Journal of Deaf Studies and Deaf Education Advance Access originally published online on August 3, 2006
The Journal of Deaf Studies and Deaf Education 2007 12(1):8-24; doi:10.1093/deafed/enl009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Focus-on-Form Instructional Methods Promote Deaf College Students' Improvement in English Grammar
National Technical Institute for the Deaf, Rochester Institute of Technology
Correspondence should be sent to Gerald P. Berent, Department of Research and Teacher Education, National Technical Institute for the Deaf, Rochester Institute of Technology, Rochester, NY 14623-5604 (e-mail: gerald.berent{at}rit.edu).
Received March 27, 2006; revised July 6, 2006; accepted July 9, 2006
Focus-on-form English teaching methods are designed to facilitate second-language learners' noticing of target language input, where "noticing" is an acquisitional prerequisite for the comprehension, processing, and eventual integration of new grammatical knowledge. While primarily designed for teaching hearing second-language learners, many focus-on-form methods lend themselves to visual presentation. This article reports the results of classroom research on the visually based implementation of focus-on-form methods with deaf college students learning English. Two of 3 groups of deaf students received focus-on-form instruction during a 10-week remedial grammar course; a third control group received grammatical instruction that did not involve focus-on-form methods. The 2 experimental groups exhibited significantly greater improvement in English grammatical knowledge relative to the control group. These results validate the efficacy of visually based focus-on-form English instruction for deaf students of English and set the stage for the continual search for innovative and effective English teaching methodologies.
The English Grammar Challenge
For the majority of deaf students, a major obstacle to the development of overall English literacy skills is the difficulty of mastering many of the most fundamental English grammatical forms and structures. The English grammar challenge is well documented and indicates that, on average, deaf students lag years behind their hearing peers in knowledge of English grammar (Berent, 1988
, 1996; Quigley & King, 1980; Wilbur, Goodhart, & Montandon, 1983). Additionally, a study of deaf and hard-of-hearing students' (ages 8–18) performance on the Stanford Achievement Test, 9th edition, showed that even when deaf students are performing at grade level, their language skills are lower than the language skills of their hearing peers on the Language Subtest (Traxler, 2000, p. 347).
Much of the earlier literature on deaf students' English grammatical knowledge had focused on deaf students at the primary and secondary levels of education (Quigley & King, 1980). Therefore, Berent (1993) assessed the English grammatical knowledge of college-level deaf students with the goal of determining whether these students would continue to show an increase in their English grammatical knowledge at college. College-level deaf students exhibit great variability in English language proficiency, and admission to certain degree programs and other academic options depends crucially on their ability to attain requisite levels of English proficiency. Accordingly, the ability to continue improving in English can be a critical determinant of educational success in college.
Berent (1993) assessed the English grammatical knowledge of 105 undergraduate deaf students using the Revised Test of Ability to Subordinate (RTAS, Berent, 1988), a sentence-combining multiple-choice measure for assessing students' knowledge of nine target English formations. The 105 students were divided into low, mid, and high English proficiency groups on the basis of their scores on the Michigan Test of English Language Proficiency (1977). Students were assessed using the RTAS at the beginning of the academic year and again at the end of the academic year to determine whether and to what extent students might exhibit gains in grammatical knowledge over time. Results revealed that all three student groups made significant improvement on the RTAS between the beginning and end of the academic year. However, the low English proficiency group improved to a significantly greater extent than the other two groups. Berent concluded that the amount of explicit English language instruction provided to deaf college students at lower proficiency levels should be greater than the amount provided to students at higher proficiency levels because it is the students at the lower English levels who exhibit the greatest potential for improvement in English grammatical knowledge.
At the National Technical Institute for the Deaf (NTID) at the Rochester Institute of Technology (RIT), deaf students who test at the lower levels of English language proficiency, in fact, currently receive additional English grammar instruction as a corequisite to their academic writing courses. This corequisite course consists of 1 hr of remedial grammar instruction per week in addition to the four contact hours of academic writing instruction that they receive. In this grammar course, nine target grammatical forms and structures are identified for mastery and are listed in Table 1. For each formation, the table also indicates a code that will be subsequently discussed in association with an instructional intervention, along with an example or description of the form or structure.
|
In addition to the benefit of increasing explicit grammatical instruction for deaf students, the search for more effective English teaching methodologies is an ongoing effort, and the critical need for instructional experimentation continues to be emphasized (Berent, 2005b
). In this study we undertook an experimental approach to teaching English grammar to deaf students in classroom research that explored the use of focus-on-form English instruction (Doughty & Williams, 1998; Williams, 2005) with college-level deaf students.
Focus-on-Form Instruction
Doughty and Williams (1998) provided an overview of the development and use of "focus-on-form" methods for teaching students of English as a second language (ESL), as well as the results of earlier research that evaluated the use of these methodologies. The development of focus-on-form methods and research burgeoned out of work such as Long (1991), Doughty (1991), and Sharwood Smith (1993), who were responding to the negative outcomes of the "no grammar teaching" initiatives of the 1980s that eschewed form. Subsequent research has established that effective second-language acquisition requires not just a focus on meaning but a focus on form as well (Williams, 2005).
However, as emphasized in Long (1991), "focus on form" (singular) is crucially different from focus on forms (plural), which involves the isolation and examination of particular forms of the target language as the objects of language teaching and learning. The crucial difference is that focus-on-form approaches are intended to draw the learners' attention to specific language forms within naturalistic communicative contexts in the process of linking form and meaning. The rationale for these methods is based on the theoretical and practical assumption that the language learner needs to attend to, or notice, target language forms as a prerequisite to processing and ultimately acquiring those forms (Gass, 1997; Schmidt, 1995; Sharwood Smith, 1994). This assumption follows the fundamental postulate of virtually all theories of linguistic knowledge, namely, that language acquisition is triggered by linguistic input available in sentences of a language that occur in communicative contexts (Chomsky, 1986).
In view of the primacy of linguistic input in all language acquisition, most theories of second-language acquisition hypothesize that learning is undermined by the frequent failure of the second-language learner to notice the very input that occurs in the linguistic environment. For example, Gass's (1997) model of second-language acquisition can be summarized in the following steps:
- Noticed input: recognition that there is something to be learned; preparation of the input for further analysis
- Comprehended input: learner's understanding of the input
- Intake: selective processing and assimilation of linguistic data
- Integration: hypothesis confirmation, rejection, or modification for later confirmation
- Output: overt manifestation of the acquisition process
As indicated by the five steps, cognitive noticing (or apperception in Gass's terms) is at the front door of the entire second-language acquisition process. The ultimate goal of focus-on-form instruction, therefore, is to facilitate Step 1 so that the noticed input can activate the remaining four steps. Schmidt's (1995) "noticing hypothesis" explicitly tied focus-on-form instruction to the achievement of this goal.
