A growing number of adolescents and adults in the United States use a language other than English at home and require support to develop spoken and written English. In the United States, of the 280.8 million people ages 5 and older, 55 million (19.6 percent) speak a language other than English at home (U.S. Census Bureau, 2005-2009, American Community Survey). More than 18 percent of those who speak a language other than English at home are below the poverty level (versus 11.6 percent of those who speak only English at home), and 31.2 percent have less than a high school education (versus 11.7 percent of English only speakers). The percentage of those without a high school education is higher among those who speak Spanish or Spanish Creole at home (more than 41 percent).
According to the National Assessment of Adult Literacy (NAAL) (Kutner et al., 2007), which in 2003 assessed the literacy of native- and foreign-born adults living in the United States, approximately 11 million adults (5 percent of the U.S. population) were estimated to be nonliterate in English (though not necessarily in their first language) and so lacked sufficient English language proficiency to be assessed in English (Kutner et al., 2007). Among those with some English proficiency, the percentage of Hispanics with below average English prose and document literacy increased from 1992 to 2003.
English language learners are the largest group enrolled in adult education programs, with 43 percent of adult learners enrolled in English as a second language (ESL) programs in the 2001-2002 program year (Tamassia et al., 2007). In the 2006-2007 program year, more than 1 million adults were enrolled in ESL programs that were part of state-administered, fed-
erally funded adult education programs. This figure is likely to be an underestimate because it does not include nonnative speakers in adult basic education and adult secondary education (general educational development [GED]) classes or in ESL classes offered by private organizations.
The adults who participate in ESL classes are diverse in terms of languages spoken, education levels, literacy skill in the first language, and knowledge of English (Burt, Peyton, and Adams, 2003). Some are highly educated in their home countries and have strong academic backgrounds; others are recent immigrants with low levels of education and first language literacy. The numbers of adults in ESL classes who have limited education in their home countries continues to grow (Center for Applied Linguistics, 2010; Condelli, Wrigley and Yoon, 2009; Purcell-Gates et al., 2002; Strucker and Davidson, 2003). Other adults are born in the United States or came to the United States as young children but have grown up with a home language other than English (Tamassia et al., 2007). Though educated in U.S. schools, these adults can be unprepared for work and higher education (Burt, Peyton and Adams, 2003; Thonus 2003; Wrigley et al., 2009), and many drop out before completing high school.
Despite the need for English language and literacy instruction, adult ESL programs have had limited success. A 7-year longitudinal study of noncredit ESL classes showed that only about 8 percent of more than 38,000 learners made the transition to other academic (credit) studies (Spurling, Seymour, and Chisman, 2008). In fact, 44 percent advanced only one literacy level, as defined by the U.S. Department of Education’s National Reporting System for adult literacy programs. Persistence was also an issue. Half of the learners who did not advance attended fewer than 50 hours of instruction. Most of those who advanced received 50 or more hours of instruction, taking on average 50 to 149 hours of attendance (usually referred to as “100 instruction hours”) to advance one level.
This chapter has four parts. Part one presents a brief orienting discussion of the component skills of English learners. Part two summarizes research on the various factors (cognitive, linguistic, social, affective, and cultural) that influence the development of literacy in a second language. Part three identifies practices to develop language and literacy instruction that warrant application and further study with adults developing their English language and literacy skills outside school. The available research does not allow for conclusions about effective approaches to literacy instruction. Thus, the chapter concludes with a summary and discussion of priorities for research to develop effective approaches to instruction for this population.
In this chapter, we draw on several recent systematic reviews of research on effective instructional practices for English language learners, augmented with targeted searches to update or expand on previous find-
ings. The available research is quite limited. In their study of “what works” for English language learners in adult literacy education, Condelli and Wrigley (2004) identified only one study of ESL students that measured a literacy outcome and included a design without confounds. Similarly, Torgerson and colleagues (2004) examined almost 5,000 reports on adult literacy and numeracy interventions, and only 3 randomized controlled trial designs focused on English as a second language. Adams and Burt (2002) cast a much wider net in their search for research on adult language learners between 1980 and 2001 to include experimental, descriptive, and practitioner studies from journals, books, reports, and dissertations. The 44 studies reviewed had methodological weaknesses, such as too few participants, unreliable measures, inadequately described practices and outcomes, and no comparison tasks or groups, which prevented drawing conclusions about the effectiveness of the approaches. Several of the studies focused on language learners in English preparatory classes before attending college, who are likely to differ in several ways (education level, first language literacy proficiency, socioeconomic status) from the broader population of English language learners. These results are consistent with a recent review of adult literacy instruction research available from the U.S. Department of Education (Kruidenier, MacArthur, and Wrigley, 2010). Similarly, the committee located four studies (two of adults in adult education and two of students in developmental college education courses) from 1990 to 2010 with the criterion that the research include at least one quantitative measure of literacy skill (see Appendix C). Because studies are so few and the ones available suffer from various methodological constraints, it is not possible to draw strong conclusions about effective instructional practices.
Given the limited research on the literacy development of adult English language learners in the United States, we also draw from a broader base of knowledge on second language and literacy development, which includes relatively well-educated adults and young children in K-12 education. Because a main challenge of literacy development for this population is learning a second language, we review research related to the development of both spoken and written language.
For simplicity, we use the term English language learners in this chapter to refer to foreign-born and native-born adults who are developing their English language skills and refer to other adults as native English speakers. On occasion we use more specific terms provided by study authors when referring to individual research studies. The research and sources of information reviewed in this chapter often do not include, however, precise or consistent ways of defining particular subgroups of the English learner population. In future research, more standard terms and definitions will be needed to refer to segments of this population to facilitate the accumulation of reliable, valid, and more interpretable research findings.
The available research, though limited, suggests that, compared with adult native speakers with low literacy in adult education programs, adult English language learners with low literacy in these programs show weaker vocabulary, passage comprehension, and sight word reading skills but better phonological processing (decoding nonwords) and somewhat better phonological awareness (Nanda, Greenberg, and Morris, 2010; see Chapter 2 for discussion of the components of reading). Similarly, Strucker et al. (2007) find that adult native speakers and English language learners tend to have different patterns of strengths and weaknesses as beginning readers. Language learners show weaknesses in vocabulary and comprehension but relative strength in decoding, whereas native speakers with low literacy tend to show the opposite pattern (Alamprese, 2009; MacArthur et al., 2010a). Even for those highly literate in their first language, some explicit teaching of English decoding rules may be needed to fill gaps in knowledge (Davidson and Strucker, 2002).
Findings for poor readers in middle school, who are more likely than proficient readers to need literacy instruction as adults, show a range of difficulties that are comparable for both native speakers of English and students with a different home language (Lesaux and Kieffer, 2010). Some students show global difficulties with language, decoding, and comprehension of text. Others have accurate and automatic decoding but poor general and academic vocabulary that affects comprehension. Still others have accurate but slow decoding and so are not fluent readers.
With good instruction, young adolescent language learners can perform at similar levels to native speakers on word recognition, spelling, and phonological processing tasks (Lesaux, Rupp, and Siegel, 2007). Similarly, adult language learners can develop decoding skills that are equivalent to native speakers (Alamprese, 2009). For both native speakers and language learners, once decoding is efficient, English oral proficiency (usually assessed by vocabulary and listening comprehension) predicts English reading comprehension, in higher grades (Lesaux and Kieffer, 2010). However, young language learners often score considerably lower than native speakers on English reading comprehension tasks (Goldenberg, 2008; Nakamoto, Lindsey, and Manis, 2008). Although adult language learners (and native speakers) can establish basic decoding skills quickly with good instruction, they need help with developing their reading skills beyond the intermediate fourth and fifth grade levels (Sabatini et al., 2010; Strucker, Yamamoto, and Kirsch, 2007). Vocabulary and comprehension skills have been particularly difficult to change with instruction, however.
