized classroom register. Academic language, by contrast, demands, for example, the ability on the part of the student to understand and generate the complex syntax of Standard English in formal oral and written expression (cf., Wright & Kuehn, 1998; Chamot & O'Malley, 1994).

Solomon and Rhodes (1995), reinforcing distinctions advanced by Cummins' language model, state that one distinction separating conversational language from academic language is that more contextual support is readily available to support conversational language. In other words, it is easier to express oneself and understand others using informal language than when confronted with the formal academic language of the classroom. As Cummins (1996) states, "As students progress through the grades, they are increasingly required to manipulate language in cognitively demanding and context-reduced situations that differ significantly from everyday conversational interactions" (p. 58).

Language Transfer

Chomsky (1979) theorized that there are underlying principles of grammar which children acquire naturally. Chomsky refers to this as universal grammar. According to his theory, humans produce language through a deep structure that enables them to generate and transfer their own grammar to any other language (Lakshmanan, 1994, p. 22). As people learn languages, they develop certain skills. They naturally transfer the skills learned in the first language (L1) to the second language (L2). The level of proficiency reached in L1 influences the development of their proficiency in L2.

Language transfer causes the learner to project the features of his own language onto the target language. According to Corder (1994),

O'Malley and Chamot (1990) define transfer as "the use of previous linguistic or prior skills to assist comprehension or production" (p. 120). Their research supports the idea that second language acquisition is "the operation of transfer of both declarative and procedural knowledge from the first to the second language" (p. 192). Declarative knowledge is the knowledge of factual data which resides in memory and the structural order of such data or the rules which surround language. Procedural knowledge is how the data is processed and the strategies used for communication.

Faerch and Kasper (1987) define transfer "as the process by which L2 learners' active L1 knowledge in developing or using their interlanguage (linguistic system between L1 and L2, to be defined later), and point out that the process may either support (positive transfer) or detract (negative transfer) from learning" (O'Malley & Chamot, p. 148). The study of language transfer strategies may help us to better understand the positive influence first language knowledge has in the learning of another language. This study, therefore, focused on the positive transfer of knowledge from L1 to L2.

O'Malley and Chamot (1990) suggest that this transfer of "concepts or skills known in the first language" to the second language is not necessarily automatic and that cultural experiences may also have an effect "on the development on various types of schemata and the ease or difficulty with which cultural inferenced story grammar, event structure, or discourse organization schema might accommodate new information of this type learned in the second language" (p. 192). These concepts form the basis for the Cognitive Academic Language Learning Approach (CALLA), part of which involves the teaching of learning strategies.

Empirical studies have shown the transfer of prior linguistic and cognitive knowledge from L1 to L2. O'Malley and Chamot's study of 64 Spanish and 34 Russian students showed that the beginning to intermediate level students used more transfer strategies in learning new languages than the more advanced students (p. 124). The subjects in this particular study may support the hypothesis that beginning to intermediate level college students will also use more transfer strategies than more advanced college students. These transfer strategies can be directly taught. For example, one of O'Malley and Chamot's studies showed that students were being taught specific cognitive reading strategies, which included transfer of cognates and similar sounding words in the L1 languages that could be applied to understanding new words in the L2. This showed that transfer and other strategies can be effectively taught in the classroom.

The Relationship Between L1/L2 Academic Language Proficiency and Transfer

According to Cummins (1981), there are two important types of language proficiency which he labeled Cognitive Academic Language Proficiency (CALP) and Basic Interpersonal Communication Skills (BICS).Cummins' threshold hypothesis states that BICS must develop to a certain level before CALP can be effectively added to it (p. 32). In other words, students must reach a certain level or threshold of informal proficiency before they will be able to develop CALP in that language. Students with strong academic skills in L1 generally tend to acquire the needed information in L2 more quickly than those without sufficient formal schooling in L1. However, the latter students may learn ways to overcome this.

Cummins' Common Underlying Proficiency (CUP) Model of Bilingual Proficiency indicates the progression from literacy in L1 to fluency in L2, including academic language literacy. The model presents the concept of CUP which focuses upon the relationship between language and thought. According to Cummins, "Concepts are most readily developed in the first language and, once developed, are accessible through the second language. In other words, what we learn in one language transfers into the new language" (in Freeman & Freeman, 1994, p. 176).

