Jim Cummins, Minoo Ardeshiri, Sarah Cohen
In a knowledge-based global economy, the literacy level of the population is increasingly seen by policy-makers as crucial to economic growth. The Canadian Education Association (2004), for example, notes that, "a 1% increase in adult literacy produces a permanent 1.5% increase in the gross domestic product. For Canada, that would result in a permanent increase of approximately $18 billion dollars per year" (p. 2). Within this competitive global economic context, schools in countries around the world are under increasing scrutiny to ensure that they develop high levels of literacy skills among the student population.
The complexity of the pedagogical challenge faced by schools in attempting to fulfill this mandate is immediately obvious in relation to two dimensions of contemporary economically advanced societies. First, as a result of global population movements, these societies are becoming increasingly diverse, with many students coming from home backgrounds other than that of the dominant language(s) of the schools and society. For example, in the Canadian cities of Toronto and Vancouver, about 50 percent of the student populations come from language backgrounds other than English or French (the two Canadian official languages). As discussed below, these students typically require at least five years to attain grade expectations in literacy-related aspects of language proficiency. Thus, during this extended period, they are attempting to access curricular texts that are likely to be considerably beyond their current level of proficiency in the language of instruction with respect to vocabulary, grammatical constructions, and discourse organization.
The second dimension of contemporary societies that contributes to the pedagogical challenge faced by schools involves the rapid technological changes that have undermined the dominance of traditional notions of literacy, understood as print-based reading and writing skills, in the lives of a growing number of students. Outside of school, electronic literacies increasingly mediate the interpersonal interactions and learning activities of adolescents, with the result that less time and inclination are available to pursue traditional print-based reading and writing activities (see Warschauer, this issue). Schools, however, remain largely focused on traditional literacy activities, partly because accountability systems typically assess only these dimensions of literacy.
The gap between electronic and hard copy texts with respect to interactivity is enormous. In electronic environments, words and phrases anywhere in the text can be located within seconds, the appearance of the text (e.g., font size, color) can be changed at will, and a variety of text comprehension resources (e.g., dictionaries, thesauri, and web-based visual, auditory, and print information) can be accessed quickly and efficiently. Additionally, the dynamic nature of electronic text can be enhanced by supports that have been specifically programmed to increase reading comprehension or vocabulary acquisition. For example, The Thinking Reader (Rose & Dalton, 2002) provides meaning and reading strategy supports to readers at various stages of their reading of the text.
None of these supports are available in hard-copy textual environments. The text is static, linear, and non-interactive. Thus, schools are faced with the pedagogical challenge of teaching traditional forms of literacy at a time when students have grown accustomed to the dynamic interactivity of electronic text, the immediacy of hyper-linked supports and multi-modal presentation, and the collaborative possibilities afforded by networked communication.
These pedagogical challenges are taken up in the present paper by examining the extent to which some of the dynamic features of electronic environments are capable of supporting traditional forms of literacy development among culturally and linguistically diverse students. We report on a study that investigated the effectiveness of certain features of electronic environments in scaffolding students' access to meaning and their learning of new vocabulary. The context of our inquiry is the challenge that many students experience, particularly after the early grades of elementary school, in reading texts that are becoming increasingly complex with respect to both language and conceptual content. This challenge is particularly acute for immigrant students who are learning the language of instruction at the same time they are attempting to access curriculum content which is expressed in that language. If effective dynamic supports could be incorporated into electronic texts, access to curriculum content and the development of academic language proficiency might be achieved much more rapidly by ELL students than is presently the case.
The nature of the academic challenge faced by ELL and other struggling students, and the potential afforded by scaffolding text in electronic environments, is elaborated below with reference to the nature of academic language and the importance of extensive reading for students' overall educational development.