Vocabulary and background knowledge are usually underdeveloped for English learners, in part because they lack the English skills needed to learn
through the texts and social and instructional interactions in schools, which are in English. Like native speakers, English language learners must gain facility with academic English, which has some features that differ from conversational English (Snow, 2010). For language learners, conversational English can develop in a few years (Collier, 1987), but becoming proficient with an academic language takes longer because it has its own jargon, linguistic structures, and formats, which can be specific to a discipline. These features of academic language need to be explicitly highlighted and supported during instruction (Achugar and Schleppegrell, 2005; de Jong, 2004; Schleppegrell, 2007). Some researchers emphasize that mastery of academic language is the single most important determinant of academic success for adolescents who have been in U.S. schools for less than 2 years (Francis et al., 2006).
Several factors affect the development of language and literacy in a second language and are important to consider in the design of effective instructional practices for segments of the English learner population. These factors include degree and type of first language knowledge, education level, English language proficiency, age, aptitude for language, reading and learning disabilities, and cultural and background knowledge.
Among adults, years of education in the primary language correlates with English literacy development (Condelli, Wrigley, and Yoon, 2009; Fitzgerald and Young, 1997; Strucker and Davidson, 2003). A detailed statistical analysis involving thousands of immigrants in Australian literacy programs shows that age and education in the home country were the two main predictors of literacy (Ross, 2000). Research with young students, including instructional intervention studies, also shows that to the degree that students have a strong literacy foundation in a first language, their first language literacy proficiency helps English literacy development (Farver, Lonigan, and Eppe, 2009; Goldenberg, 2008; for a meta-analysis, see Slavin and Cheung, 2005). For adolescents, self-reported first language and English proficiency in eighth grade predict English reading comprehension outcomes in grades 8, 10, and 12 as well as postsecondary achievement (occupational prestige, postsecondary education). Using data from the National Education Longitudinal Study (NELS), Guglielmi (2008) found that self-reported language proficiency of Hispanic learners predicted both initial levels of English reading and rates of improvement and through
that, high school and post–high school achievement. (Similar results were not found, however, for Asians who spoke various first languages, such as Chinese, Filipino, or Korean.)
Effects of the first language on second language processes. Precisely how language and literacy in a first language affects second language development needs to be studied more thoroughly to understand how best to facilitate second language acquisition, especially for less educated adults. The extensive literature on bilingualism (knowledge of two spoken languages) is beginning to suggest ways in which a first language may help to support second language growth. Although more experimental research is needed, modern research methods that include behavioral, psychophysiological, and neuroimaging techniques have been used to study questions of bilingualism, such as how two languages are represented in the brain and whether parallel lexicons coexist for bilinguals or if they possess one integrated lexicon. Less is known about the development of more than two languages, and so we have restricted our focus to the bilingual case.
Psycholinguistic research has mainly looked at how knowledge of two languages affects comprehension and production of each one. Does a bilingual person using one language activate the same information in the other language while listening or speaking? Such parallel activation across languages has been observed in many experiments, in the form of cross-language ambiguity effects, for example: Whereas “hotel” has the same meanings in Dutch and English, “room” has different meanings (it means “cream” in Dutch). A Dutch-English bilingual will briefly (and unconsciously) activate both meanings of the word “room,” quickly choosing the one that is appropriate to the language being used. Similarly, words that are pronounced differently in two languages (e.g., “coin” in French and English) produce interference in silent reading compared with words with very similar pronunciations (e.g., “piano”); (Kroll and Linck, 2007, 2009). Similar effects occur in comprehending sentences, as measured by word-by-word reading times, eye movements, and evoked potential measures. These effects are modulated by such factors as an individual’s familiarity with each language and the relative frequencies of the word in different languages. However, they suggest that knowledge of a second language becomes closely interlinked to knowledge of a first language, making it difficult to inhibit activation of the alternative language under many conditions.
Studies using functional magnetic resonance imaging also support that the two languages share brain structures and circuits instead of having segregated ones (Abutalebi, 2008; Abutalebi, Cappa, and Perani, 2001). The degree of overlap appears to depend on such factors as the age at which the second language was learned and second language proficiency. Individuals whose knowledge of the second language is relatively weak, for example, have shown greater activation of frontal regions that reflect more cognitive
effort and use of working memory. For skilled bilinguals, switching between languages involves increased attention or executive functions also associated with the frontal lobe, areas that are not as activated in monolingual language processing. These additional processes can be expected to provide cognitive benefits, specifically enhanced executive function and skill in allocating attention (see Bialystok et al., 2005).
Adults bring an already well-developed system for processing a first language that affects processing specific features of the second language. For example, language learners appear to be aware of grammatical structures that are similarly marked in both of their languages (e.g., auxiliary verbs used in progressive tense: “estar” in Spanish versus is in English) or unique to a second language (determiner gender marking “un/una” in Spanish). However, if a linguistic structure is marked differently in the second language, it may not be noticed (e.g., determiners of number agreement in “el/los” in Spanish versus “the/the” in English) (Tokowicz and MacWhinney, 2005). To summarize, recent findings from behavioral and neurobiological research imply that the role of the primary language cannot be ignored during the learning of a second language.
Literacy skills across languages and possibilities for transfer. Transfer from a native language to English depends on the overlap in characteristics between the two languages. Learning to read English involves matching distinctive visual symbols to units of sound in the spoken language (see Chapter 2; Ziegler and Goswami, 2006). Language learners may be familiar with writing systems that differ in their degree of similarity to English; for example, a native Spanish speaker will be familiar with an alphabetic system like the one for English, whereas a native Chinese speaker will know a nonalphabetic system. Moreover, some languages do not have a written form. Languages that do have a writing system represent their oral languages in different ways, both in terms of the symbols used as well as the phonological units that are represented in print. Some languages are nonalphabetic and represent morphological and phonological rather than purely phonological information (e.g., Japanese Kanji and Chinese). Some languages have alphabetic writing systems but use a non-Latin script (e.g., Korean, Russian, Hebrew). Even alphabetic languages that use the Latin script can be very different from English (e.g., Malay, Turkish, Welsh).
Languages differ in availability (which phonological units are more salient in the spoken language), consistency (the number of possible mappings in word recognition and spelling—more specifically, the number of different pronunciations for orthographic units and the number of different spellings for phonological units), and granularity (the nature of orthographic units that need to be learned to access the phonology). For example, in Chinese many different characters need to be learned, whereas in languages like Spanish a small subset of letters is enough to represent phonemes accu-
rately. English is in between: single letters represent phonemes, but because of the inconsistencies at the phoneme level, larger units (such as onset rimes) provide more systematic information on how to pronounce a word. For example, vowel /a/ can be pronounced differently by itself in different words (“car/lake/pat”), but in a larger rime unit (such as “/-at/”) it is pronounced the same way (“hat/cat/mat”). Even young beginning readers are sensitive to the characteristics of their spoken language and find it easier to perform the phonological awareness tasks that focus on the salient units in their spoken language (Durgunoğlu and Oney, 1999; Ziegler and Goswami, 2006). Depending on a language’s characteristics (consistency, availability, and granularity), learning to decode can be almost trivial or take longer.
For individuals literate in their home language, the first language writing system and how it represents the oral language affects the strategies used in English decoding. For example, when college students who are highly literate in a first language are learning English, Japanese and Chinese speakers rely on visual cues more than Korean or Persian speakers because the latter two groups have a phonologically based rather than morphemically based writing system, although all four of the groups use non-Latin scripts (Akamatsu, 2003; Hamada and Koda, 2008; Koda, 1999).