Evidence for transfer of language skills also comes from research done by Cummins (1979). He examined the relationship between bilingual children's academic skills in the primary and second language. Correlations between L1 and L2 proficiency ranged between .60 to .80, demonstrating that children who read well in their first language were likely to read well in the second language (Cummins, 1979b).

The authors examined the same issue (relationship between bilingual's academic language skills in L1 and L2) in this research but with a focus on adult learners. The results complement and contribute to Cummins' and other related research regarding the relationship between bilinguals' academic proficiency in L1 and L2 and the issue of transfer (Bossers, 1991; Canale, Frenette, & Bellanger, 1988; Carson & Kuehn, 1994; Cummins, 1984; Hoover, 1983; Verhoeven, 1991).

As people learn languages, they develop certain skills which may naturally transfer from L1 to L2. The level of proficiency reached in L1 does influence the development of their proficiency in L2 (Odlin, 1986). Empirical studies have indicated that the "transfer of ability to L2 can only occur if individuals have already acquired that ability in their L1" (Carson & Kuehn, 1994, p. 260).

When students of second languages test out hypotheses concerning the nature of the second language, they may transfer certain cognitive knowledge learned in their first language to the second language. Specifically, they may transfer their reading and writing knowledge from L1 to L2, including their use of strategies. If they make assumptions which are true about the second language, the transfer will be a positive one. If not, the transfer may be negative, causing interference with the learning process. If teachers focus on the positive transfer skills that students have, it may be possible to teach these students ways to help them use the second language more effectively.

The purpose of this research was to examine the factors associated with the development of ESL students' English academic language and the role of transfer (the use of previous linguistic or prior skills to assist comprehension or production) at the post secondary context. These factors included first language proficiency, English language proficiency, and L1 to L2 transfer. The researchers sought to understand the following:

Method

Participants and Setting

The research was conducted at a community college in California during the fall semester of the 1998-1999 academic year. The two ESL courses from which the 22 volunteers were selected were low intermediate ESL courses taught by the researchers. The 22 volunteers who were the focus of the study were classified into two groups. Group 1 (late immigrants) consisted of 13 (59%) students who had at least 10 years of L1 education before arriving in the United States. Group 2 (early immigrants) consisted of nine (41%) students who had fewer than 10 years of education in the United States, specifically, at least a partial secondary education or a complete secondary education, and some of these also had a partial elementary education in the United States. The late immigrant group had a mean age of 33.7 years, with an age range of 20 to 43 years. Their mean number of years in the United States was 7.9 years, with a range of 1.8 to 25 years. The early immigrant group was a more traditional college age, with a mean age of 22.2 years (range 18 to 34 years). The early group had been in the United States an average of 9.8 years, with a range of 5 to 19 years. The late immigrant group included five Spanish, two Hmong, and six Russian speakers. The early immigrant group included two Spanish and seven Hmong speakers.

Both types of ESL adult students wanted to learn more English in order to succeed in academic mainstream courses or to get a better job in the near future. Students were told that the study would be done on a voluntary basis if they wanted to participate. The major language groups included in the study sample—Spanish, Hmong, and Russian—were chosen because these were the three largest language groups in both classes and are also representative of large immigrant groups in the college and in the local area.

The researchers implemented the Academic Language Assessment and Development for Individual Needs (ALADIN) curriculum (Kuehn, 2000) designed to meet the needs of students under-prepared in academic English proficiency. Students were also taught specific strategies to prepare them for L2 academic success.

Data Collection

Academic language pre-test and post-test

The instruments used to measure students' English academic language proficiency developed by Kuehn (1997) were a parallel pre-test and post-test which were used to assess and evaluate the students' achievement and progress in their learning of academic language. Scores cover nine test sub-sections. The pre-test and post-test consisted of both performance-based test tasks and self-assessments. The test sections: (a) lecture note-taking, (b) lecture and short answer question, (c) dictation of lecture sentences, (d) self-assessment of academic vocabulary (dictation), (e) self-assessment of academic vocabulary (reading), (f) reading sentence completion (modified cloze), (g) reading underlining of important information, (h) reading note-taking, and (i) reading short answer question.

Academic language curriculum

The academic language curriculum was designed to help students to prepare for mainstream academic college courses by giving them the foundation they need. There are five sections in each of the three lessons. Each lesson is based on a videotaped lecture and text excerpts from a different academic discipline, including psychology, literature, and business law. Each lesson includes lecture note-taking strategies, academic vocabulary building, reading skills and strategies, sentence complexity strategies, and academic culture. In this research project, the students were taught only the first lesson of the academic language curriculum because of time constraints and the low ESL level of the target student group.