Focus-on-Form Research
Actual research on focus-on-form interventions over the past 15 years has yielded mixed results with respect to their efficacy. Jourdenais, Ota, Stauffer, Boyson, and Doughty (1995) obtained positive results in the use of visual input enhancement (specifically, "textual enhancement"), one type of focus-on-form methodology (see below). They employed textual enhancement for facilitating the noticing of the Spanish past tense forms (preterit and imperfect) by English-speaking second-language learners of Spanish. Their enhancement protocol involved the use of underlining and a different font to highlight the past tense forms in a printed text along with the bolding of the preterit forms and the shadowing of the imperfect forms. The results of their study revealed that the students who received the input enhancement treatment more readily detected the target forms, relative to their control group, and produced more of the target forms in obligatory contexts in their written productions. Among other studies that obtained positive outcomes from focus-on-form interventions are Doughty (1991), Leeman, Arteagoitia, Fridman, and Doughty (1995), and Shook (1994, 1999).
Particular studies that failed to establish the efficacy of focus-on-form interventions include Izumi (2002), Leow (1997), and Overstreet (1998). For example, Izumi studied the impact of visual input enhancement on the learning of English relative clauses by college-level ESL students. The participants read passages in which relative clauses were textually enhanced through underlining and in which relative pronouns (e.g., which) were further enhanced using bolding and a different font. Izumi's pretest/posttest design employed several assessments including sentence-combining and sentence completion tests, an interpretation test, and a grammaticality judgment test. Despite evidence that participants noticed the target forms in the input, the study's results revealed that the visual input enhancement did not facilitate the learning of English relative clauses.
Han (2005) and Han, Park, and Combs (2005) provided a comprehensive critique of 16 focus-on-form research studies on second-language learning conducted over the past 15 years. Han et al. (2005) ascertained that many previous studies that failed to show the efficacy of focus-on-form instruction had flaws in their theoretical assumptions or in their research designs. In contrast, the studies that demonstrated the efficacy of focus on form possessed arguably positive design characteristics. Among these characteristics are the following. The studies were long-term rather than short-term studies, they targeted "ready learners," the interventions provided participants the opportunity to "act upon" noticed input, and the interventions allowed participants to process the target input for meaning before processing it for form. These positive design characteristics were incorporated into this study and are explained in the Method section.
Focus-on-Form Instruction as Visual Input
All the studies critiqued by Han (2005) and Han et al. (2005) were focus-on-form investigations that involved visual input enhancement. Focus-on-form methodologies in general have been developed in the context of second-language instruction for hearing learners. With respect to English teaching and learning, whereas hearing ESL learners learn English under the cognitive constraints imposed on second-language acquisition, deaf learners of English are constrained by severely restricted access to spoken English input and, in many if not the majority of cases, by a delay in the onset of English language learning. Once deaf learners begin focused English language learning, often through formal schooling, English input is available to them largely through vision. Nevertheless, like all language learners, deaf learners must notice linguistic input before they can comprehend it, process it, and ultimately integrate it as acquired grammatical knowledge. Because deaf learners primarily have access only to visual linguistic input (although advancements in technology related to audition are changing that situation), any attempt to facilitate the noticing of English input through focus-on-form instruction must obviously employ methodologies that lend themselves to visual presentation.
The focus-on-form methods examined in Doughty and Williams (1998) are listed and briefly described in Table 2. Many of these methods lend themselves to visual implementation and, in fact, many are presented visually to hearing ESL learners. As already seen above, input enhancement is often implemented as (visual) textual enhancement. Input flood, which is essentially a bombardment on the learner of many instances of a target form, equally lends itself to visual presentation. Recast can involve the noticing of nontarget forms in print and their reformulation as correct forms in print. Other methods might conceivably be adapted to the visual modality; however, some descriptions imply implementation through spoken conversational interaction.
|
 |
Method |
|---|
 TOP Method ResultsDiscussionConclusionAppendixReferences |
|---|
Research Questions
The purpose of this classroom research study was to assess the efficacy of visually based focus-on-form English teaching methods for helping deaf college students improve their knowledge of English grammar. Under the assumptions of a theory of language acquisition in which noticing the linguistic input is a prerequisite for subsequent learning, the fundamental research question of this study is the following: Will visually based focus-on-form instruction facilitate deaf learners' noticing of the target language input, resulting in significant improvement in knowledge of a set of specific English grammatical forms and structures (see Table 1) over a 10-week period? Secondly, will the employment of different focus-on-form methods have a differential effect on the improvement of English grammatical knowledge?
Participants
The deaf participants in this study comprised three groups of college-level deaf students taking 10-week remedial English grammar courses that focused primarily on the nine forms and structures listed in Table 1. There were two experimental groups that were taught with a different mix of focus-on-form methods and one control group of deaf students that was taught without the use of any focus-on-form methods. Additionally, two control groups of hearing college students were recruited for the purpose of determining how hearing native speakers of English would perform on the assessment measures used in the study to assess deaf students' grammatical knowledge.
The two experimental deaf student groups are labeled the Input Group and the Dictogloss Group. These labels are based on the particular mix of focus-on-form methods employed in each group, which are described below. The deaf control group is designated as the Conventional Group, simply to reflect the fact that they did not receive focus-on-form instruction but rather received an instructional combination of grammatical explanation, drill and practice, and crosslinguistic comparisons with American Sign Language. The background characteristics of the three deaf student groups are displayed in Table 3.
|
The means and standard deviations in Table 3 are provided for each group's age, reading level as measured by the California Achievement Tests (Tiegs & Clark, 1957
), writing skills as measured by the NTID Writing Test (Albertini et al., 1986), general English language knowledge as measured by the Michigan Test of English Language Proficiency (1977), and Pure Tone Average hearing loss. The mean scores related to reading, writing, overall language proficiency, and placement in the NTID reading curriculum show that the deaf student groups in this study reflected a low level of English language knowledge in comparison with other deaf students enrolled at NTID or enrolled in baccalaureate programs in other colleges of RIT. For example, the California Reading grade levels of these participating students place them in developmental English courses at NTID. Research shows that 92% of bachelor degree and 82% of associate of applied science degree NTID/RIT deaf graduates have a ninth-grade reading level or higher (Cuculick & Kelly, 2003, p. 282). The participating students' NTID Writing Test scores also place them below the NTID student average and place them in the second lowest level out of four levels in the NTID writing curriculum. NTID students' average on the Writing Test is 47 (SD = 14.5) with a scoring range from 21 to 85 (Schley & Albertini, 2005, p. 99). The participating students' Michigan scores in the range from 50 to 60 also indicate their lower language abilities. Research shows that 87% of NTID/RIT baccalaureate-level deaf graduates and 78% of associate of applied science degree graduates have a Michigan score of 70 or higher (Cuculick & Kelly, 2003, p. 283). The participating students' NTID Reading Placement scores also place them in the second lowest level out of four levels in the NTID reading curriculum. The possible NTID Reading Placement Test percentile scores range from 0 to 200. Students with a percentile score of 150 or above are excused from the NTID reading curriculum requirements (Crandall, 1997). One-way analyses of variance (ANOVAs) on the group means listed in Table 3 revealed no significant differences between these groups on any of the measures.