Academic language can be defined operationally for educational purposes (grades K-12) as the totality of the vocabulary, grammatical constructions, and discourse conventions (e.g., paragraph formation) that students are exposed to and expected to learn between Kindergarten and grade 12. Thus, it reflects the language students encounter in the context of schooling and which they need to master in order to succeed in school. Academic language proficiency reflects the extent to which an individual has gained access to and command of the oral and written linguistic registers of schooling. In other words, it refers to students' ability to understand and express, in both oral and written modes, concepts and ideas that are relevant to success in school. Valdes (2004) summarizes the conception of academic language embedded in the TESOL Standards document (TESOL, 1997):
In the Standards document, for example, we are told that to achieve academically students will use English to follow oral and written directions both implicitly and explicitly, request and provide clarification, request information and assistance, explain actions, negotiate and manage interactions, and ask and answer questions. They will also use English to obtain, process, construct, and provide subject matter information in written form. They will retell information, compare and contrast information, persuade, argue, and justify, analyse, synthesise and infer from information. They will also hypothesize and predict, understand and produce technical vocabulary and text features according to the content area. (2004, p. 121)
Thus academic language proficiency includes knowledge of the less frequent vocabulary of English that is relevant to schooling as well as the ability to interpret and produce increasingly complex written and oral language. As students progress through the grades, they encounter far more low frequency words (primarily from Greek and Latin sources), complex syntax (e.g. passives), and abstract expressions that are virtually never heard in everyday conversation. Students are required to understand linguistically and conceptually demanding texts in the content areas (e.g. literature, social studies, science, mathematics) and to use this language in an accurate and coherent way in their own writing.
Acquiring academic language is challenging for all students. For example, schools spend at least 12 years trying to extend the conversational language that native-speaking children bring to school into these more complex academic language spheres. It is hardly surprising, therefore, that research has repeatedly shown that immigrant students, on average, require at least five years of instructional exposure to the school language to come close to native-speaker norms (Cummins, 1981; Hakuta, Butler & Witt, 1999; Klesmer, 1994; Shohamy, Levine, Spolsky, Kere-Levy, Inbar, Shemesh, 2002). Many ELL students who have acquired conversational fluency and decoding skills in English are still a long way from grade-level performance in academic language proficiency (e.g., reading comprehension) (Cummins, 2000, 2001).
Conversational and academic language registers represent subsets of what Gee (1990) has termed primary and secondary discourses. Primary discourses are acquired through face-to-face interactions in the home and represent the language of initial socialization. Secondary discourses are acquired in social institutions beyond the family (e.g., school, business, religious and cultural contexts) and involve acquisition of specialized vocabulary and functions of language appropriate to those settings. Secondary discourses can be oral or written and are central to the social life of both non-literate and literate cultures. Examples of secondary discourses common in many non-literate cultures are the conventions of storytelling or the language of marriage or burial rituals which are passed down through oral tradition from one generation to the next. Thus, the academic language of schooling represents just one among many specialized secondary discourses. However, unlike many other secondary discourses (e.g., the language of horticulture), the development of expertise in understanding and using academic language determines the life chances of students in literate societies around the world.
Because academic language is found almost exclusively in written texts, we encounter this language primarily when we read. Therefore, reading extensively is of crucial importance for the expansion of vocabulary and the growth of reading comprehension abilities in both L1 and L2. There is also evidence that reading by itself is less powerful than when it is combined with an instructional focus on comprehension strategies and vocabulary consolidation (e.g., Laufer, 2003; Paribakht & Wesche, 1997; Pressley, Duke, & Boling, 2004).