If adult English learners are not literate in their first language, then literacy development in English has to include instruction to develop sensitivity to the phonological units of English, the English alphabet, and the mappings at both phonemes and larger units. For individuals who are already literate in their first language and already have a metacognitive understanding of spelling-sound mappings, word recognition and spelling skills develop rapidly (Burt, Peyton, and Adams, 2003), especially when instruction highlights the specific characteristics of English.
If certain skills and strategies are available to a learner in a first language, building on them may help to develop literacy in a second language (for reviews see Dressler and Kamil, 2006; Durgunoğlu, 2002, 2009; Genesee and Geva, 2006). For language learners, proficiency in phonological awareness is positively related across two languages, even when the first language is not similar to English (for a review, see Branum-Martin et al., 2006; Genesee and Geva, 2006; Swanson et al., 2008). Decoding skills in a first language overlap with decoding skills in English as the second language, even across a span of 10 years (Sparks et al., 2009a, 2009b), suggesting that decoding skill in a first language supports decoding in a second language. As children gain more experience in English, English decoding becomes a stronger predictor of English reading comprehension than Spanish decoding (Gottardo and Mueller, 2009; Manis, Lindsey, and Bailey, 2004; Nakamoto, Lindsey, and Manis, 2008). The results look different for spelling: spelling in a first language (mostly Spanish) is either not
related or negatively related to proficiency with English spelling (Rolla San Francisco et al., 2006).
Vocabulary knowledge across the two languages of language learners is relatively independent (Cobo-Lewis et al., 2002; Nakamoto et al., 2008). However, the metacognitive aspects of vocabulary knowledge, such as knowing how to construct formal definitions, are related across the two languages (Durgunoğlu, Peynircioğlu, and Mir, 2002; Ordoñez et al., 2002). In addition, proficiency with explicit analytic processing or awareness of language (e.g., of morphology or cognates) in a first language correlates with having these skills in the second language (Deacon, Wade-Woolley, and Kirby, 2007; Nagy et al., 1993).
Good readers use similar comprehension strategies in both of their languages (Jiménez, 1997; Langer et al., 1990; van Gelderen et al., 2007). Writing proficiency is also correlated across the two languages of language learners: good writers use similar writing strategies in both of their languages (Durgunoğlu, Mir, and Arino-Marti, 2002; Schoonen et al., 2003).
These findings point to possibilities for applying knowledge and skills in a first language to the second language when the literacy tasks involve analyzing language structure (phonology, morphology) or using metacognitive strategies. When the tasks involve language-specific patterns (e.g., orthographic rules for spelling, meanings of items), the data suggest limited or no transfer. The available data are correlations, however. Experiments are still needed to determine whether specific literacy skills may be leveraged for the development of more efficient instructional approaches. In addition, it is not yet known how much these relationships are due to the learner transferring a specific skill from the first language to the second language and how much they are due to common underlying proficiencies that may be less sensitive to instruction. For example, although metacognitive strategies in a first language may be spontaneously accessed and used in the second language, or have the potential for transfer with instruction, other shared cognitive processes (e.g., working memory in phonological awareness) may be less amenable to change.
For young language learners, proficiency with speaking English strongly predicts growth in English reading comprehension, and those with higher English proficiency reach reading comprehension levels of their native speaker peers (Kieffer, 2008). One crucial influence on reading comprehension is vocabulary. Grabe and Stoller (2002) and Laufer (1997) estimated that one needs at least 3,000 words in a second language to read independently in that language. The greater the number of unknown words in a text, the more text comprehension suffers (Hsueh-chao and Nation,
2000). Zareva, Schwanenflugel, and Nikolova (2006) found that in order to comprehend a college-level academic text, a vocabulary of about 9,000 words is needed. In addition to vocabulary breadth, the depth of one’s vocabulary correlates with reading comprehension (Qian, 1999). Based on their empirical work, Perfetti and Hart (2002) proposed the lexical quality hypothesis, which states that rich, stable, and integrated word knowledge (that includes orthographic, phonological, syntactic-semantic information) facilitates word recognition, especially when decoding cues are weak (see also Stanovich, 1980).
Explicitly teaching vocabulary can lead to significant improvement in word knowledge and comprehension for both monolinguals and language learners (August et al., 2009; Carlo et al., 2004; Lesaux et al., 2010; McKeown et al., 1985; Vaughn et al., 2009). Vocabulary develops not only through explicit teaching but also through routine exposure to language, especially print, which contains words and word structures used less often in speech (Nagy, Herman, and Anderson, 1985). In native speakers, literacy and degree of print exposure both predict growth in reading comprehension. Individuals with high levels of literacy do more reading and so develop their vocabulary, comprehension, and general knowledge through text, whereas those with lower proficiencies get less and less benefit from print. Not only in childhood, but across the life span, vocabulary and knowledge are predicted by print exposure (Stanovich, 1986; Stanovich, West, and Harrison, 1995). This pattern has not been studied specifically with adult language learners, but it is reasonable to expect that increased opportunities to learn from print and other exposure to spoken English beyond explicit instruction would help all learners.
For language learners in elementary and middle school, proficiency in oral communication develops rapidly, whereas decontextualized and formal language structures, such as those in academic settings, tend to take longer to acquire through exposure to varied texts and routine social interactions that support learning and practicing those forms of spoken and written language. The development of academic language has not been systematically investigated with adults, but a similar pattern can be expected. Most adult language learners, especially if they were born in the United States, report having good speaking skills, but according to the NAAL only a third had literacy skills beyond the basic level (Wrigley et al., 2009).
An analysis of U.S. census data (Batalova and Fix, 2010) showed that adults (both nonnative and native English speakers) who self-reported poor oral English skills (ratings of not very well/not at all) also had poor document literacy, but self-reports of good oral proficiency (ratings of very well/ well) did not predict literacy performance. For example, only 13 percent of native speakers and only 9 percent of nonnative speakers (and only 13 percent of native speakers) who reported having good spoken English skills
were proficient on document literacy tasks. Although these self-reported results need to be interpreted with caution, they suggest a difference between everyday communication skills in English and the English language skills needed to comprehend more sophisticated material in different domains. (The report did not state how many of the native speakers were second-generation immigrants or Generation 1.5 who had not completed their education.)
An important question in the teaching of adults is whether age affects the ability to acquire spoken and written language. In childhood, a first language is learned rapidly and without explicit instruction or consistent feedback. Children exposed to two languages are able to learn both (Bialystok and Hakuta, 1995), and hearing children of deaf parents become bilingual in both speech and sign (Mayberry, 2009). Because the bilingual’s learning task is more difficult, there are some differences in patterns of language development compared with the single language learner (Genesee, 2001). Some have hypothesized that a critical period for developing language ends with puberty (Lenneberg, 1967), and others propose that the window closes earlier (e.g., Pinker, 1994). Regardless of the exact timing, it is well established that the ability to learn a second language declines with age. The declines observed do not suggest, however, that literacy in a second language cannot be achieved in adulthood at the levels required for career and academic success. What they do imply is that learning a second language will take more time and practice at later ages, and that even at high levels of second language facility differences in spoken language might be expected between a native and nonnative English speaker.
There are competing explanations for why the decline occurs, which differ in their emphasis on biological versus environmental influences. One theory emphasizes the role of neurobiological development (Newport, 1990; Stromswold, 1995): whereas the young brain is well suited to acquiring languages rapidly and effortlessly, this capacity decreases because of neurodevelopmental processes, such as dendritic proliferation and pruning, and synapse elimination (Buonomano and Merzenich, 1998; Hensch, 2003). These neurodevelopmental changes are seen as similar to ones that affect other capacities (e.g., vision, Daw, 1994) and occur in other species (Doupe and Kuhl, 1999). This theory predicts an age-related discontinuity in second language attainment associated with the closing of the critical period for acquiring the skills of a native speaker (Johnson and Newport, 1989). Data from a recent large-scale study using U.S. census responses show linear age-related declines in second language attainment but not the
discontinuity that would implicate a biologically determined window of opportunity (Hakuta, Bialystok, and Wiley, 2003).