Assessment Procedures
Essay.
During the first week of the course, all students were required to type a short essay on any one of the following three topics: "My Best Friend," "My First Job," or "My Worst Day." This First Essay served to establish a baseline, for each student, of productive grammatical knowledge of the nine target formations (Table 1). During the 10th week of the course, each student was required to type a Last Essay on one of the two remaining topics (i.e., not on the topic chosen for their first essay). Essays were coded and scored following the coding guidelines discussed below, and difference scores based on the First and Last essays formed the basis for assessing change in productive grammatical knowledge over the 10-week period.
Essays were coded and scored following the procedures described in Berent (2005a). Student essays were collected in both print and electronic format. Essays were assigned random numbers for later matching of essays with students' names, which were removed from the essays during the coding and scoring process to reduce the possibility of rater bias. Each student's essay was coded using the codes associated with the nine target formations shown in Table 1. When a student produced one of the nine formations successfully, the relevant code, preceded by a plus (+) sign, was typed into the Microsoft Word version of the essay immediately before the formation. When a student's formation was unsuccessful, the relevant code was preceded by a minus (–) sign. Only correct and incorrect productions of the target formations were coded, no other nontarget formations. The following paragraph illustrates the procedure. Target formations are italicized here for clarification.
Last year I –PST work first day at my city's Community College and –PST/PROG I'm doing landscaping job. First day in morning, my boss +PST told me +INF to air blow all on sidewalk on campus. Then I +PST went –INF get air blow in shop and –PST start –INF blow on sidewalk. In 30 minute later, my boss –PST check on me and –PST make sure +THAT –PST I'm all right and –PST/PROG do the job. I –PST keep blow and walk on sidewalk and finally –PST finish in one and half hours. Then I –PST return to boss's office and –PST ask him what –PST ___ next list +INF to do? He –PST give me list +ADJC that I –PST have +INF to do for day. Then I –PST look at paper and –PST see what I –PST suppose +INF to do.
The procedure for determining whether a target formation is successful (+) or unsuccessful (–) is guided by three criteria: (a) the formation is one required by the established discourse context, (b) the target formation is formed properly, and (c) the formation actually occurs where required. In the paragraph coded above, the topic My First Job as well as the student's use of the time expression last year establish that the paragraph is relating past events. Accordingly, the scorer assigned –PST before the verbs work, start, check, and so forth, because the student produced bare forms of the verb where past forms are required by the discourse. The underscore in the sequence what –PST ___ next list signals that the required form was omitted, in this case, was, for example. The –INF before get in the sequence went –INF get air blow in the shop indicates that the infinitive required after the verb went in this structure is formed improperly, lacking the sign of the infinitive to. Because two of the target formations, THAT and ADJC are clauses (Table 1), some formations occur within other formations. For example, in the above sample, the coding of the sequence list +ADJC that I –PST have +INF to do for day indicates that the complete sequence contains the correctly formed adjective (i.e., relative) clause beginning with that and that within the ADJC, there is an unsuccessfully formed past tense and a correctly formed infinitive. Note that lexical choice is not a component of this coding set; within the ADJC, the student's production of to do (rather than to complete, e.g.) is semantically close enough not to destroy the student's intended meaning or disguise the required grammatical structure.
For each student, a percentage score of productive grammatical knowledge was calculated by dividing the total number of correct productions by the total number of attempted productions, both successful and unsuccessful. The student sample above contains 8 successful formations and 19 unsuccessful formations. Therefore, the sample receives a score of 42.1% correct (8/27). The difference score for each student in the study is the percentage difference between the percentage correct on the First Essay and the percentage correct on the Last Essay.
Interrater reliability among scorers of the essays was calculated using a measure of correlation as the procedure (Borg & Gall, 1983). High interrater reliability of .96 was obtained across three raters. The internal rescoring consistency and reliability of an individual rater was .98. Such high interrater reliability and internal rescoring consistency indicate that the coding process is clear and more objective than subjective.
The complete coding of one student's First Essay and Last Essay is illustrated in the Appendix, along with the percentage score for each essay. From the Appendix and the sample coded above, it should be apparent that the coding system employed in this study is a largely objective procedure. It is driven by the student's established discourse and the target grammatical principles inherent in the student's attempted productions. In these respects, the procedure differs from the procedures employed in writing assessment that involves the holistic scoring of essays (e.g., Albertini et al., 1986), which can be considerably more subjective.
Grammar test.
In addition to the essay procedure, an in-house indirect assessment of grammatical knowledge was developed—the Test of Grammar for Academic Writing (T-GRAW)—that focused specifically on the nine formations (Table 1) targeted in the grammar courses. The T-GRAW consisted of 18 items that elicit grammaticality judgments (Part 1) and nine sentence rewrite items (Part 2). Part 1 included two items for each target formation, one that contained a correct formation and one that contained an incorrect formation. In Part 2, each of the nine sentences (one per target structure) contained an error that was expected to be corrected in the student's rewritten sentence. There were two randomized versions of the T-GRAW, A and B. Students who completed Version A as a pretest at the beginning of the course completed Version B as a posttest at the end of the course; those who completed pretest B completed posttest A.
The T-GRAW Part 1 items illustrated below include one grammatical sentence that should elicit a YES response and one ungrammatical sentence that should elicit a NO response. The first assesses adjective clause knowledge; the second assesses knowledge of the present progressive verb formation.
|
The following Part 2 item should elicit a rewritten sentence containing a correctly formed that-complement as illustrated in the italicized response below the item sentence. Following the syntax of English, a correct response could include the complementizer that or not.
- 3. My best friend thinks will pass his science test.
- My best friend thinks that he will pass his science test.
- My best friend thinks that he will pass his science test.
The maximum score on the T-GRAW is 36. Students' difference scores based on their pretest and posttest T-GRAW performance provided an indirect assessment of change in grammatical knowledge of the nine target formations over the 10-week period.
Two assessments, a direct and an indirect measure, were included in this study with the intention of possibly obtaining cross-validation of the study's results. However, it was not known to what extent the indirect measure (the T-GRAW) might be a valid assessment of deaf students' target language knowledge. With respect to deaf college students' writing skill assessment, Berent et al. (1996) found that, indeed, indirect measures of deaf students' writing ability had questionable validity in comparison to direct measures. However, because the issue of direct and indirect assessments of deaf students' grammatical knowledge has not been researched, it was decided to include both a direct and an indirect assessment for possible cross-validation.