Large-scale research has consistently shown a strong relationship between the amount that students read and their reading achievement (see Guthrie, 2004, Krashen, 2004, McQuillan, 1998, for reviews). The overall pattern of relationships can be illustrated in relation to the PISA study that investigated the reading performance of 15-year olds in more than 30 countries (Organization for Economic Cooperation and Development [OECD], 2004).The report points out:
The PISA results underline the importance of student engagement. For example, students who are habitual readers and who enjoy reading are also more likely than others to have high levels of reading literacy. Greater engagement in literacy can be a consequence, as well as a cause, of higher reading skill, but the evidence suggests that these two factors are mutually reinforcing. Indeed, the level of a student's reading engagement is a better predictor of literacy performance than his or her socioeconomic background, indicating that cultivating a student's interest in reading can help overcome home disadvantages. (2004, p. 8)
Guthrie (2004) has similarly drawn attention to the centrality of literacy engagement for reading achievement. Drawing on both the 1998 NAEP data from the United States and the results of the PISA study of reading achievement in international contexts, he notes that students whose family background was characterized by low income and low education, but who were highly engaged readers, substantially outscored students who came from backgrounds with higher education and higher income, but who themselves were less engaged readers. Based on a massive sample, this finding suggests the stunning conclusion that engaged reading can overcome traditional barriers to reading achievement, including gender, parental education, and income. (p. 5)
Guthrie notes that the term engagement incorporates notions of time on task (reading extensively), affect (enthusiasm and enjoyment of literacy), depth of cognitive processing (strategies to deepen comprehension), and active pursuit of literacy activities (amount and diversity of literacy practices in and out of school).
Two of the components of Guthrie's notion of reading engagement, namely affect and depth of cognitive processing, are also evident in the construct of involvement load proposed by Laufer and Hulstijn (2001) as a key factor in incidental L2 vocabulary acquisition. Their thesis is that "retention of hitherto unfamiliar words is claimed to be conditional upon the amount of involvement while processing these words" (p. 1). Involvement is proposed as a motivational-cognitive construct comprised of need, search, and evaluation. Need is defined as the learner's "drive to comply with task requirements" (p. 14); search is the attempt to find the meaning of an unknown word by consulting a dictionary or another authority (e.g., a teacher), while evaluation is the process of deciding on the appropriateness of the word for a particular context. Involvement load is the extent to which a particular task induces the motivational/affective dimension of need, and the cognitive processing dimensions of search, and evaluation. According to Laufer and Hulstijn, "[o]ther factors being equal, words which are processed with higher involvement load will be retained better than words which are processed with lower involvement load" (p. 15). In other words, the involvement load hypothesis proposes that the meaning of words will be retained better when the learner is strongly motivated to complete the task (e.g., read a passage) and commits more cognitive energy into processing unfamiliar words. Their reference to "other factors being equal" acknowledges that the involvement load hypothesis will operate only within certain limits of task difficulty." If the task is too far beyond the learner's current L2 proficiency, frustration will ensue and both need and the impulse to search and evaluate will likely dissipate.
The constructs of reading engagement and involvement load focus attention on the challenges for immigrant students who are trying to gain access to the curriculum and catch up academically to their peers. How can we promote reading engagement and optimize involvement load during the period of five or more years that ELL students are in the process of catching up to grade expectations? Reading engagement is likely to be compromised by students' limited English academic proficiency which reduces their access to grade-level academic text and potentially increases the frustration associated with trying to read text that is frequently far beyond their current proficiency level. Under these conditions, many students may experience involvement overload and withdraw from academic effort. Text that is presented in a static hard-copy environment provides few opportunities to customize support for the needs of ELL students. However, if we think of the possibilities of scaffolding academic text that is presented in electronic form, far greater opportunities are available to modify the involvement load such that frustration does not undermine need and the cognitive elaboration processes of search and evaluation are enabled to operate even among students whose knowledge of the target language is limited.
The e-Lective Language Learning computer program (Chascas & Cummins, 2005), which provided the electronic environment for the present study, was developed in order to support learners in gaining access to academic texts that are beyond their current level of target language proficiency. Although the involvement load hypothesis was not considered in developing the program, it provides a useful framework both for discussing the scaffolds build into the e-Lective program and interpreting the outcomes of the present study.
The e-Lective Language Learning program (Chascas & Cummins, 2005) attempts to enable ELL students (and other learners) to extend their knowledge of academic language by means of extensive reading. Any text in electronic form can be imported into the program and students can get rapid access to L1 and English dictionary support to facilitate understanding of the meaning of individual words and sentences. In addition, the program "remembers" the words that each individual student has looked up (unknown words) and provides individualized practice to students to assist them in learning this vocabulary. These practice exercises employ several varieties of receptive and productive cloze procedure and can be set at five levels of difficulty. For example, the Blank Words procedure blanks out the words the student looked up and then students select the correct word from a set of alternatives. The difficulty level, which can be adjusted by teachers or students, determines the number of words from which students select. The Partial Words procedure blanks out part of the unknown word and students type the entire word into a pop-up box. The program gives immediate feedback as to the correctness of the choice of word or spelling of the word. The difficulty scale determines the proportion of letters (ranging from 90 to 10%) that are deleted from the word. The default difficulty level, used in the present study, is 3 which blanks out 50% of the target word. More information on the e-Lective program can be obtained at www.e-Lective.net.