Language use questionnaire

A Language Use Questionnaire was designed, field tested, and administered to measure the students' background information, including their L1 language level, ethnicity, years of L1 education, years of L2 education, years in the United States, self-assessment of their proficiency in L1 and L2, their purpose of learning ESL and English academic language, and their goals for the future.

Interviews

Student interviews were designed and field-tested by the researchers and conducted individually during the semester to investigate the students' use of strategies in both L1 and L2 language activities, evidence of language transfer, their prior EFL and/or ESL learning experiences related to affixes, and their perception of the academic language curriculum.

L1 and L2 writing samples and inter-rater reliability

The students were also asked to write an essay in L1 with the same writing prompt translated into Spanish, Hmong, and Russian. Students also wrote an essay in English (L2) using specific writing prompts (see Carson & Kuehn, 1994). The individual's L1 and L2 academic writing proficiency were measured based on a six-point scoring rubric.

Scoring of the L1 writing samples combined a six-point holistic with five analytical scores measuring the overall writing, organization, content, grammar, and vocabulary of the students (Kroll, 1998). The L1 writing prompts were translated by bilingual teaching professionals whose native languages were Spanish, Hmong, and Russian. A second native-speaker of each target language checked each translation for accuracy. The six point scale rubrics for each of the L1 writing groups were adapted for each language group by the raters. Two native-speaker raters in each language group scored each writing sample.

To ensure the reliability of the agreement between the raters, two methods were used: (a) the correlation between Rater A and Rater B in each language group was determined on each of the five scoring areas (overall writing, organization, content, grammar, and vocabulary), and (b) the percentage of agreement based on the number of pairs of scores that were within one score point difference was determined. The inter-rater reliability was determined to be adequate. For the English writing sample, the researcher and another experienced ESL writing instructor scored the writing using the same rating scale and criteria.

Since the inter-rater reliability for each L1 and L2 writing was verified, the two scores for each of the five areas of the writing were averaged for the purpose of data analysis. Also, a new composite variable for both L1 and L2 writing, the total L1 writing and total L2 writing, was created by combining all the writing subscores to compare the students' L1 and L2 writing in general.

Academic vocabulary translation (L2 to L1)

A 25-word list of academic vocabulary from the first lesson of the curriculum was developed and given to students to translate into their L1 as a measure of their level of L1 academic vocabulary knowledge. The same teachers who scored the L1 essays scored the L1 vocabulary translations.

Results

Research Question 1: Do Early and Late Immigrant Students Vary in Their Educational Backgrounds and L1 Language Proficiency?

Results of the t-tests comparing the two groups (see Table 1) showed that the early and late immigrants were significantly different in their total years of L1 education, total years of L2 education, and in their years of English as a Foreign Language (EFL) education in their own countries, but not significantly different in their years of stay in the United States. This fact can be explained by the age difference between the two groups (11 years), which was significant. The early immigrant group had been in the United States longer than the late immigrant group, but they were also much younger than the ones in the late immigrant group. Early immigrants had significantly less L1 education, significantly more L2 education, and significantly less EFL. On the two L1 language tasks, as expected, late immigrants were significantly more proficient than the early immigrants.

Table 1
Comparison of Background Differences for Late (n = 13) and Early (n = 9) Immigrants: t-test Results

The results suggest that late immigrants have more L1 CALP than did early immigrants, which supports Cummins' threshold hypothesis and provides more conducive condition for late immigrant to transfer L1 CALP in their learning of L2 CALP (academic English).

Research Question 2: What is the Relationship Between the Students' L1 Academic Language Proficiency and the Rate of Development of Their English Academic Language Proficiency?

Although placed in the same ESL level, there were some differences between the early and late immigrant groups on the academic language pre-test tasks. The differences were examined using ANCOVA, controlling for years of EFL and years in the United States, both of which could be contributing to these language differences observed. (ANCOVA is analysis of covariance. It allows the researcher to control for one variable, like age or years in the US, while looking for differences in other variables, like test scores.) As with all the statistical tests done for this research, it should be kept in mind that the small sample sizes of the two groups make it difficult to find statistical significance in the differences observed due to a lack of power.