Hearing control groups.
The two hearing control groups consisted of students in NTID's interpreter-training program and its teacher-training program that prepares teachers of deaf students at the secondary level of education. One group of 11 hearing students typed essays following the procedure described above. Their coded essays revealed productive knowledge of the nine target English formations that one would expect native speakers to exhibit. Their overall performance yielded 99.5% correct formations.
Another group of 18 hearing NTID students completed the T-GRAW (half Version A and half Version B). They achieved a total T-GRAW score of 93.8% correct. Their Part 1 score was 95.1%, and their Part 2 score was 92.6%. Although these hearing students' T-GRAW scores did not approach 100% correct, their scores were considered sufficiently high enough to approximate native speakers' grammatical knowledge of the nine target formations.
Instructional Methods
Selection of focus-on-form methods.
Because this was the first study of the efficacy of focus-on-form methods in teaching English to deaf students, it was important to select focus-on-form methods that lent themselves best to visual presentation. The various methods that have been employed in second-language teaching (Table 2) were reviewed for their adaptability to the visual modality and for their ease of assessability. Because input enhancement has been commonly employed in second-language research, especially in the form of textual enhancement, it was felt that the textual enhancement of select reading assignments would be a convenient and justifiable methodology to employ in the instructional process.
Moreover, because Han et al. (2005) had argued that a positive design characteristic of effective focus-on-form studies was to allow participants to act upon noticed input, the essay procedure (Berent, 2005a), which was used for the direct assessment of grammatical knowledge, was also included in the instructional procedure as a focus-on-form intervention. As discussed below, the essay procedure elicited participants' written productions in the form of communicative essays. The coding of those essays converted them into visually (textually) enhanced input for students to act upon in the instructional process through essay revision.
Of the remaining focus-on-form methods, it was realized that the dictogloss procedure, which involves spoken dictation, could be adapted to the visual modality as a "visual dictation" in the form of a projected paragraph for students to read and subsequently discuss and reconstruct following the other steps of the conventional dictogloss (see Table 2). Thus, the three methods—textual enhancement, essay coding for revision, and the visual dictogloss—were employed as focus-on-form methodologies designed or adapted for access by deaf students in the visual/print modality.
Textual enhancement.
During the 10-week course, both the Input Group and the Dictogloss Group were assigned readings in which some of the target grammatical formations of the grammar course were textually enhanced to focus the deaf students' attention on those formations. The following paragraph is taken from an assigned reading adapted from "Hollywood and Broadway" (n.d.). In the assigned reading, verb formations reflecting perfect aspect were visually enhanced through the use of a larger, bold font.
American movies and live theater have influenced the entire world, and this influence has grown stronger and stronger over the years. People in almost every country in the world have seen movies that were produced in Hollywood and plays that began on Broadway.Accordingly, discussion of such course readings and related in-class and homework activities were carried out on input-enhanced materials.
Essay coding and revision.
The essay coding procedure was used as a primary instructional tool for the Input Group and the Dictogloss Group but not for the Conventional Group. Although the First and Last Essays were used for assessing the grammatical knowledge of all three groups, students in the Conventional Group (the control group) never saw their coded essays.
For the two experimental groups, the coding of the First Essay provided visual focus-on-form input enhancement through the explicit coding of students' productions. This procedure constituted the conversion of the students' own communicative output, their essays, into enhanced input in the form of plus and minus codes preceding the target formations. Although the coding is explicit, the feedback is implicit in the sense that the instructor does not correct the students' unsuccessful productions. Instead, the coding forces the students to notice and reflect on their successful and unsuccessful productions. The successful productions serve as reinforcement of students' correct grammatical production appropriate to the communicative discourse contexts that they established, and the unsuccessful productions serve to motivate reflection on the target formations to be "mastered" in the course, to be compared against the successful productions in the essay, and to induce "learning" through the conscious effort to correct all unsuccessful productions during the essay revision process. The course activities of the Input and Dictogloss Groups included two to three revisions over time of each student's First Essay in which students were told explicitly to reformulate their unsuccessful formations by comparing them with and reflecting on their successful formations.
Visual dictogloss.
The visual dictogloss, or "visuogloss," involved the development of short paragraphs that could be displayed through projection on a screen to serve as a visual dictation. Following the guidelines of conventional dictogloss procedure, a visual dictogloss contained several occurrences of one target formation (Table 1) within the paragraph. This device actually incorporates an input flood (Table 2) within the visual dictation. However, the target formations are not textually enhanced. The rationale behind the dictogloss is that it should force the noticing of the recurring English formation and that student discussion and collaborative reconstruction should draw further attention to the formation.
With a conventional dictogloss (Doughty & Williams, 1998), the instructor reads the paragraph to the class several times. After the dictation, students discuss and collaboratively reconstruct the text (written or typed) based on their memory of the dictation. In this study, the visual dictogloss was projected and displayed long enough for students to read it three or four times. Afterward, students discussed the paragraph for a few minutes generally in groups of three; then one member of the group typed a reformulated paragraph on a computer with the collaborative input of group members as best as they could recall the original paragraph.
An example of a visual dictogloss is the following, in which the recurring formation is the past progressive verb formation, italicized only here for illustration:
Sharon was walking down the sidewalk with her friends. They were all discussing their summer plans. Sharon was explaining that she and her family are planning to visit California this summer. Marcia was telling everyone about her plans to visit Europe. When they were crossing the street, they were not paying attention, and a speeding car almost hit them.
This focus-on-form activity was provided only to the Dictogloss Group (hence, the group name). Accordingly, instruction provided to the two experimental groups differed only in this regard. That is, both groups participated in the revision of their input-enhanced First Essays and in activities related to textually enhanced readings. The visual dictogloss was a third focus-on-form activity provided only to the Dictogloss Group.
Design characteristics.
As noted in the introduction, Han et al. (2005) analyzed the results of 16 research studies on the use of visually based focus-on-form methods with second-language learners in an effort to identify characteristics of research design that may account for divergent results among those studies. They argued that the second-language studies that demonstrated the efficacy of focus-on-form methods exhibited particular design characteristics that were lacking in the studies that failed to establish the efficacy of focus-on-form methods. In this study we incorporated four of the positive design characteristics identified in Han et al.