The study was conducted in a highly diverse K-5 elementary school whose 690 students represented more than 30 different home language backgrounds. The e-Lective program was piloted in the school in fall 2004 and winter 2005 with individual students in grades 4 and 5 as a form of tutoring for those with identified with language and literacy needs. From February through May 2005 all of the grade 5 students were introduced to the program and used it over the course of several weeks in the computer lab. We termed this experience the Big Words for Big Minds project and most students appeared to be highly motivated to use the program to learn new vocabulary encountered in the texts they read.
Prior to the quantitive study, grade 5 students were surveyed for their opinions on using the e-Lective program. Seventy-five students participated. With respect to overall attitude, 88% of students gave the e-Lective program a rating of 3 or higher when they were asked to rate the e-Lective program on a scale of 1-5 (with 5 being the highest). Comments by the students included:
The quantitative study involved 46 grade 5 students who were familiar with the e-Lective program as a result of their use of it during the previous three months. Sixty-one percent of the participants spoke two languages, 19.5% three languages, and 19.5% spoke only English. Approximately half of the participants indicated that they could read and write in their home language.
Four reading texts were adopted from Aesop's fables. Each text was edited to be a similar length (192- 207 words) and to contain a similar number of low frequency words (between 16 and 19, assessed using the Word Frequency identifier built into the e-Lective program). The reading texts also included an illustration in order to be more engaging for students. The materials were imported into the e-Lective Language Learning program and were also prepared in a photocopied version. Nine low-frequency vocabulary items from each text were identified as the target words, for a total of 36 words in the four texts. Students were randomly assigned to the texts with the constraint that each student would read two texts in each condition (electronic and hard-copy). For example, if student 1 was administered fable A and B using e-Lective, this student then read fable C and D in the hard-copy condition on the following occasion. Thus all students read all four of the fables, two in each condition. There was a three-day gap between each experimental condition. The target words in each of the fables are shown in Table I.
|
The experimental procedure consisted of four stages: (a) pre-test, (b) reading the texts in two sessions followed by an immediate test of the nine target words in each text, (c) post-test administered two weeks after the pre-test, and (d) delayed post-test administered three weeks after the pre-test. A more detailed description of each of the stages follows below.
Pre-test of target words.The Vocabulary Levels Test (VLT) (Nation, 2001; Schmitt, Schmitt, and Clapham, 2001) format was utilized in constructing the pre- and post-tests designed to assess students' knowledge of the 36 target words from the four texts. The procedures recommended by Schmitt, Schmitt, and Clapham (2001) were followed in constructing the tests. Twenty minutes were allotted for participants to take the pre-test. The pre-test provided a baseline against which to interpret students' immediate recall of the nine items in each of the Fables Tests and also the words retained in the post-test and the delayed post-test. The format of the items is illustrated in the example below:
|
The students read the fables with a focus on learning vocabulary that was unfamiliar to them. They were given 30 minutes to read two fables, look up unknown words, carry out practice exercises in the electronic condition if they chose to do so, and respond to the questions on the Fables Tests. The major differences between the two conditions were in how students searched for the meaning of unknown words and in how they attempted to remember the meanings of these words. The hard-copy group was instructed to read the texts using the photocopied version, to write down all the words whose meanings they were unsure of, and look up the unknown words in a conventional dictionary. The electronic text group read the texts on the computer and looked up unknown words using the Babylon English/English on-line dictionary (see www.babylon.com). Access to word meanings was extremely rapid, requiring only that students place the cursor on the target word and click Alt + right mouse button to bring up the dictionary. Students also had the option to practice the unknown words utilizing the Blank Words and Partial Words exercises. After reading each fable, students were given a vocabulary test in multiple-choice format (the Fables Test) that required students to read the target words in a sentence and choose the closest synonym from a list.Our purpose here was to assess students' immediate recall of the nine target words in each fable. An example of the test item format is provided below:
The immediate test after reading each fable also served to motivate students to remember the unknown words. Students were already highly motivated as a result of participating in the Big Words for Big Minds project over a period of several weeks during which time they had used the e-Lective program. Our observations suggested that students took the task very seriously and were highly motivated to perform well on the various tests.