A second hypothesis is that plasticity declines because of success in learning a first language (Bever, 1981; Seidenberg and Zevin, 2006), rather than brain development. Learning a second language requires adjusting neural networks that support the first language. Adjusting existing neural networks for second language processing is very difficult, especially since in adults those networks have been stabilized and are still successfully used in first language processing (Seidenberg and Zevin, 2006). A third possibility is that critical period effects reflect changes in the conditions (social, environmental) under which the second language is learned (Flege, Yeni-Komshian, and Liu, 1999). Older learners of a second language may have more restricted exposure to the second language or less motivation to use it, limiting what is learned.
Regardless of the underlying explanation, age constraints on language learning may help to explain slower growth in older adults’ reading comprehension in general or second language reading comprehension (Alamprese, 2009) and other basic reading skills (Condelli, Wrigley, and Yoon, 2009). Certain linguistic structures in a second language may be more difficult to automatize and integrate later in life, which may affect comprehension of text. For example, Jiang (2007) found that late Chinese-English bilinguals were accurate in detecting violations of English morphological structure (plural -s) in unspeeded, written tests, indicating they had explicit knowledge of this structure. However when faster, computer-based tests were used, these bilinguals showed less sensitivity to such errors. Age-related differences in working memory also may affect second language and literacy acquisition, rather than a biological window for learning a language (Birdsong, 2006). For example, working memory affects second language acquisition, since it is involved in the implicit recognition of statistical properties and patterns of language, such as memory for instances and associations (see Ellis, 2005, for a review; McDonald, 2006).
General language aptitude predicts second language proficiency (Abrahamsson and Hyltenstam, 2008). In studies with high school and college students learning a second language in school, general second language aptitude as assessed using the Modern Language Aptitude Test (MLAT) strongly predicts the potential for reading comprehension, writing, listening, and speaking in a second language (Sparks et al., 2009b). It measures aptitude using tasks, such as learning the numbers or word meanings in a made-up language, mapping nonsense syllables to their transcriptions, and identifying grammatical functions. These tasks require phonological,
orthographic, semantic, and syntactic processing and inductive reasoning and so, as would be expected, the aptitude scores also relate to first language skills (Sparks et al., 2009b). It should be noted that although aptitude as measured by the MLAT can indicate how much effort and instruction will be needed to teach a second language, the measure is not designed to diagnose a learning disability.
When language learners experience reading and writing difficulties in a second language, it is hard to determine whether the cause is a true disability or not-yet-developed second language skills (Klingner, Artiles, and Méndez Barletta, 2006; Lovett et al., 2008a; McCardle et al., 2005). First language literacy levels can be a useful indicator: if learners have not developed first language literacy skills despite having had opportunities to do so, then a disability diagnosis can be considered (Durgunoğlu, 2002). For young language learners with reading difficulties, weak word and nonword recognition skills and phonological processing problems are found in both of their languages (Manis and Lindsey, 2010). This group can be distinguished from children who have reading comprehension problems because of underdeveloped oral language skills or sociocultural barriers. Just as with struggling native speakers, struggling language learners have similar risk factors, such as low socioeconomic status and attending low-achieving schools (de Jong, 2004; Grubb, 2008).
Children who have reading difficulties not related to exposure or quality of instruction show similar precursors and profiles and benefit from similar types of interventions regardless of whether they are native speakers or second language learners (Lesaux and Geva, 2006; Lovett et al., 2008b). More research is needed to determine how best to diagnose and intervene to develop literacy of both adult English native speakers and English language learners who have disabilities.
Research suggests that it is important in both practice and research to understand variations in learners’ cultural knowledge in order to develop effective learning environments. As described in Chapter 2, the component skills of literacy develop through participating in routine literacy practices in a culture for particular purposes and with the materials and tools available in the culture, which include uses of technologies for reading and writing inside and outside the classroom. Even when basic decoding is mastered, readers can struggle depending on the particular type of text they are asked to read, their level of background knowledge or interest, and the task they
are asked to do (Moje, 2009). While these contextual factors affect literacy instruction and performance for native English speakers, they are especially important to consider in practice and research with adults learning a second language, because these adults bring more diverse cultural and educational backgrounds and literacy experiences.
Decades of literacy research have shown that comprehension involves interpreting the meaning of text using preexisting knowledge, beliefs, and opinions. The more one knows about a topic, the better one comprehends the material (Anderson and Pearson, 1984a, 1984b; McNamara, de Vega, and O’Reilly, 2007). Errors in comprehension can occur if the reader has incorrect or misleading information (Kendeou and van den Broek, 2005). For language learners, lack of cultural knowledge can hamper reading and listening comprehension (Center for Applied Linguistics, 2010; Droop and Verhoeven, 1998; Lesaux et al., 2006). Even for college students learning a second language, cultural knowledge plays a role in their comprehension (Brantmeier, 2005; Carrell, 1984; Fitzgerald, 1995). Although advanced language learners have very rich semantic networks in English, these networks may differ from those of native speakers (Zareva, Schwanenflugel, and Nikolova, 2005) indicating that differences in cultural experience shape how learners create word associations (or understand and remember relations between words).
Cross-cultural studies show certain cognitive processes are not necessarily universal, even for highly educated college students. Some basic processes, such as categorization, perception of an object in relation to its background, and making causal attributes, have been shown to be affected by the cultural context in which an individual was raised and educated (Ceci, 1991; Choi, Koo, and Choi, 2007; Choi, Nisbett, and Norenzayan, 1999; Nisbett et al., 2001; Norenzayan and Nisbett, 2000). Behavioral and brain scan data also show that monolinguals who did not have the opportunity to develop literacy skills because of social or cultural obstacles (rather than a neurological problem) perform more poorly on certain cognitive tasks, such as two-dimensional naming, phonological processing, memory, verbal abstraction hypothesis testing, and decision making, but not on verbal fluency or word repetition (Dellatolas et al., 2003; Reis, Guerreiro, and Petersson, 2003; Stanovich, West, and Harrison, 1995). All of these results highlight the importance of studying how culture affects cognition related to language and literacy.
In addition to cognitive factors, affective factors associated with culture can influence language learning. In particular, language learners can feel insecure about their English skills and cultural differences and feel conflicted between a desire for cultural integration and a desire for preserving their own home culture. Anxiety and motivation also have been shown to relate to second language achievement. For instance, Sparks et al. (2009a) found
that U.S. high school students’ second language achievement correlated with self-reported motivation and anxiety with respect to learning English. A Canadian descriptive study showed that, especially for individuals with minimal literacy and schooling in their first language, ESL classes can be frustrating and embarrassing, leading to an unwillingness to participate in classroom settings (Klassen and Burnaby, 1993). A meta-analysis showed that identity with the second language community and effort and desire to learn the second language correlated with second language achievement among Canadian students (Masgoret and Gardner, 2003), although it should be noted that the participants were required to learn an official second language and so the particular findings may not generalize. Nonetheless, findings such as these show the importance of considering adults’ affective responses to instruction and the learning environment in research and practice.