First, this study involved focus-on-form interventions (with the two experimental groups) that spanned a 10-week period. Han et al. (2005) had found that it was the longer term second-language studies that yielded positive results in contrast to short-term studies that involved one-time or short-term interventions. Secondly, this study targeted ready learners, who are defined as having some degree of knowledge of the target formations. Despite the relatively low level of English grammatical knowledge exhibited by the participants in this study, they have all been exposed to English in various forms throughout their education and are therefore ready learners. Third, the focus-on-form interventions employed in this study allowed students to act upon noticed input, another positive design characteristic noted by Han et al. For example, essay revision, exercises based on textually enhanced readings, and reconstruction of visual dictogloss input were activities that required students' active involvement. Finally, this study employed a methodology—the essay procedure—that enabled the "sequential processing," rather than requiring the simultaneous processing, of meaning and form. Enabling sequential processing is another characteristic of the successful studies reviewed by Han et al.
The essay procedure involves sequential processing of meaning and form by virtue of the use of students' own output (i.e., their essays) as visually enhanced input. That is, in producing communicative, experiential essays, the students automatically understood their own output. Once the output was visually enhanced through the coding process and returned to the students in the experimental groups, they were forced to focus on target forms within contexts that they already understood. The procedure therefore involves the sequential processing of form and meaning. In sum, this study incorporated specific design characteristics that have been observed in previous focus-on-form studies that yielded positive results.
|
Results |
|---|
|
TOPMethodResultsDiscussionConclusionAppendixReferences |
|---|
Essay Analysis
For each of the three deaf student groups, total scores were calculated to determine each group's percentage change between the First Essay and the Last Essay. Because of the number of student participants in each group, the percentage scores were converted to arcsine values following the procedures recommended by Snedecor and Cochran (1980)
for handling a zero proportion and a 100% proportion when n is less than 50. These percentages transformed to arcsine values were then used in all statistical tests. To examine the changes between the First and Last essays, a 3 (Group) x 2 (Essay) ANOVA with repeated measures on the dependent variable was conducted. The main effect for the repeated measures variable showed that there was overall student improvement from the First to the Last Essay, F(1, 65) = 38.80, p < .0001. However, a significant Group x Essay interaction, F(2, 65) = 16.90, p < .0001, showed that it was the two experimental groups (Input and Dictogloss) that improved significantly between the First and Last Essay. The control group receiving conventional instruction did not experience significant improvement. Figure 1 illustrates the Group x Essay interaction. The percentage values indicate that the Input Group improved by 13 percentage points and the Dictogloss Group by 14 percentage points in their production of the nine target formations between the First Essay and the Last Essay. In contrast, the Conventional control group showed no change between their First and Last essays, thus creating the significant interaction effect with the two experimental groups. As a supplement to the Figure 1 group values, Figure 2 displays, for each of the three groups, the percentages of students who experienced a positive change between the First and Last Essay (i.e., a higher percentage score on the Last Essay than on the First Essay) and the percentages who experienced a negative change (i.e., a lower percentage score on the Last Essay than on the First Essay). Figure 2 reveals the high percentages of students in the experimental groups, who improved, relative to the low percentage in the control group, who experienced considerably more negative than positive change.
|
|
The essay difference scores between the First and Last essays were analyzed using a one-way ANOVA to measure students' increase in "mastery" of the nine target structures between the First and Last essays, where mastery was defined as
80% correct per grammatical formation. The one-way ANOVA for Group x Essay Difference score yielded a significant difference between the three student groups, F(2, 65) = 6.38, p = .003. Post hoc pairwise comparison of means using Fisher's Protected Least Significant Difference (PLSD) showed that the Input Group differed significantly from the Conventional control group, Fisher's PLSD mean difference = 1.9, p = .0007. The Dictogloss Group comparison with the Conventional control group approached, but did not attain, significance (Fisher's PLSD mean difference = 1.1, p = .063). Therefore, on this specific pre/postassessment, the Input Group, relative to the Conventional Group, exhibited a significant increase among its students in the mastery of the target formations between the First and Last Essay.
Grammar Test Analysis
For each of the three deaf student groups, total scores were calculated to determine each group's change between the pretest and the posttest of the T-GRAW. A 3 (Group) x 2 (Pretest, Posttest) ANOVA with repeated measures on the dependent measure yielded a significant main effect for the repeated measures between Pretest and Posttest, F(1, 65) = 12.64, p = .0007, indicating that all three deaf student groups improved significantly between the pretest and posttest with no interaction effects (p = .8381). Collectively, the groups increased from 65% to 70% correct between pretest and posttest, with no significant differences between groups.
To determine the source of improvement on the two-part T-GRAW, separate ANOVAs with repeated measures were calculated for Part 1 and Part 2. The Part 1 ANOVA did not attain significance (p = .21), indicating that the overall Part 1 increase from 76% to 78% was due to chance. However, the ANOVA analysis for Part 2 yielded a significant main effect for the Pretest/Posttest, F(1, 65) = 17.26, p = .0001, verifying that the overall Part 2 increase from 53% to 61% was responsible for students' significant improvement on the T-GRAW as a whole.
|
Discussion |
|---|
|
TOPMethodResultsDiscussionConclusionAppendixReferences |
|---|
Efficacy of Focus-on-Form Methods
The results of the essay analysis strongly support the efficacy of visually based focus-on-form methods for teaching English grammar to deaf students receiving remedial English instruction at the college level. Both the Input Group and the Dictogloss Group demonstrated significant improvement in their productive knowledge of a group of nine target grammatical formations (Table 1) over a 10-week period (Figure 1). In contrast, the Conventional Group, which received no focus-on-form instruction, demonstrated no such improvement in grammatical knowledge. The significance of the experimental groups' improvement is bolstered by the fact that 98% of the students in the Input Group and 78% of the students in the Dictogloss Group experienced a positive change between the First and Last Essay. In contrast, only 38% of the students in the Conventional Group changed in a positive direction.
There was a difference between the Input Group (75%) and the Dictogloss Group (44%) with respect to achieving mastery of the target grammatical structures. Under a 80% criterion for mastery, the Input Group's change in mastery over 10 weeks compared to the control group (35%) was significant, but the Dictogloss Groups' change compared to the control group approached, but did not attain, significance (p = .063). This difference in mastery may be due to chance or due to other factors that were not explicitly controlled for in this study. For example, the Dictogloss Group received one additional focus-on-form treatment than the Input Group did: the visual dictogloss. Within the constraints of the 10-week treatment period, more interventions reduce the percentage of time allocated to each individual intervention. Future research needs to explore the effects of both type and quantity of focus-on-form methods employed with deaf students.
Direct and Indirect Assessment
Whereas the essay analysis uncovered significant group differences and robust improvement by the experimental groups, the grammar test analysis revealed a significant 5% increase between the T-GRAW pretest and posttest but no differences between groups. It was also seen that the observed significant improvement on the T-GRAW was attributable to the groups' significant improvement only on the Part 2 sentence rewrite task.