On the pre-test, the early immigrant group scored significantly higher than the late immigrant group on one listening-related task: lecture note-taking (F = 6.78, p = .018). It should be noted that on the dictation word endings (F = 3.58, p = .075), and lecture question content (F = 3.12, p = .094), the early immigrant group also scored higher than the late immigrant group and the differences approached statistical significance. On two vocabulary-related tasks, however, the late immigrant group scored significantly higher than the early immigrant group: vocabulary self-assessment (F = 6.51, p = .02) and vocabulary dictation (F = 9.60, p = .006). The higher scores on the listening-related pre-test tasks were not unexpected since the early immigrant group had had significantly more L2 instruction and L2 educational experience than the late immigrant group.

On the post-test, however, the late immigrant group closed the gap and scored equally well on the three listening tasks on which they had been behind the early immigrant group on the pre-test. In other words, on the post-test, there were no significant differences between the two groups on the three listening tasks. On the two vocabulary-related test tasks, on which the late group had out-scored the early immigrant group on the pre-test, the late group maintained this significant difference on the post-test. On one test task, the modified cloze test (reading sentence completion), the late immigrant group scored significantly higher than the early group, even though it had scored equally on the pre-test.

In summary, comparing the post-test results to the pre-test results, it is clear that the late and early groups made different progress on different sections of test after instruction in academic language. The late immigrant group made better progress than the early immigrant group as a result of instruction, perhaps gaining an advantage over the early group because of transfer of L1 skills and strategies relating to academic language development (see Jiang, 1999).

In terms of overall gains, both groups made statistically significant gains on the majority of the test sections. The late group made statistically significant gains on all but two sections of the test. The early group made statistically significant gains on all but three sections of the test. The instruction was effective for both groups, but the late group made more significant gains than did the early group, even though both started at relatively similar levels.

The gains made by each groups on the post-test compared to the pre-test were also tested for significance. The gains made by the late group compared to the early group were significantly higher on two sections of the test: lecture note-taking (t = 2.44, p = .024) and cloze sentence completion (t = 2.10, p = .049). On the lecture question content, the late group also scored higher, and the difference approached significance (t = 2.34, p = .076). In contrast, the early immigrant group made significantly more gain than the early group on only one section of the test: the vocabulary dictation (t = -3.09, p = .006). On all other sections of the test, the two groups had similar score gains.

In summary, both the late and the early groups made significant progress on their development of English academic language skills as a result of the academic language instructional intervention. The late group had higher gains on three sections of the academic language test: lecture notes, cloze sentence completion, and lecture question content.

Other Evidence and Implications of Transfer: Correlation Between L1 and L2 Writing and Correlation Between Years of L1 Education and L1 and L2 Writing

The result of the correlations computed for the entire group showed a moderate positive correlation between students' L1 and L2 writing scores (r = .382). Scores in each of the five sub-score areas of the L1 writing also showed moderate positive correlations with the corresponding area of the L2 writing. The result indicated a positive relationship between the general and specific writing areas in L1 to the same areas in L2. A 1990 study also suggested a similar correlation, i.e., that "literacy skills can transfer across languages" (Eisterhold, 1990, p. 95). This conclusion is also supported by Canale, Frenette, and Bellanger (1988), who found positive correlation between L1 and L2 writing, "suggesting a common underlying proficiency in writing ability across languages" (p. 95).

The students' total years of L1 education was positively related to their L1 and L2 writing performance in the five areas, especially in the area of vocabulary (L1 education and L1 writing, r = .571; L1 education and L2 writing, r = .324). The positive correlation between the years of L1 education and the students' L2 writing level was also statistical evidence of transfer, further supporting Cummins' Hypothesis of Common Underlying Proficiency.

Research Question 3: What Role Does Transfer Play in the Development of the English Academic Language Proficiency Among the Early and Late Immigrant Students?

Interview analysis

Student interviews were designed (and field-tested) by the researchers and conducted during the semester to investigate the students' use of strategies in both languages, evidence of transfer, their prior EFL and/or ESL learning experiences related to affixes, and their perception of the academic language curriculum. The following questions were asked:

In comparing responses of the late group and early group to the first question, the percentages showed that more students in the late group reported using context clues (39%) and making inferences in solving L1 new word problems (39%) than students in the early group (11% and 22% respectively). More late immigrants tended to ask someone about a new word (62%) than did the students in the early group (67%), but both groups reported a high use of this strategy.