This section describes promising approaches to consider in the development of effective second language literacy instruction for adult learners. We draw on three sources of information. First, given the importance of language development to literacy, we review research on teaching a second language to high school or college students. This research has focused mainly, however, on developing linguistic structures and vocabulary for those who are highly literate in their first language. We also review studies of children who have limited literacy in their first language and who are developing both oral and written language skills in English. These findings may provide insights into effective practices for adults with no or very limited education or literacy facility in their first language. A third source of information is practitioner descriptions of practices that appear to be effective or ineffective in ESL adult education classes and that may be studied more systematically in future research. The practices gathered from these sources of information, which we describe next, suggest the importance of (1) integrated explicit instruction and opportunities for the implicit learning of language and literacy, with a focus on both linguistic form and meaning with feedback, (2) development of vocabulary and content knowledge for learning and reading comprehension, (3) extensive practice outside the classroom, (4) leveraging knowledge of the first language, (5) multimodal instruction, (6) attention to writing, (7) attention to the affective aspects of learning and instruction, and (8) sound assessment of literacy skill and affective and psychological outcomes of instruction.
Second language learning involves both implicit learning as well as explicit knowledge about language. Across the years, methods for teaching a second language have fluctuated between emphasizing sequenced explicit instruction of grammatical structures and using language to communicate for a purpose (Long, 2009). One promising approach is task-based language teaching (see Box 8-1) (for a thorough review, see Ellis, 2005; Vouloumanos, 2008).
Research is required to know precisely how to configure instruction with the appropriate balance and emphasis and whether it depends on characteristics of the learner and other factors (Long, 2009; Long and Crookes, 1992; Norris and Ortega, 2000; Spada and Lightbown, 2008). Most empirical evaluations of instructional approaches have been short-term studies of very specific language structures (e.g., use of Spanish clitic pronouns) in college foreign language classes. In one study of 256 adults in ESL programs in the United Kingdom, however, the main factor that predicted increased
Task-Based Language Teaching
The task-based language teaching method is a promising approach that integrates explicit instruction and implicit learning and emphasizes that language is learned from communications used to accomplish certain tasks and goals (Ellis, 2005; Long and Crookes, 1992; Robinson and Ellis, 2008). Language instruction occurs as tasks are performed. The tasks are selected to be relevant and meaningful to the learners, consistent with observations that relevance and connections to communicating for real-world purposes are especially important for adult language learners (Condelli, Wrigley, and Yoon, 2009). In task-based language teaching, the first step is to analyze the learner’s practical literacy needs (e.g., reading technical manuals, communicating with a child’s teacher, navigating bureaucratic mazes, taking lecture notes) and the learner’s developmental levels. This information is used to inform the design of systematic and structured instruction. Instruction involves gradually increasing the complexity of the communicative and conceptual demands of the tasks, while directing the learner’s attention to the language structures and tools (such as those used for understanding the referents of pronouns) available in the language.
Genuine materials are used which differ from “authentic” materials in actual use in that they are designed to have features that systematically scaffold learning while enabling engagement with meaningful and valued content. Task-based language teaching includes a systematic focus on the grammatical form of language and not only a focus on meaning.
oral language proficiency (grammar and vocabulary) was the observed balance and variety of instruction, which was defined as connections among lessons; integrated reading, writing, and speaking; and use of a wide range of activities and materials (Baynham et al., 2007). Longer term intervention studies are needed with a broader range of language structures and populations to determine how best to integrate explicit instruction techniques and genuine experiences with communicating, depending on the language development needs of learners in adult literacy education.
A principle of learning is that most students have trouble discovering important principles on their own, without careful guidance, scaffolding, or well-crafted materials. English language learners are usually exposed to much more input than they can process, and learners continually test hypotheses and filter input through knowledge of their first language. Instruction that focuses the learner’s attention is needed to isolate and highlight the crucial parts of the input, especially with complex structures of syntax (Gass, Svetics, and Lemelin, 2003). For example, second language cues that are absent or marked differently in the first language may have low salience for a learner. Cues that are redundant or accompanied and resolved by other contextual resources may not be explicitly noticed and learned either (such as the two cats, in which both the inflection and the number term indicate plurality).
In a meta-analysis to evaluate the effectiveness of different instructional techniques, Norris and Ortega (2000) found that focused second language instruction was more effective than the control conditions. Explicit teaching that included rule explanation as part of the instruction produced stronger effects than implicit teaching that included neither rule presentation nor directions to attend to particular linguistic forms (see also Long, 1983; Nassaji, 2009). A third finding was that instruction that begins with learners doing meaningful tasks but also presents opportunities to provide information about linguistic structures (e.g., an error) produced comparable results as instruction in which an ordered sequence of linguistic forms is taught outside a communicative context (e.g., teaching a sequence of grammatical structures), especially with explicit instruction in both (Norris and Ortega, 2000). In addition, Ellis (2005) reports that teaching formulaic expressions (e.g., I don’t understand) and basic rules of chunked materials (e.g., I don’t + verb) may help learners develop familiarity with the English language and master structures used frequently for communicating. Less is known about how instruction interacts with contextual variables, such as the proficiency of the learner (Gass, Svetics, and Lemelin, 2003), type of language outcome measured (Norris and Ortega, 2000), complexity and
type of linguistic structure (DeKeyser and Sokalski, 1996; Gass, Svetics, and Lemelin, 2003), and characteristics of learners’ first language (Williams, 2005).
As described in Chapter 2, research with young language learners also shows the benefit of direct instruction on phonological awareness, decoding, vocabulary, comprehension, and writing (Goldenberg, 2008). Sometimes derided as “drill and kill,” direct and explicit instruction simply means making the components and requirements of complex literacy tasks obvious and salient, and it generally can be done in motivating ways. The most effective language instruction for young children combines interactive instruction involving discussions among teachers and learners and direct approaches that involve explicit teaching, modeling of correct language usage, and feedback (Genesee et al., 2006; Goldenberg, 2008).
Focus on Both Form and Meaning and Providing Feedback
There is much debate about how to draw the learner’s attention to an error in a linguistic structure without disrupting the communicative interactions that are also needed for learning. Similar issues arise in any situation in which the long-term flow of an activity is part of the instruction. A standard approach in other areas is to use video or other capture techniques and then replay the moment at which a small correction is needed, after the overall event occurs, during postproblem reflection opportunities (Katz, Connelly, and Allbritton, 2003; Lesgold and Nahemow, 2001).
A principle of learning is that students benefit from feedback on their performance in a learning task. Feedback is especially important in language development because, despite some linguistic errors, the meaning of a communication may still be clear. Because communication is not disrupted, incorrect forms may not be challenged and continue to be used. Therefore, second language educators emphasize the importance of giving feedback on what is not an acceptable form, even when the form is comprehensible, as well as explicitly teaching the structures to prevent and correct such errors (Ellis, 2005; Lyster, 1998). These findings are consistent with research on “negative suggestion effects” showing that learning wrong information can be reduced when feedback is immediate.
A common tool used in classrooms is to provide language feedback: “recasting” or responding to a learner’s error by restating what the learner has said while modeling the correct form (e.g., Learner: She go to school. Teacher: She goes to school? with stress on “goes.”). Recasts are useful because they can occur as part of the conversation and do not disrupt the flow of communication. They temporarily focus the learner’s attention on language itself. Because recasts are usually subtle, they may not be noticed
or used, and so they tend to be more effective when made more explicit (e.g., signaling the error with stress).
In a study of eight Somali-speaking young adults, those with higher literacy in their first language were better able than those with lower literacy to learn to form questions in English using recasts that teachers provided (Bigelow and Tarone, 2004; Tarone, Bigelow, and Hansen, 2007). At this point, it is not clear whether the development of complex syntactic structures, such as relative clauses, requires more explicit or additional or more varied supports (e.g., presenting information visually, in writing) for those with low literacy. Learning from recasts can be limited unless classroom instruction focuses explicitly on language itself (through explicit prior instruction on the form, for example) or unless, consistent with the Somali study, the learners are at a more advanced stage (Ellis, Loewen, and Erlam, 2006; Lyster, 1998; Nassaji, 2009; Nicholas, Lightbown, and Spada, 2001).