First of all, Part 1 and Part 2 are fundamentally different types of indirect assessment. As a grammaticality judgment task, Part 1 contains both grammatical and ungrammatical sentences presented in isolation from any meaningful context. As students consider the acceptability of each sentence, there is no surrounding naturalistic discourse to guide attention to specific English forms, whether during pretesting or during posttesting. The same can be said of Part 2. However, in Part 2, students are told explicitly that every single sentence contains an error of some kind that must be repaired in their rewritten sentences. Because of greater familiarity with the target formations at posttesting, as a result of classroom exposure irrespective of actual learning that might have taken place, students in general would be more likely to achieve greater change on Part 2, relative to Part 1, simply through actively repairing every sentence. This is a different process from "feeling" the grammatical status of sentences in a grammaticality judgment task in Part 1. On the controversy over grammaticality judgments in ESL research see, for example, Cowan and Hatasa (1994) and Gass (1994).
Independent of these factors is the fact that, unlike the T-GRAW, the essay procedure constitutes a direct assessment of grammatical knowledge based on naturalistic student-generated written productions possessing communicative content within connected English discourse. Growth in grammatical knowledge over time is revealed directly in the extent to which students deploy discourse-appropriate grammatical formations in communicating life experiences in naturalistic contexts. It is through such direct assessment of grammatical knowledge that the Input and Dictogloss Groups displayed significant improvement over the Conventional Group, supporting the efficacy of focus-on-form interventions.
Though not a focus of this study, the robust significant differences obtained through the essay procedure relative to the T-GRAW suggest that, as in the assessment of deaf students' writing skills (Berent et al., 1996), it is likely that direct assessments of deaf students' grammatical knowledge have more effect power and thus greater reliability and validity than indirect assessments. This issue should be examined directly and systematically in future research.
Positive Design Characteristics
A significant factor that likely contributed to the success of this study in supporting the efficacy of visually based focus-on-form methods for promoting deaf college students' improvement in English grammatical knowledge is the positive design characteristics that this study incorporated. Based on the conclusions of Han (2005) and Han et al. (2005), previous studies that have supported the efficacy of visually based focus-on-form interventions with hearing second-language learners all possessed characteristics that the studies that failed to support such focus-on-form interventions did not possess. This study incorporated four of those design characteristics.
Specifically, this study involved focus-on-form interventions that were long-term (spanned a 10-week period) relative to studies that involved short-term or "one-shot" interventions. Secondly, this study targeted ready learners, who had some degree of prior knowledge of the target English formations. Third, the three focus-on-form interventions employed in this study allowed students to act upon noticed input in various ways (reading-related assignments, essay revision, and visual dictogloss reformulation). Finally, the essay procedure permitted the sequential rather than simultaneous processing of meaning and form. That is, the students' own essays were automatically understood (processed for meaning) by the students before they were enhanced and then processed for form during the revision process.
Weaknesses of the Study
Despite the positive design characteristics discussed above, there are two areas of weakness associated with this study. First, there was no tight control for treatment integrity other than the fact that each participant group was taught following a syllabus devoted to mastery of the nine target English formations listed in Table 1. It was impractical, if not impossible, to guarantee equal relative proportions of class time devoted to each formation across the three participant groups. It is very difficult to control for treatment integrity in classroom research of the type carried out in this study, especially in student-centered classrooms where student needs can alter the most carefully designed lesson plan. Nevertheless, future research of this type should make every effort to monitor treatment integrity across groups as closely as possible.
Secondly, group improvement over the 10-weak period was assessed on difference scores between the First and Last Essay and between the T-GRAW pretest and posttest. The study did not include a follow-up delayed posttest to assess the extent to which learning that occurred during the study period was retained after a delay of several weeks. A delayed posttest should certainly be incorporated into the design of future research of this type on focus-on-form interventions with deaf students of English. Interestingly, Ellis (2002) reviewed several focus-on-form studies of hearing second-language learners and observed that all the studies that demonstrated learner improvement between pretesting and posttesting also demonstrated retention of that improvement in the results of delayed posttesting.
|
Conclusion |
|---|
|
TOPMethodResultsDiscussionConclusionAppendixReferences |
|---|
Overall, the findings of this study support the efficacy of visually presented focus-on-form methods in teaching English grammar to deaf college students. The employment of these methods resulted in students' significant improvement in knowledge of a set of basic grammatical formations over a 10-week period. More specifically, the use of a combination of methods that enhance the English input provided to deaf students at a remedial English level yielded significant improvement that was not observed in a control group of deaf students who did not receive focus-on-form instruction. Importantly, the results of this study validate the effectiveness of theory-based teaching methods that postulate "noticing" as a prerequisite to the further processing of input and the ultimate acquisition of new grammatical knowledge. It is therefore essential that further research on English language teaching to deaf students capitalize on the centrality of noticing linguistic input and explore all avenues for facilitating the noticing of English input.
It was noted in the introduction that, based on Berent (1993), explicit English grammar instruction should be increased for deaf students with lower overall English skills because they exhibited the greatest potential for improvement. This does not mean that the search for innovative and highly effective grammar teaching methods should always focus on deaf students with lower English skills. On the contrary, the efficacy of promising methodologies such as visually based focus-on-form instruction should be explored at all levels of English language proficiency and at all ages. Nevertheless, although the opportunity to enhance English language input for deaf students might lead to significant growth in English language knowledge and improved educational success at any level, it is critical to determine just how early and in what format focus-on-form interventions can be provided to younger deaf students. Because early intervention is one of the most critical factors leading to successful language development, any teaching methods for which efficacy is established for use with deaf students should be further studied for implementation as early as possible.
The positive results of this study underscore the need for more classroom-based research for comparing methodologies used in teaching English to deaf students and for discovering which methodologies hold the greatest promise for promoting deaf students' early and ongoing English language development. The promising methodologies that were the focus of this study were adapted from the field of second-language teaching and research. In view of this fact, researchers and educators of deaf students have an obligation to seek answers not only within "the field of deaf education" but also through careful and focused cross-disciplinary inquiry.
|
Appendix |
|---|
|
TOPMethodResultsDiscussionConclusionAppendixReferences |
|---|
One Student's Coded Essays
The two essays below were produced by the same student: the First Essay My Worst Day and the Last Essay My Best Friend. The coded essays contain the relevant symbols for the nine target structures listed in Table 1. The symbols precede the target formations, which are italicized in this appendix for illustration. Symbols associated with the student's successful productions are preceded by a plus sign; symbols associated with unsuccessful productions or missing productions required by the discourse context are preceded by a minus sign. The rationale for most of the codes used below is apparent from the discourse context. Otherwise, superscripts within each essay refer to notes that follow the essay and provide clarification for the reader.
The student's First Essay contains 14 successful and 8 unsuccessful formations, yielding a percentage correct score of 63.6%. The Last Essay contains 30 successful and 11 unsuccessful formations, yielding a percentage correct score of 73.2%.