The post-test was designed to measure how much vocabulary had been learned and retained during the intervention. The post-test was administered two weeks after the pre-test. The same procedure as the pre-test was followed; however the order of items within the test as a whole and the order of words within items were changed. Again, 20 minutes were allotted for the post-test.
A delayed post-test was performed one week following the post-test in order to determine the retention of vocabulary over time. The same procedure was followed as for the post-test.
In the electronic condition (mean = 7.40, SD = 4.00) students looked up significantly more words than in the hard-copy condition (mean = 5.40 words, SD = 3.1) (t = 3.5, p<.001). Overall, for the entire sample, students gained an average of 6 words between the pre-test and the post-test (mean pre-test = 20.23, SD = 7.43 versus mean post-test?= 26.64, SD = 7.23, t=9.8, p<.001). As indicated in Figure 1, the delayed post-test showed a drop in performance of only .86 from the time of the post-test, indicating that students retained most of the vocabulary they had learned despite an additional week since the actual reading of the fables.
Although both treatments registered significant gains in vocabulary that persisted in the delayed post-test, no significant differences were observed between the electronic and hard-copy conditions on any of the comparisons (immediate Fables Tests, Post-test, Delayed Post-test). The means and standard deviations for the Fables Tests (maximum score 18) are shown in Table II. The relatively high scores on this test reflect the fact that it was administered for each fable immediately after the student had read the fable and looked up any unknown words. The differences between electronic and hard-copy conditions on the pre-test, post-test and delayed post-test are depicted by bar graphs in Figure 2 (for the total sample) and Figure 3 (individual students) and the means and standard deviations are shown in Table III. For these analyses, N = 40 due to the exclusion of students who were absent for one or more tests.
Figure 2. Comparison of electronic and hard-copy conditions on the pre-test, post-test and delayed post-test.
Figure 3.Performance of individual students on pre-test, post-test, and delayed post-test
|
|
An additional analysis was carried out to assess the extent to which students learned words that they did not know on the pre-test. As shown in Table IV, Student A did not know six out of the 18 target words from the two fables which he or she read in the electronic condition. This student did not know four of the 18 words that appeared in the fables which he or she read in the hard-copy condition. Student A demonstrated knowledge of all but one of these previously unknown words (seize) on the post-test (electronic condition) (= 83.33%) and in the hard-copy condition he or she learned 100% of the words that were previously unknown. On the delayed post-test he or she demonstrated knowledge of all but one of these words in each condition (seize in electronic and vanity in hard-copy conditions). The percentages of previously unknown words that were learned were calculated for both conditions and are shown in Figure 4.
| |||||||||||||||
This analysis revealed a high level of retention of newly learned vocabulary in both conditions. In the electronic condition on the post-test, students learned 66.4 percent of the words that they had not known on the pre-test. In the hard-copy condition, the figure was 68.9 percent. In the delayed post-test, students retained an average of 60 percent in the electronic and 64.5 percent in the hard-copy condition. A paired-samples t test showed that these differences were non-significant.
The three major findings of the present study are as follows: (a) students looked up significantly more words in the electronic condition than in the hard-copy condition; (b) in both electronic and hard-copy conditions, students learned and retained an impressive amount of vocabulary (60-69% of previously unknown words) on the basis of a relatively short exposure to these words in the context of reading four fables; and (c) electronic and hard-copy conditions were equally effective in supporting students' acquisition and retention of vocabulary.