Encouraging a learner to generate the correct form can be effective for learning (Loewen and Philip, 2006). This finding is consistent with the generation effect, which states that learning is enhanced when learners produce answers in comparison to having them recognize answers. Such “desirable difficulties” also can present challenges that require cognitive effort (e.g., to retrieve the correct form) and thereby lead to longer retention and learning. In sum, although recasts have been studied extensively and systematically in second language classrooms, research is needed to determine how best to use recasts to facilitate learning for adults.
Rich and Elaborated Input
Rich and elaborated input provides the learners with opportunities to experience and learn certain structural patterns in the language related to phonology, morphology, syntax, or pragmatics, although they may not be aware of learning the grammatical rules that govern the structures. Through intensive exposure to language, learners develop an implicit understanding of the patterns in a language—for example, how often certain structures or words are used. Nevertheless, these incidental opportunities for learning about the structure of a language are not sufficient, especially for learning the more complex structures presented in text, and require explicit attention to form (Ellis, 2005; see Chapter 3 for a similar description of language and literacy development in a first language).
As described in Chapter 3, adult literacy instruction often includes authentic reading materials. The term authentic often carries the incorrect connotation that explicit teaching is not necessary for learning to read and that exposure to “real-world” language and text is enough (see Condelli and Wrigley, 2004, for a similar argument). For language learners, and even for adults developing literacy in a first language, the spoken and especially
more formal written input in authentic communications can be overwhelming. Although one option could be to simplify the input, such simplified materials do not allow learners to experience the complex structures of the second language that they need to learn. A more promising approach is to use “elaborated” input (Long, 2009) that includes linguistic supports, such as redundancy, paraphrasing, synonyms, clear signaling, and marking to increase topic salience, making the information flow chronologically, using shorter sentences, and so on. Likewise in written and especially spoken language elaborations, instruction includes frequent clarification requests and comprehension checks (Long, 2009; Yano, Long, and Ross, 1994). However, making the language more comprehensible does not mean using child-like content. Adult learners need materials that are interesting and relevant to their knowledge development needs.
Francis and colleagues (2006) have compiled research-based recommendations for helping adolescent newcomers in schools who have limited English proficiencies and have difficulty especially with reading and writing academic texts. The literacy difficulties of these students may stem from limited oral proficiency in English, limited exposure to English texts, and possible gaps in background knowledge for the topic. Taking all of these factors into consideration, it is suggested that effective instruction for adolescent newcomers includes content-based literacy instruction, with an emphasis on developing academic language. In this approach, there are dual, integrated objectives: teachers address content through language and teach language through content. In addition, explicit instruction is used to teach reading comprehension and writing for academic purposes.
Effective vocabulary instruction for adolescent newcomers is explicit, systematic, extensive, and intensive (Francis et al., 2006). Explicit instruction involves not only direct instruction of the meanings of specific key words but also direct instruction in effective word learning strategies, such as breaking words down into parts, using contextual clues, and using dictionaries as references. Systematic instruction requires teachers to thoughtfully choose the key words that they teach and create multiple opportunities for meaningful exposure to the words and their meanings. Extensive vocabulary instruction is incorporated into every lesson, integrated across the curriculum. Finally, intensive vocabulary instruction provides depth of knowledge, such as an understanding of multiple meanings of words, their different forms, and different contexts of use and situated in larger conceptual frameworks. These instructional strategies are accompanied by high-quality ongoing classroom assessments to monitor students’ progress
and, if needed, appropriate intervention for newcomers with word reading difficulties (Francis et al., 2006).
Initiatives to improve the academic vocabulary of language learners in upper elementary and middle schools (August et al., 2009; Lesaux, Kieffer, and Kelley, 2009; Vaughn et al., 2009) have integrated the teaching of academic vocabulary with learning in content areas, such as social studies and science. Instead of teaching vocabulary as an itemized list of new words, the instruction integrates the words into discussion of what Vaughn et al. (2009) calls “big ideas,” such as human rights. The target words are taught using a combination of strategies, including reading, writing, oral discussions, and multimedia (e.g., videos to build background knowledge). The words are used and practiced in different contexts. In addition to teaching specific words, these programs also explicitly model and teach word analysis and comprehension strategies. Some also use students’ first language, Spanish, as a resource. There is also peer support, dyad and group work of the learners.
ALIAS (Academic Language Instruction for All Students) is a good example of such a program developed for middle school students (Lesaux et al., 2009). As the name implies, this program targets all students, both native speakers and English language learners. The curriculum provides rich and systematic instruction of high-utility academic words. There are multiple, planned exposures to each word through reading, writing, class discussions, and group activities. The students are encouraged to talk about these concepts, engage through personal connections and class discussions, and finally use the words in writing. The evaluation of the program indicates that for native speakers as well as language learners, there was significant growth in the targeted vocabulary, in word analysis, and, most importantly, in reading comprehension—although it should be noted that the size of the gain was relatively small compared with the amount of gain needed, and the practical meaning of the gains was not clear.
As discussed in Chapter 3, contextualized literacy instruction is an approach that is consistent with principles of learning and has sufficient preliminary support to warrant further research on its effectiveness with adults. Few data exist for adult language learners. The Integrated Basic Education and Skills Training program, or I-BEST, in Washington state is a program in which basic skills instructors and college-level career-technical faculty jointly design and teach college-level occupational courses for adult basic skills students. The program aims to increase successful completion of postsecondary occupational education and training (Jenkins, Zeidenberg, and Kienzl, 2009). The instruction of basic skills is integrated with instruction in college-level career-technical skills courses. The tracking of these students for 2 years showed that I-BEST students (both adult basic education and ESL students) had better basic skills (assessed by Comprehensive
Adult Student Assessment Systems) and the persistence to continue their education (e.g., earning college or vocational credits, certification). These students were not randomly assigned, however, but rather self-selected into I-BEST programs, so the results need to be interpreted cautiously. The effectiveness of embedded programs has not been evaluated systematically. Observations and interviews indicate that teacher specialization or boundaries and coordination within programs can present challenges to effective implementation (Cara et al., 2006; Guenther, 2002).
Learning continues outside the classroom where adult language learners can experience continued interactions in both spoken and written English (Reder, 2008). Successful language learning requires extensive second language input and opportunities to interact with others and to use language to express their own ideas, thoughts, and views (Ellis, 2009). Exposure to rich language patterns is also helpful, because learners are quite sensitive and readily notice the common patterns in a language (Vouloumanos, 2008). Thus, it is important not to isolate language learners from native speakers and to maximize exposure to the second language using many different venues.
Technology is a promising tool to provide practice outside the classroom through opportunities to use Internet sites, distance learning, and email. Although some new research has accumulated with adult language learners, most research on technology’s effectiveness with this population is old and ambiguous (Abraham, 2008; Torgerson, Porthouse, and Brooks, 2003). A new generation of research is needed because the feasibility, usefulness, and effectiveness of self-access models via technologies for adult language learners have not been fully explored (Wrigley, 2009).
Given the possibilities of transfer discussed earlier, more needs to be known about how best to use the first language to support development of English literacy. It is also reasonable to expect that acknowledging and valuing a learner’s first language are motivating, since they acknowledge and build on the knowledge and capabilities of the learner. When the first language is used as an aid to clarify instructions and tasks, learners show more growth in second language reading comprehension and oral proficiency (Condelli, Wrigley, and Yoon, 2009). Systematic use of the first language may not be feasible in many languages other than Spanish because of lack of qualified teachers and materials.