First Essay
My Worst Day
I +PRS have a worst day. I +PST went to Freeport, Bahama with my friends +ADJC who +PST were from Buffalo.1 We +PST got big mistake about our vacation plans. We +PST ordered the tickets by the Travel Helpers, but they +PST made mistake online for vacation. When we –PST get there and we –PERF been waited for six hours2 +INF to get our ride from the bus, but they still –PST don't have our names on the lists or files for the hotel. The travel helpers +PST gave us the awful hotel in downtown. We +PST paid the ticket –ADJC with –PST include for foods, VIPs, hotel, and bus3 but we +PST got nothing. We –MOD cant afford4 for fun or something else. We +PST waited for 3 days –INF to got checks by the travel helpers. We +PST stayed there extra days for free. But I really +PST enjoyed on my vacation. ___ –PST Solve on my trouble vacation.5
Last Essay
My Best Friend
I –PST had a best friend, Tim –ADJC who we +PST grew up in first grade from our old school. We +PST graduated high school in 1998. I +PST did not see my old best friend anymore because I +PST went to different state because of deaf college. Tim –PST has +INF to stay at home for Darden Community College in Darden, Missouri, where I +PST lived. I +MOD couldn't hold of him because he –PST did not had1 any computer and I +PST had computer as such as email and internet. In 1998, we –PST did not had any computer with an internet. Few years ago, I +PST went Tim's house and his mean mother +PST told me +THAT that he just –PST ___ kicked out2 because he +PST did not pay her mother's bills –ADJC that she +PST wanted him +INF to pay bills. She +PST got no idea where he +PST went. I +MOD would like –INF to ___ my best friend if we +PST had the computer, but he +PST had –INF to gone somewhere in Missouri. I +PST knew +THAT that everybody –PST get older3 and +PST became family where we +PST finished in high school. I +PST felt despondently +THAT that my best friend +PST was gone because we +PST promised before ___ +PST graduated from high school and we +PST agreed –INF for being4 best friend for ever until we +PRS die.5
|
Acknowledgments |
|---|
Preliminary results of this research were presented by Berent et al. (2005)
. We gratefully acknowledge the participation of the NTID students who served as subjects in this research study. We also thank Kathleen Eilers-Crandall for her facilitation of data gathering. No conflicts of interest were reported. |
Notes |
|---|
1 The sequence ... who were from Buffalo is a successfully produced adjective clause (+ADJC) containing a successfully produced past tense form (+PST), were.
2 The discourse context requires a verb reflecting perfect aspect and, in this case, past tense, perfect aspect: we had waited for six hours. The coding system uniquely identifies formations, with no double descriptions. Thus, the combination of present or past tense and perfect aspect represent a PERF formation, irrespective of the tense.
3 The most likely interpretation of the sequence ... the ticket with include for foods, VIPs, hotel, and bus is "... the ticket which included food, VIP [passes?], hotel, and bus." Accordingly, the sequence following the ticket is coded as an unsuccessful adjective clause containing an unsuccessful past tense form.
4 Consistent with the uniqueness principle of the coding system, MOD refers to both the modal verb and the bare form of the main verb with which the modal co-occurs. Because the discourse requires couldn't afford, the formation, can[']t afford is designated –MOD.
5 Although not terribly frequent in the students' essays, sequences like Solve on my trouble vacation diverge somewhat from grammatical English structure. In such cases, the most plausible interpretation was employed in coding. In this case, the verb solve has no subject as required in English finite sentences. Therefore, an underscore was provided to signal a missing element (the sentence subject). Regardless of the structure inferred, this sequence would require a past tense verb, hence, –PST Solve.
1 In keeping with the uniqueness principle of the coding system, did not had in the sequence he did not had any computer, is coded –PST because it is an ill-formed past tense formation despite the fact that it contains the negative not and therefore the auxiliary verb did.
2 The student's production, that he just kicked out, contains the required elements and occurs in an appropriate environment (after told me) to be judged a properly formed that-complement (+THAT). However, within the that-complement, there is a missing required element for a properly formed past tense formation, specifically was or got to create a past tense, passive voice formation: was kicked out or got kicked out. Because passive is not one of the target formations, the relevant target is the missing past tense required in the formation, therefore, the code –PST.
3 In a situation in which the discourse contains recurring past tense forms, get in the sequence that everybody get older and became family is coded –PST rather than –PRS because the surrounding environment makes it clear that the student did not intend to produce the generic present tense form gets.
4 In the case of we agreed for being best friend, syntactically, the complement of agreed here could be an infinitive, to be, or a prepositional phrase beginning with on followed by a gerund (on being). Because an infinitive is more commonly the complement of agree (we agreed to be best friends) and because the student did not produce the preposition on, the error was scored as –INF. Furthermore, we agreed on being best friends has a different connotation than the construction with the infinitive. It should be noted, however, that we agreed on being best friends would have been regarded an acceptable production because this study involved the assessment of grammatical, not stylistic, knowledge.
5 In cases where either of two alternatives is grammatically acceptable, either production is coded as correct. That is, in until we die, die is correctly +PRS. If the student had instead produced until we died, died would have been coded +PST because past tense is also acceptable here following English sequence of tense principles.
|
References |
|---|
|
TOPMethodResultsDiscussionConclusionAppendixReferences |
|---|
-
Albertini J, Bochner J, Cuneo C, Hunt L, Nielson R, Seago L, Shannon N. (1986) Development of a writing test for deaf college students. Teaching English to Deaf and Second-Language Students 4:25–11.
Berent GP. (1988) An assessment of syntactic capabilities. In Strong M (Ed.). Language learning and deafness(Cambridge University Press, Cambridge, MA) pp. 133–161.
Berent GP. (1993) Improvements in the English syntax of deaf college students. American Annals of the Deaf 138:55–61.[Web of Science][Medline]
Berent GP. (1996) The acquisition of English syntax by deaf learners. In Ritchie W and Bhatia T (Eds.). Handbook of second language acquisition(Academic Press, San Diego, CA) pp. 469–506.
Berent GP. (2005a) Coding deaf and hard-of-hearing students' successful and unsuccessful English productions. In Janáková D (Ed.). Proceedings 2004: Teaching English to deaf and hard-of-hearing students at secondary and tertiary levels of education in the Czech Republic(Eurolex Bohemia, Prague, Czech Republic) pp. 109–115.
Berent GP. (2005b) Input enhancement in teaching English to deaf and hard-of-hearing students. Proceedings 2004: Teaching English to deaf and hard-of-hearing students at secondary and tertiary levels of education in the Czech Republic (Eurolex BohemiaIn Janáková D (Ed.). , Prague, Czech Republic)75–87.