The students' response to the second question (solving new word problems in L2) showed the same trend with a bigger difference between the two groups. More students in the late group (46%) reported using context clues and making inferences (62%) in solving L2 new word problems than did students in the early group (22% and 33%). Late immigrants tended to ask someone for the answer less often (39%) compared to early immigrants (67%).

The percentage results from the response to the first two questions indicated that students with more L1 education used more cognitive strategies in both L1 and L2 to solve new word problems than students with less or no L1 education. The fact that more students in the late group used strategies and fewer students just asked somebody for the meaning of the new words than in the early group also suggests the transfer of prior cognitive knowledge in students' language learning.

In response to the third question, 62% of the students in the late group said they thought about their L1 grammar when learning English grammar. In contrast, only 22% of the students in the early group had the same answer to the question. In other words, students with more L1 education used their prior knowledge in learning the same domains of knowledge in L2 to facilitate their learning, while the ones with less L1 or no L1 had less or nothing to fall back on in their L2 learning.

In response to the fourth question 85% of the late group and 100% of the early students thought that knowledge of L1 helped them, at least to a certain extent, in learning English. The unanticipated result was that students in the early group realized the importance of L1 education even though they had little or no L1 education. The fact that about 15% of the late students did not think their L1 knowledge helped them learn English might indicate their frustration with the negative transfer from L1 to L2 or interference of L1 in L2 learning. The fact that 13% of the 15% Spanish-speaking students in the class thought that L1 knowledge did not facilitate their English learning. This finding also supported this hypothesis of interference of L1 when the target language was similar to their native language.

In summary, the interview procedures showed that late and early immigrant students reported differences in several areas of language learning strategies. In general, these responses support the other analyses in the study and suggest that transfer of L1 prior knowledge and strategies does occur for students with higher L1 language proficiency.

Conclusion

The statistical tests done on the two groups compared in the study, the late and early immigrants, yielded several significant differences despite the small sample sizes and resulting lack of statistical power.

Difference Between the Two Groups (L1 CALP and BICS)

The t-test results showed that the late and early immigrants were significantly different in their L1 education and L2 education with the former having more years of L1 education and EFL instruction and the latter more years of L2 education. No significant difference was found between the two in their years of stay in the United States. In both the L1 vocabulary translation and the L1 writing exercises, the late immigrants' results were significantly better than those of the early immigrants.

Stotsky's (1983) survey of L1 correlation studies showed positive correlations between "reading achievement and writing ability . . . writing quality and reading experience," and "reading ability and measures of syntactic complexity in writing" (p. 88). In other words, "Better writers tend to be better readers" (Eisterhold, 1990, p. 88). The result of the late immigrant group's significantly higher performance in L1 writing, and significantly higher rating of their L1 reading ability than the early immigrant group, supports Stotsky's findings on the positive correlations between L1 reading and writing.

The results of comparing the mean score of each group from the pre- and post- test on each test task showed that the late group with high L1 CALP made significant gains on all 13 individual tasks on the pre-test, except for two: the reading note-taking and vocabulary dictation. Their gains in these areas approached significance. The early group with low L1 CALP and higher L2 BICS made significant gains on all the individual tasks of the post-test except for three tasks: lecture note-taking, reading note-taking, and the cloze test of sentence completion. However, as reported, the late immigrant group made more progress than the early group in several areas.

These results support the hypothesis that students with higher L1 CALP can transfer skills and strategies that facilitate their development of L2 CALP, despite not having had formal instruction in an L2 educational context. In spite of the significantly more years of education in the United States that the early immigrants had received, the lack of differences between the two groups on the pre-test suggests that their lack of L1 CALP was related to their relatively low L2 CALP. Clearly, the transfer of skills from L1 to L2 can only take place when those skills are already present in L1 (Carson & Kuehn, 1994). The gains made by both groups in academic language were similar, although the late immigrant high L1 CALP group out-performed the early immigrant group in several areas. Finally, the positive correlation between L1 writing and L2 writing further supports the evidence of positive transfer.

Interview findings

The results of the student interviews provided further evidence of the students' use of strategies in both languages, as well as evidence of transfer of their prior EFL and/or ESL knowledge and the way in which they perceived the academic language curriculum. Student responses to the first two questions indicated that more late immigrant students (high L1 CALP) used cognitive strategies in both L1 and L2 to solve new word problems than the early students did. The results suggest that more students with more L1 education and CALP transferred the same strategies in L2 word-solving tasks.