Research with monolinguals indicates that higher order comprehension skills necessary for reading can also be developed through discussions of material presented in different modalities, such as visual or auditory (Kendeou et al., 2008). Using technology to present information in a variety of modalities shows particular promise for language instruction, since language and content presented in a variety of modalities (visual, auditory, text-based) reinforce each other. In addition, visual and auditory presentations can provide varied input that is not available in print, such as regional accents, speed of discourse, pronunciation, and pragmatic uses of language. Research with monolinguals indicates that higher order comprehension skills necessary for reading can also be developed through discussions of material presented in different modalities, such as visual or auditory (Kendeou et al., 2008). As Hanley, Herron, and Cole (1995) report, visual support in the form of descriptive pictures significantly improved comprehension scores for English-speaking students learning French.
Anecdotal evidence from the adult literacy field consistently stresses that adults in literacy programs enjoy using technology (Benbunan-Fich and Hiltz, 1999; Parke and Tracy-Mumford, 2000). Cromley (2000) suggests that access to technology results in greater learner engagement and retention. Technologies for acquiring English have shown positive impacts on the frequency of revision and the complexity of content in the writing of adults learning a second language (Li and Cumming, 2001).
As explained earlier, speaking, listening, reading, and writing are all interrelated modes of communication (Hornberger, 1989). Even with very young language learners, providing exposure to oral and written language together is more effective in developing vocabulary and phonological awareness (Farver, Lonigan, and Eppe, 2009). Likewise, for adults, it is not a good strategy to provide only oral language instruction while waiting until reading and writing reach a certain level of proficiency. It is also useful to include and integrate both decoding and comprehension instruction (see Chapter 6 for further discussion of instructional approaches).
As for native speakers, writing is an essential part of instruction for adult language learners. It offers an opportunity to practice second language skills related to both reading and writing, and it can be another way to track second language proficiencies. Writing can also help to meet the learner’s practical needs for communication because those with limited literacy in their first language tend to take notes (and use other cognitive strategies) to overcome this limitation (Klassen and Burnaby, 1993).
Cross-sectional studies suggest that uses of vocabulary, syntax, mor-
phology, signaling, and rhetorical devices in writing improve with second language proficiency, as does the coherence and fluency of writing (Chenoweth and Hayes, 2001; Cumming, 2001; Sasaki, 2000; Sasaki and Hirose, 1996). First and second language writing processes are fundamentally similar, and knowledge of the first language is used in second language writing. But those with weak second language skills tend to devote more attention to form (e.g., finding the right word or syntactic structure in the second language by translating from the first language) and thus devote less attention to the macro processes of generating ideas, planning, revising, and editing (Sasaki, 2000). A promising avenue for research is to understand more about how to develop these macro processes in second language instruction (Sasaki, 2000). Other promising instructional strategies provide additional scaffolds and support, such as prediscussions of the writing topic, peers evaluating and responding to each other’s work (Berg, 1999), and teacher-student dialogue journals (Peyton and Seyoum, 1989). As Cumming (in press) summarizes, educators can facilitate second language writing development by providing extensive opportunities to write and by responding to that writing, modeling relevant text types and discourse interactions, by enhancing students’ self-control over their composing and learning processes, and by organizing curricula and assessments appropriate to learners’ abilities, purposes, and interests. Finally, extensive reading and vocabulary development in the second language are also helpful for writing.
Writing is a complex cognitive skill that is influenced by social and cultural aspects of the learner’s environment. Although the basic components of writing discussed in Chapter 4 apply to all adults, several other factors affect writing for second language learners. These include significant variability in first language background, educational level, second language proficiency, length of time in the new country, acculturation and familiarity with second language writing contexts, and the purposes and needs for writing. This complex web of factors has yet to be considered in a comprehensive model of second language writing development (Cumming, in press). Research is needed, especially with adult language learners in adult education settings, to track learners’ progress in the use of text features, their use of composition processes for different tasks and writing environments, and how progress changes as a function of different types of instruction (Cumming and Riazi, 2000).
As for reading, more needs to be understood about how to develop second language writing in content domains and how to support writing outside the classroom. A rare study of second language writing in the workplace illustrates how the specific style of writing and vocabulary required in a particular workplace evolves. In this observational research, newly graduated Francophone nurses who received mentoring, were encouraged to interact informally with peers, and had opportunities to observe others who modeled forms of communication in the workplace developed both
their expressive communication skills in the second language and the specific workplace genre for completing patient charts and discharge papers (Parks, 2001; Parks and Maguire, 1999). Such studies show the potential for developing language, reading, and writing in a second language outside the classroom.
Field research indicates the importance of attending to the affective aspects of instruction (Wrigley, 2009), although more systematic research on English language learners’ affective responses to literacy instruction is needed to develop motivating and supportive approaches. Field observations show that beginning learners are reluctant to use English inside and outside the classroom because they may feel insecure about their linguistic skills. English learners can become demotivated, frustrated with the slow pace of literacy instruction; repetitive instruction (e.g., as teachers try to catch up students who have missed a class); a focus on topics that are not well matched to the learner’s education level, interests, or familiarity with U.S. culture (e.g., a focus on holidays when content related to science and technology and topical discussions is preferred). Those whose goal it is to transition to training or postsecondary education mention the lack of focus on academic vocabulary in high beginning or intermediate classes.
As mentioned earlier, the general principles for supporting motivation and persistence in Chapter 5 are likely to apply to language learners. Given the unique contexts that surround language learners, it is important to make the learning environment safe, supportive, and comfortable (Hardman, 1999); to make instruction useful and valuable to the learner (Burt, Peyton, and Adams, 2003); to encourage support through collaborations and peers (Baynham et al., 2007; Cener for Applied Lingistics, 2010; Slavin, 1996; Taylor et al., 2007; Watanabe and Swain, 2007; see Torgerson, Porthouse, and Brooks, 2003, for a review); and to use relevant topics, activities, and texts for instruction. Although not yet systematically evaluated, cooperative learning and other forms of peer support may matter even more for adult language learners. Even when adults in certain ESL classes reported feeling frustrated at times, they reported enjoying meeting people and getting to know other speakers of Spanish (Klassen and Burnaby, 1993), so the social aspects of the instructional environment may be especially powerful in motivating persistence.
Adequate assessments are lacking for English language learners. The need to develop more valid and comprehensive approaches to the assessment of adults’ reading and writing skills also applies to this population.
Four additional issues specific to the English language learner population have emerged in research and practice: (1) assessment of learners’ linguistic and cultural backgrounds and existing language proficiencies, (2) the need to avoid the use of tests developed for native speakers of English, (3) assessment of incremental progress in subcomponents of spoken and written language proficiency, and (4) assessment of affective and psychological outcomes.
Learner Background and Existing Proficiencies
The heterogeneity of adult English language learners requires having systematic ways to assess the backgrounds and such factors as first and second language and literacy proficiencies that influence English literacy development. Currently, teachers report that it is a challenge to provide instruction that is sufficiently common to all in a classroom while differentiating instruction to meet the needs of all learners (Wrigley, 2009). As more is understood about the factors that affect English language and literacy development for different English language learner populations, more reliable and valid assessments can be developed to help make placement decisions and inform instructional planning.
Use of Tests Developed for Native English Speakers
Assessments in English that are developed for native speakers may not provide valid information about language learners for two reasons. If linguistic complexity rather than the content of a test item causes low performance, the assessment does not reliably assess that content knowledge (Abedi, 2006). This type of measurement error occurs on tests other than reading, such as mathematics tests, which are sometimes incorrectly assumed to be relatively independent of linguistic proficiency (Abedi, 2002, 2006; Abedi and Lord, 2001).