These findings suggest that the greater ease of access to meaning support within the electronic environment might have been offset by the more intensive focus on meaning within the hard-copy environment. Students in the hard-copy environment had to write down the words they didn't know and then make the effort to look up these words in a conventional dictionary, whereas in the electronic condition access to meaning support required much less effort (and perhaps less cognitive processing). In addition, the practice exercises students carried out in the electronic condition did not require an intensive focus on meaning or an in-depth evaluation of the appropriateness of the words for the particular textual context. Blank Words required students to recognize and choose the correct word from a drop-down list, while Partial Words focused on correct spelling of the target word. In the hard-copy condition, students did not carry out any practice exercises and thus they may have had more time to focus on the words they had looked up and evaluate the relationship of the target word to the various synonyms or explanations provided in the dictionary.
The pedagogical implications of these findings can be considered in relation to Laufer and Hulstijn's (2001) involvement load hypothesis. Specifically, both conditions appear to be equivalent with respect to students' need or drive to acquire the new vocabulary but the hard-copy condition may have generated a greater involvement load with respect to the cognitive processes of search and evaluation that offset the advantages of the electronic environment with respect to speed of access to meaning support. Paradoxically, within certain limits, learning may be enhanced by the greater cognitive effort required to access meaning in hard copy conditions.
However, there is a delicate balance between involvement load and involvement overload. Students are unlikely to sustain the motivation to continue to look up words within a hard-copy environment for extended periods. The short period of time that was involved in each condition in the present study, together with students' high motivation, enabled them to focus efficiently on the task at hand.Over the longer-term, however, the ease of computer-supported access to dictionaries and the individualized scaffolding and reinforcement provided by practice exercises may represent a more effective environment for sustained engagement with academic text. In terms of the involvement load hypothesis, students' drive to gain access to meaning and complete the reading task is likely to dissipate rapidly when the search and evaluation conditions become too arduous. When this happens, involvement load becomes involvement overload and frustration undermines task completion.
In short, the findings of the present study suggest that electronic and hard copy textual environments can play complementary roles in supporting students' comprehension of text and expansion of their academic language proficiency. Scaffolded electronic environments can provide effective supports for reading comprehension and vocabulary expansion that are not available within hard-copy environments. Thus, ELL students and struggling readers can be enabled to read grade-appropriate texts that would otherwise have been beyond their comprehension level or zone of proximal development. Teachers who are already using predominantly paper-and-pencil supports for academic language development should therefore be encouraged to explore the possibilities of electronic environments that can scaffold difficult texts and reduce the cognitive and linguistic overload experienced by many ELL (and other) students as they struggle with grade-level hard-copy texts. In view of the well-documented problems of integrating computer technology into school systems (particularly in low-income areas), reliable access to computers in many schools will remain sporadic for some time to come (Cuban, 2001; Warschauer, Noble, & Stone, 2004). Thus, hard-copy texts will remain the norm in most schools. Under these conditions, schools need to decide what kinds of access to technology are most fruitful for under-achieving or ELL students and plan accordingly. The findings of the present study suggest that access to scaffolded text in electronic environments should be seriously considered as a potentially effective tool for complementing regular classroom instruction and promoting reading comprehension and vocabulary growth.
However, the findings of the present study also suggest that paper-and-pencil activities can usefully complement technology-supported reading. The fact that the hard-copy condition produced vocabulary growth that was equal to the electronic condition suggests that paper-and-pencil activities (e.g., keeping a vocabulary notebook or creating a word bank) can provide additional rehearsal and elaboration opportunities to complement the cognitive processing that is supported in an electronic environment. This multi-modal approach would likely increase the involvement load in productive ways and ensure that students are focused on meaning more than they might be within the electronic environment alone.