Late immigrant student responses to the third question indicated that they used (transferred) their prior linguistic knowledge in learning the same domains of linguistic knowledge in L2, such as the learning of grammatical features (O'Malley & Chamot, 1990). The early students reported little or no linguistic strategy use to fall back on in their L2 learning.

Student responses to the fifth question may indicate that the knowledge of L1 grammar may cause both positive and negative transfer depending on the learner. More early students thought their knowledge of L1 helped them at least to a certain extent in learning English. The surprising fact was that almost 90% of the students in the early group realized the importance of L1 education even though they had little or no L1 CALP. They may have realized the importance of grammar skills during the process of their education.

In summary, the findings from this study support Cummins' model of Common Underlying Proficiency (CUP), and the Interdependence Hypothesis between L1 and L2 complement previous studies on positive correlation between literacy development in first and second languages (Hoover, 1983; Cummins, 1984; Bossers, 1991; Verhoeven, 1991; Carroll, 1991; Canale, Frenette, & Bellanger 1988; Carson & Kuehn, 1994), and provide both quantitative and qualitative evidence on positive transfer of prior linguistic and cognitive skills from L1 to L2.

Transfer of CALP

The results indicated that students in the late immigrant group were able to out-perform the early immigrants in vocabulary and some reading-related tasks, despite low L2 educational experience. Their performance suggested the transfer of their prior linguistic skills, especially in vocabulary and reading (both declarative and procedural knowledge) which are more related to CALP.

There were differences between late and early students in understanding normal English speech, isolated sentences, and single vocabulary. In other words, early immigrants may have acquired the native speaker's speech pattern due to their longer exposure to the language, but seem to have acquired more L2 BICS than CALP, which may have be related to their lack of L1 education.

Implications for Instruction

The study presented here indicates the importance of the differences in language backgrounds of students who have high and low levels of L1 education and therefore of L1 CALP. If we picture the process of learning academic language as a journey, perhaps by train, students may reach the final destination at different speeds, at different times, and by using different methods to get there. Students in the late immigrant group discussed in this research who have L1 education and prior knowledge may progress faster because of the active transfer that may help to facilitate their learning process and accelerate them on their journey.

Students in the early immigrant group who have limited or no L1 education or prior knowledge may accelerate more slowly. In order to help the latter students to reach their final destination, more intervention and support may be needed from the institution and from teachers. There may also be ways to facilitate their learning and to provide them with faster means of transportation.

The differences between the late immigrant students and the early immigrant students revealed in this study may shed some light on the understanding of the different needs that the two groups have in the development of their English academic language proficiency. Since the late immigrant students had significantly higher L1 education and thus had relatively more cognitive and academic skills to transfer in their learning of English academic language, they can progress with the intervention of academic language curriculum at a faster pace.

In contrast, the early immigrants had very little L1 education with little L1 CALP to facilitate their learning of L2 English academic language. It would take much longer for these students to develop the L2 cognitive and academic skills required for English academic language proficiency. In other words, they need not only to develop English academic language proficiency but also the cognitive and academic skills to help them in language learning and academic study.

The implications of this research and the other studies cited for ESL and remedial programs may be to provide two different kinds of courses: one for students who have had L1 education and have high L1 academic language proficiency and cognitive skills, and the other for students who have had little or no L1 education with little L1 academic proficiency and cognitive skills. In this way, teachers can address the different needs of the two groups and provide more appropriate instruction for each type of student.

For students of the second group, those who have little or no L1 academic and cognitive skills, an implication is that the teaching of cognitive and metacognitive learning strategies, along with academic language, may help these students to be successful, at least in answering the type of comprehension questions required in academic courses. Another implication is that these students need to learn the strategies in the context of developing CALP in the target language. Therefore, teachers who work with ESL students with little or no L1 education should provide instruction in the use of appropriate cognitive and metacognitive strategies in the context of their curriculum.

An implication for the immigrants who have more L1 education, and therefore more background knowledge, is that they are more likely to transfer their cognitive skills and L1 academic language concepts and schema to the learning of the target language. In other words, teachers should understand the advantages that students who have had L1 education have, such as the ability to transfer their prior linguistic knowledge, prior skills, or existing schemata to facilitate their acquisition of the target academic language.

Specifically, such students may transfer their reading and writing knowledge from L1 to L2, including their use of strategies (as discussed earlier). However, sometimes the students may not be aware of the knowledge that they already have because of the challenge of other cultural, social, or psychological factors. Thus the teacher can facilitate the use of this prior knowledge through specific teaching strategies. In other words, transfer may not be automatic and teachers need to provide instruction to activate it through instruction (Eisterhold, p. 98).