Second, because reading comprehension, whether in a first or second language, is tied to background knowledge (Garcia, 1991; Lesser, 2007), a language learner may show poor comprehension not because of poor language or comprehension ability but because the topic is unfamiliar. In fact, for Spanish college students learning English, discipline-related background knowledge and language proficiency compensated for each other. Those with low English proficiency could read texts successfully if they had prior knowledge about the topic, and those with high English proficiency comprehended texts even if they had low background knowledge about the topic (Uso-Juan, 2007).
Such interrelationships between language proficiency and background knowledge have not been systematically explored with adult language learners. The possibility of existing background knowledge (in the first
language) compensating for low linguistic proficiency is intriguing, and further research can identify if and how background knowledge in the first language can be used as an instructional tool while building English proficiency.
Assessment of Incremental Progress in All of the Subcomponents of Spoken and Written Language
Another challenge in assessing language learners is the complexity of language acquisition. Understanding spoken or written language requires integrating multiple sources of information, such as word meanings, syntactic rules, and background knowledge. Because assessments usually tap into only a subset of language skills (e.g., vocabulary) or include measures that are too broad (e.g., oral proficiency), they may not assess the full range of language skill. Moreover, different components of language develop at different rates and in an incremental fashion. For example, vocabulary knowledge is not a simple dichotomy of knowing or not knowing a word’s meaning. Rather, knowledge is a continuum that ranges from not knowing a word, to recognizing it, to knowing it roughly, to describing it very accurately and knowing its uses in different contexts (Schoonen and Verhallen, 2008; Vermeer, 2001). Such incremental growth in linguistic knowledge is not reflected in vocabulary tests. An analysis of the 19 most common assessments for language learners (many of which are not widely used or standardized) identified the need for assessments that measure a greater range of language skills and more detailed proficiency levels (Center for Applied Linguistics, 2010). Assessments are needed for different purposes. Although global measures at the program level may be sufficient for accountability purposes, more fine-grained assessments are needed at the individual level to assess language and literacy growth for planning instruction and providing feedback to learners (see Center for Applied Linguistics, 2010, for a review).
Given the integrated development of spoken and written language, proficiencies in both written and spoken English should be assessed. Language and literacy in the first language and level of education are also important to assess to guide instruction because, as reviewed earlier, these are closely linked to second language development.
Assessments that involve selected or constrained responses (e.g., multiple choice or completion) show the largest effects of instruction because they match instruction closely. Free response tasks that require spoken or written answers are better measures of learners’ second language proficiency, however, because they relate most closely to language use outside the classroom (Ellis, 2009; Norris and Ortega, 2000). Developing tests of the second type, especially for language learners, is a challenge, but, as for all adults in literacy instruction, it is important to develop reliable and valid
measures to assess performance on relevant real-world tasks (Purcell-Gates et al., 2002).
Assessment of Affective and Psychological Outcomes
A range of affective and psychological influences on learning and desired outcomes are important to evaluate in addition to language and literacy skill. These include self-efficacy in the use of spoken and written English, effortless and confident navigation of new contexts in the culture, and the ability to interact comfortably with native speakers, all of which remain difficult to quantify.
The number of adults who need to develop their English literacy skills in the United States is substantial and growing, and the population is extremely diverse. These adults differ in languages spoken, education levels, literacy skill in the first language, knowledge of English, familiarity with U.S. culture, and other characteristics. The adults differ in the component skills they need to develop and bring to the challenging task of learning to use and comprehend a second language. Some English learners need to understand how the English writing system represents the spoken language and how to decode and read words in English. They often need to develop vocabulary and knowledge of linguistic features, such as syntax and morphology. Background knowledge related to the culture, the texts to be comprehended, and purpose of a literacy task all may need attention to help adults use their skills to make inferences and create a rich mental representation of the meaning of text. Communicative expression may need to be developed in both spoken and written modalities.
Various cognitive, linguistic, social, affective, and cultural factors influence the development of literacy in a second language. These include education and proficiency in the first language, age, type and degree of existing English proficiency, aptitude for language, possible learning disabilities, cultural and background knowledge, and interest. All of these factors must be considered in the development of instruction for adults learning English as a second language.
Research on effective practices for developing English language and literacy in adults is severely limited, especially those with low levels of education and literacy in the first language. Thus, this chapter reviews three additional sources of information to identify promising practices to study further with adult English language learners: (1) studies of second language teaching in high school and college settings, (2) studies of children who have limited literacy in their first language and who are developing both oral and written language skills in English and thus may provide insights
Practices to Apply and Study with English Language Learners
Engaging and differentiated instruction for adults who vary in
• English language and literacy skills,
• first language proficiency,
• educational background, and
• familiarity with U.S. culture.
Instruction that integrates explicit instruction with opportunities for the implicit learning of language and literacy, with a focus on
• both linguistic form and meaning with feedback,
• development of vocabulary and content knowledge for learning and reading comprehension,
• extensive practice outside the classroom,
• leveraging knowledge of the first language,
• multimodal instruction,
• attention to writing,
• attention to the affective aspects of learning and instruction, and
• sound assessment of literacy skill and affective and psychological outcomes of instruction.
into effective practices for adults with limited education or literacy facility in a first language, and (3) practitioner descriptions of practices used in adult education ESL classes and that warrant more systematic research attention. Box 8-2 shows practices to apply and study in future research. A particular challenge is the need to differentiate instruction for adults in a classroom who vary in first language proficiency, educational background, and familiarity with U.S. culture.
Box 8-3 summarizes directions for research. The overarching priorities for this research agenda are to (a) develop and evaluate effective instructional methods for diverse populations of English language learners; (b) develop adequate assessment methods; (c) identify or develop the technologies that can facilitate the learning of language and literacy skills for adult English language learners who differ in their knowledge of English language and literacy, first language literacy, and educational and linguistic backgrounds; and (d) specify the training and supports instructors need to implement the instructional approaches effectively. Standard terms and definitions for describing the subgroups of this diverse population of adults will need to be used in this research to produce more reliable, valid, and interpretable information about the approaches that generalize across subgroups and the specific approaches that meet a particular group’s literacy development needs.
Directions for Research on English as a
Second Language Instruction
• Experiments to identify effective instructional practices for different groups of language learners (with varying first languages, knowledge of English, first language literacy skills, educational backgrounds, and reasons for attending instruction) to help instructors differentiate instruction.
• Studies to specify the length, type, and intensity of instruction that is the most effective for different language learner groups.
• Systematic and longitudinal analyses of language teaching practices (integrating language structures with language use and meaningful content) and documentation of outcomes for adult language learners.
• Comprehensive description and analysis of the components of effective programs at multiple levels (instructional content, teaching practices, student interactions, and so on) using quantitative and qualitative methods that link components to outcomes.
• Background variables that have an impact on outcomes and that are important to assess at program entry and for differentiated instruction.
• Characteristics of learners and aspects of language exposure (both inside and outside the classroom) that predict learning and a range of other desired outcomes that include persistence, continuation with further education, finding employment, and lifelong learning.
• The relation between first language skills and the development of spoken and written English skills and identification of opportunities for transferring skills and strategies.
• Ways to provide effective multimodal language instruction (speaking, reading, writing, visual presentations) and technology.
• Ways to integrate classroom instruction with informal learning opportunities provided by interactions in communities and through the use of technology.
• The most effective ways to integrate language and literacy development with content instruction.
• Development and evaluation of “integrated instruction” models that combine language and literacy education with academic and career education.
• Assessments that (a) provide enough information about language and literacy skills and progress to be useful for planning instruction and providing feedback to learners, (b) are valid measures of practically important language and literacy competencies, and (c) measure affective, cultural, and psychological factors that affect learning.
• Teacher knowledge and professional development to effectively administer and use assessments and flexibly adapt the curriculum to meet learners’ needs.