Berent GP, Aldersley S, Kelly RR, Schmitz KL, Khalsa BK, Panara J, Keenan S. ( June 2005) Focus-on-form teaching methods promote deaf college students' improvement of English grammar skills(Paper presented at the Fourth Meeting of the English Think Tank, Literacy in the English Classroom and Beyond, National Technical Institute for the Deaf, Rochester, NY).
Berent GP, Samar VJ, Kelly R, Berent R, Bochner J, Albertini J, Sacken J. (1996) Validity of indirect assessment of writing competency for deaf and hard-of-hearing college students. Journal of Deaf Studies and Deaf Education 1:167–178.
Borg WR and Gall MD. (1983) Educational research: An introduction 4th ed (Longman, Inc, New York).
Chomsky N. (1986) Knowledge of language: Its nature, origin, and use(Praeger, New York).
Cowan R and Hatasa YA. (1994) Investigating the validity and reliability of native speaker and second-language learner judgments about sentences. In Tarone EE, Gass SM, Cohen AD (Eds.). Research methodology in second-language acquisition(Lawrence Erlbaum Associates, Hillsdale, NJ) pp. 287–302.
Crandall KA. (1997) Review and selection of tests for placement of deaf college students in NTID's nonfiction reading curriculum: A Curriculum Development Team report(National Technical Institute for the Deaf at the Rochester Institute of Technology, Rochester, NY).
Cuculick JA and Kelly RR. (2003) Relating deaf students' reading and language scores at college entry to their degree completion rates. American Annals of the Deaf 148:279–286.[Web of Science][Medline]
Doughty C. (1991) Second language instruction does make a difference. Studies in Second Language Acquisition 13:431–469.
Focus on form in classroom second language acquisition (1998) (Cambridge University PressIn Doughty C and Williams J (Eds.). , New York).
Ellis R. (2002) Does form-focused instruction affect the acquisition of implicit knowledge? A review of the research. Studies in Second Language Acquisition 24:223–236.[CrossRef]
Gass SM. (1994) The reliability of second-language grammaticality judgments. In Tarone EE, Gass SM, Cohen AD (Eds.). Research methodology in second-language acquisition(Lawrence Erlbaum Associates, Hillsdale, NJ) pp. 303–322.
Gass SM. (1997) Input, interaction, and the second language learner(Lawrence Erlbaum Associates, Mahwah, NJ).
Han Z-H. ( February 2005) Input enhancement: Untangling the tangles, Keynote presentation at the 27th Winter Applied Linguistics Conference, New York State Teachers of English to Speakers of Other Languages, New York.
Han Z-H, Park ES, Combs C. (2005) Input enhancement: A critical meta-synthesis of the research, Unpublished manuscript, Teachers College, Columbia University.
Hollywood and Broadway, (n.d.) Retrieved September 1, 2004, from http://www.geocities.com/yamataro670/hollywood-broadway.htm.
Izumi S. (2002) Output, input enhancement, and the noticing hypothesis: An experimental study on ESL relativization. Studies in Second Language Acquisition 24:541–577.
Jourdenais R, Ota M, Stauffer S, Boyson B, Doughty C. (1995) Does textual enhancement promote noticing? A think-aloud protocol analysis. In Schmidt R (Ed.). Attention and awareness in foreign language learning(University of Hawaii Press, Honolulu, HI) pp. 183–216.
Leeman J, Arteagoitia I, Fridman B, Doughty C. (1995) Integrating attention to form with meaning: Focus on form in content-based Spanish instruction. In Schmidt R (Ed.). Attention and awareness in foreign language learning(University of Hawaii Press, Honolulu, HI) pp. 217–258.
Leow R. (1997) The effects of input enhancement and text length on adult L2 readers' comprehension and intake in second language acquisition. Applied Language Learning 8:151–182.
Long M. (1991) Focus on form: A design feature in language teaching methodology. In de Bot K, Coste D, Kramsch C, Ginsberg R (Eds.). Foreign language research in a cross-cultural perspective(John Benjamins, Amsterdam) pp. 39–52.
Michigan Test of English Language Proficiency. (1977) (University of Michigan, English Language Institute, Ann Arbor, MI).
Overstreet M. (1998) Text enhancement and content familiarity: The focus of learner attention. Spanish Applied Linguistics 2:229–258.
Quigley SP and King CM. (1980) Syntactic performance of hearing impaired and normal hearing individuals. Applied Psycholinguistics 1:329–356.
Schley S and Albertini JA. (2005) Assessing the writing of deaf college students: Re-evaluating a direct assessment of writing. Journal of Deaf Studies and Deaf Education 10:96–105.
Schmidt R. (1995) Consciousness and foreign language learning. In Schmidt R (Ed.). Attention and awareness in foreign language learning(University of Hawaii Press, Honolulu, HI) pp. 1–63.
Sharwood Smith M. (1993) Input enhancement in instructed SLA. Studies in Second Language Acquisition 15:165–179.
Sharwood Smith M. (1994) Second language learning: Theoretical foundations(Longman, London).
Shook D. (1994) FL/L2 reading, grammatical information, and the input-to-intake hypothesis. Applied Language Learning 5:57–93.
Shook D. (1999) What foreign language reading recalls reveal about the input-to-intake phenomenon. Applied Language Learning 10:39–76.
Snedecor GW and Cochran WG. (1980) Statistical methods 7th ed (Iowa State University Press, Ames, IA).
Tiegs EW and Clark WW. (1957) California Achievement Tests—Junior high level(California Test Bureau, Monterey, CA).
Traxler CB. (2000) The Stanford Achievement Test, 9th Edition: National norming and performance standards for deaf and hard-of-hearing students. Journal of Deaf Studies and Deaf Education 5:337–348.
Wilbur RB, Goodhart WC, Montandon E. (1983) Comprehension of nine syntactic structures by hearing-impaired students. The Volta Review 85:328–345.
Williams J. (2005) Form-focused instruction. In Hinkel E (Ed.). Handbook of research in second language teaching and learning(Lawrence Erlbaum Associates, Mahwah, NJ) pp. 671–691.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S.-F. Cheng and S. Rose Investigating the Technical Adequacy of Curriculum-Based Measurement in Written Expression for Students Who Are Deaf or Hard of Hearing J. Deaf Stud. Deaf Educ., October 1, 2009; 14(4): 503 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. P. Berent, R. R. Kelly, K. L. Schmitz, and P. Kenney Visual Input Enhancement via Essay Coding Results in Deaf Learners' Long-Term Retention of Improved English Grammatical Knowledge J. Deaf Stud. Deaf Educ., April 1, 2009; 14(2): 190 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Han, E. S. Park, and C. Combs Textual Enhancement of Input: Issues and Possibilities Applied Linguistics, December 1, 2008; 29(4): 597 - 618. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||