If teachers focus on the positive transfer of skills that students have, it may be possible to teach these students ways to activate their prior schemata in the teaching of English vocabulary, reading, and writing in order to help them learn the second language more effectively. For example, in learning new vocabulary, teachers can ask ESL students whose native languages are similar to English if they have similar words in their own language and if they have the same or similar meaning as in English to help students transfer their prior knowledge. For students whose native languages are not similar to English, teachers can also ask the more advanced students in the class to explain the new words in their native language to other students with the same L1 to ensure their understanding of the meaning in English.

Limitations of Study

This study may be limited by the small sample of 22 students who were selected based on three specific language groups. Because of the small sample size, the results may not have generalizability for other students or for other language groups. However, other studies have been done across the United States using the same curriculum and pre/post-test, and a comparison study would be an interesting follow-up for future research.

The study was limited by the lack of performance assessment of students' L1 reading, listening, and speaking proficiency. If there had also been reading, listening, and speaking tests in L1, the students' complete L1 proficiency could provide further validity to the study.

References

Bossers, B. (1991). On thresholds, ceilings and short-circuits. The relation between L1 reading, L2 reading and L2 knowledge. ALLA Review, 45-60.

Canale, M., Frenette, N., & Belanger, M. (1988). Evaluation of minority student writing in first and second language. In J. Fine (Ed.), Second language discourse: A textbook of current research (pp. 147-165). Norwood, NJ: Ablex.

Carson, J. E., & Kuehn, P.A. (1994). Evidence of transfer and loss in developing second language writers. In A. H. Cumming (Ed.), Bilingual performance in reading and writing (pp. 257-281). Ontario Institute for Studies in Education.

Chamot, A. U., & O'Malley, J. M. (1986). A cognitive academic language learning approach: An ESL content-based curriculum. Wheaton, MD: National Clearinghouse for Bilingual Education.

Chamot, A. U., & O'Malley, J. M. (1987). A cognitive academic language learning approach: A bridge to the mainstream. TESOL Quarterly, 21, 227-249.

Chamot, A. U., & O'Malley, J. M. (1994). The CALLA handbook. Reading, MA: Addison-Wesley.

Chomsky, N. (1979). Language and responsibility. New York: St. Martin's Press.

Collier, V. P. (1989). How long? A synthesis of research on academic achievement in a second language. TESOL Quarterly, 23 (3), 509-531.

Corder, S. P. (1994). A role for the mother tongue. In S. M. Gass & L. Selinker (Eds.), Language transfer in language learning (pp.18-31). Philadelphia, PA: John Benjamins North America.

Crandall, J. A. (1993). Content-centered learning in the United States. In W. Grabe, C. Ferguson, R. B. Kaplan, G. R. Tucker, & H. G. Widdowson (Eds.), Annual Review of Applied Linguistics: Issues in second language teaching and learning (Vol. 13, pp. 11-26). New York: Cambridge University Press.

Cummins, J. (1979a, October). Cognitive/Academic language proficiency, linguistic interdependence, the optimum age question, and some other matters. Working papers on bilingualism, 19 (1), 121-129.

Cummins, J. (1979b). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research 34, 46-49.

Cummins, J. (1981). The role of primary language development in promoting educational success for language minority students In California State Department of Education Office of Bilingual Education, Schooling and language minority students: A theoretical framework (pp. 3-49). Sacramento, CA: Author.

Cummins, J. (1984). Bilingualism and special education: Issues in assessment and pedagogy. San Diego, CA: College-Hill Press Inc.

Cummins, J. (1992). Language proficiency, bilingualism, and academic achievement. In P. A. Richard-Amato & M. A. Snow (Eds.), The multicultural classroom: Readings for content-area teachers (pp.16-26). White Plains, NY: Longman Publishers USA.

Cummins, J. (1994). In R. A. DeVillar, C. J. Faltis, & J. P. Cummins (Eds.), Cultural diversity in schools: From rhetoric to practice (pp. 360-390). Albany, New York: State University of New York Press.

Cummins, J. (1996). Negotiating identities: Education for empowerment in a diverse society. Ontario, CA: California Association for Bilingual Education.

Eisterhold, J. C. (1990). Reading-writing connections: Toward a description for second language learners. In B. Kroll (Ed.), Second language writing: Research insights for the classroom (pp. 88-101). New York: Cambridge University Press.