Thus, the question posed at the beginning of this paper regarding the relative efficacy of electronic and hard copy textual environments needs to be reformulated. Rather than pose the question in an oppositional way, it appears more productive to us at this point to ask how electronic and hard copy textual environments might complement each other in supporting students' academic growth. As indicated by the PISA results (OECD, 2004), the challenge is to promote sustained literacy engagement among students, particularly those who face challenges in accessing grade-level texts, and both electronic and hard-copy textual environments have important and complementary roles in play in this process. Because there is minimal use of scaffolded electronic textual environments in schools at this point, this conclusion implies that schools and other educational institutions should actively explore the potential of this form of curricular access for students who otherwise would be excluded from academic participation.
Canadian Education Association (2004). Policy brief: The promise and problem of literacy for Canada: An agenda for action. Toronto: Canadian Education Association.
Chascas, S. & Cummins, J. (2005). e-Lective Language Learning. (Computer program).
Cuban, L. (2001). Oversold and underused: Computers in the schools. Cambridge, MA: Harvard University Press.
Cummins, J. (1981). Age on arrival and immigrant second language learning in Canada: A reassessment. Applied Linguistics, 1, 132-149.
Cummins, J. (2000). Language, power, and pedagogy: Bilingual children in the crossfire. Clevedon, England: Multilingual Matters.
Cummins, J. (2001). Negotiating identities: Education for empowerment in a diverse society. 2nd edition. Los Angeles: California Association for Bilingual Education.
Gee, J.P. (1990). Social linguistics and literacies: Ideologies in discourses. New York: Falmer Press.
Guthrie, J.T. (2004). Teaching for literacy engagement. Journal of Literacy Research, 36, 1-30.
Hakuta, K., Butler, Y.G., & Witt, D. (2000). How long does it take English learners to attain proficiency? Santa Barbara: University of California Linguistic Minority Research Institute.
Klesmer, H. (1994). Assessment and teacher perceptions of ESL student achievement. English Quarterly, 26(3), 8-11.
Krashen, S.D. (2004). The power of reading: Insights from the research. 2nd edition. Portsmouth, NH: Heinemann.
Laufer, B. (2003). Vocabulary acquisition in a second langauge: Do learners really acquire most vocabulary by reading? Some empirical evidence. The Canadian Modern Language Review. 59, 567-587.
Laufer, B. & Hulstijn, J. (2001). Incidental vocabulary acquisition in a second language: The construct of task-induced involvement. Applied Linguistics, 22, 1-26.
McQuillan, J. (1998). The literacy crisis: False claims, real solutions. Portsmouth, NH: Heinemann.
Nation, I.S.P.(2001). Learning vocabulary in another language. Cambridge: Cambridge University Press.
Organization for Economic Cooperation and Development (2004b). Messages from PISA 2000. Paris: Organization for Economic Cooperation and Development.
Paribakht, T.S., & Wesche, M. (1997). Vocabulary enhancement activities and reading for meaning in second language vocabulary development. In J. Coady & T. Huckin (Eds.), Second language vocabulary acquisition: A rationale for pedagogy. (pp. 174-200). New York: Cambridge University Press.
Pressley, M., Duke, N.K., & Boling, E.C. (2004). The educational science and scientifically-based instruction we need: Lessons from reading research and policy making. Harvard Educational Review, 74, 30-61.
Rose, D. & Dalton, B. (2002). The brain and reading comprehension. In M. Pressley, L. Gambrell & C.C. Block (Eds.) Comprehension instruction: Building on the past and improving instruction for today's students (pp. 257-274). San Francisco: Jossey Bass Publishers.
Schmitt, N., Schmitt, D. & Clapham, C. (2001). Developing and exploring the behavior of two new versions of the Vocabulary Levels Test. Language Testing, 18, 55-88.
Shohamy, E., Levine, T., Spolsky, B., Kere-Levy, M. Inbar, O., Shemesh, M. (2002). The academic achievements of immigrant children from the former USSR and Ethiopia. Report (in Hebrew) submitted to the Ministry of Education, Israel.
Valdes, G. (2004). Between support and marginalization. The development of academic language in linguistic minority children. International Journal of Bilingual Education and Bilingualism, 7, 102-132.
Warschauer, M., Knobel, M. & Stone, M. (2004). Technology and equity in schooling: Deconstructing the digital divide. Educational Policy 18, pp. 562-588.