Faerch, C., & Kasper, G. (1987). Introspection in second language research. Cleveland, OH: Multilingual Matters LTD.

Flowerdew, J. (1994). Academic listening. Cambridge, MA: Cambridge Press.

Freeman, D. E., & Freeman, Y. S. (1994). Between worlds. Portsmouth, NH: Heinemann.

Gass, S. M., & Selinker, L. (Eds.). (1994). Language transfer in language learning. Philadelphia, PA: John Benjamins North America.

Hoover, W. (1983). Reading: A universal process. In S. Hudelson (Ed.), Learning to read in different languages (pp. 27-31). Washington, DC: Center for Applied Linguistics.

Jiang, B. (1999). Transfer in the academic language development of post-secondary ESL students. Unpublished doctoral dissertation, University of California, Davis, and California State University, Fresno.

Kasper, L. F. (1998). Meeting ESL students' academic needs through discipline-based instructional programs. In T. Smoke (Ed.), Adult ESL: Politics, pedagogy, and participation in classroom and community programs (pp.147-157). Mahwah, NJ: Lawrence Erlbaum Associate.

Kroll, B. (1998). Assessing writing abilities. Annual Review of Applied Linguistics, 18, 219-242.

Kuehn, P. A. (1997). Academic language: Assessment and development of individual needs (ALADIN). Unpublished manuscript.

Kuehn, P. (2000). Academic language assessment and development of individual needs. Boston: Pearson.

Lakshmanan, U. (1994). Universal grammar in child second language acquisition. Amsterdam: John Benjamin Publishing Co.

Odlin, T. (1986). Language transfer: Cross-linguistic influence in language learning. New York: Cambridge University Press.

O'Malley, J. M., & Chamot, A. U. (1990). Learning strategies in second language acquisition. New York: Cambridge University Press.

Saville-Troike, M. (1984). What really matters in second language learning for academic achievement? TESOL Quarterly, 18, 199-219.

Saville-Troike, M. (1991). Teaching and testing for academic achievement. Washington, DC: National Clearing House for Bilingual Education.

Solomon, J., & Rhodes, N. (1995). Conceptualizing academic language. Washington, DC: Center for Applied Linguistics, The National Center for Research on Cultural Diversity and Second Language Learning (pp. 1-11).

Spolsky, B. (1989). Conditions for second language learning. Oxford, UK: Oxford University Press.

Stotsky, S. (1979). Teaching the vocabulary of academic discourse. Journal of Basic Writing, 2 (3), 15-39.

Stotsky, S. (1983). Research on reading/writing relationships: A synthesis and suggested directions. Language Arts, 60, 627-642.

Swain, M., Lapkin, S., & Barik, H. C. (l976). The cloze test as a measure of second language proficiency. Working papers on bilingualism, 11, 32-42.

U.S. Census Bureau. (1995). The foreign-born population: 1994. Washington, DC: U.S. Government Printing Office.

Valdez, Pierce, L., & J. M. O'Malley. (1992). Performance and portfolio assessment for language minority students. Washington, DC: National Clearinghouse for Bilingual Education.

Verhoeven, L. (1991). Acquisition of biliteracy. In J. H. Hulstin & J. F. Matter (Eds.), Reading in two languages (Vol. 8, pp. 6-74). Amsterdam: AILA.

Wang, L. S. (1996). Causal modeling research on language minorities achievement. New York: Peter Lang Publishing, Inc.

Wright, E., & Kuehn, P. A. (1998, Spring, Summer). The effects of academic language instruction on college bound at risk secondary students. Journal of Educational Opportunity, 17 (1), 9-22.

Appendix A: Interview Questions

Appendix B: L1 and L2 Writing Prompts and Scoring Scales

L1 Writing Prompt Translated into Hmong, Russian, and Spanish

As certain professions become more crowded, students of today are faced with a difficult choice. Should they choose a field they wish to enter even though job prospects are limited? Or should they choose a field in which they are less interested but that offers more job opportunities? Which decision would you make? Give reasons for your answer.

L2 (English) Writing Prompt

Some people believe that it is important for them to be part of a group in order to achieve their goals. Others believe that they should try to be unique individuals in order to reach their objectives. Which position do you prefer? Give examples to support your choice.

The writing prompts are from Carson and Kuehn 1994, with the author's permission.