Using Online Chat to Foster the Written Language Development of Students Who Are Deaf

Barbara R. Schirmer
Albert L. Ingram

For a printer-ready version of this article, click here. Use your browser’s “back” button to return to the Reading Online site. Abstract The purpose of the study described in this article was to research the efficacy of an instructional strategy for fostering writing development among students who are deaf. The strategy was designed to capitalize on the advantages and motivating influence of an online chat environment among peers. Two experiments were carried out, one involving pairs of teenagers and the second pairs of younger adolescents. Each pair included one deaf student and one student with no hearing loss. The students met daily online to discuss an astronomy topic. In the instructional intervention, one of the investigators joined the discussion to recast the deaf students’ written language using language structures that were not appearing consistently in their spontaneous writing. Results indicate that the strategy was effective in promoting significant increases in the use of the targeted language structure for one teenage student who completed the experiment; although all younger adolescents completed the experiment, results did not provide evidence for the effectiveness of the strategy for this age group.

Related Postings from the Archives A Teacher Takes on the Challenges of Deaf Literacy by Michael Milone Writing and Communication Technologies by Colette Daiute Reading and Deaf Children by Mardi Loeterman et al.

In order to express themselves in writing, individuals must be able to manipulate the syntactic and morphologic structures of the language. For children with a hearing loss who are not fluent in spoken language or a face-to-face form of sign language, the task is daunting. Preschool children who are deaf have been found to demonstrate developmentally appropriate knowledge and understanding of written language and uses of literacy, even when their language acquisition is delayed when compared to that of hearing children (Rottenberg & Searfoss, 1992; Williams, 1994; Williams & McLean, 1997). However, research has shown that as deaf children are engaged in formal reading and writing instruction in school, delayed language development appears to affect their literacy development (Holt, 1993; LaSasso & Mobley, 1997).

(For further information on the education of deaf students, we encourage the reader to explore the Deaf Education website and the Laurent Clerc National Deaf Education Center website. For guidelines on using appropriate language when describing an individual with a disability, we refer the reader to the American Psychological Association website. “People-first language” is used to denote terminology that puts the person first and the disability second -- as in “child with autism” instead of “autistic child.” The term became common in the United States in1990, when the Congress amended the Education of All Handicapped Children Act and renamed it the Individuals With Disabilities Act. Note, however, that although people-first language is preferred when referring to most individuals with disabilities, in the fields of deaf studies and education of students who are deaf, disability-first language is not only considered appropriate, it is often preferred.)

Writing instruction most frequently follows the product or the process approach. Product approaches focus the teacher’s and the child’s attention on how well the child’s completed compositions represent the skills and rules he or she has been taught and has been given opportunities to practice (Tompkins, 2000). Process approaches focus the teacher’s and the child’s attention on the child’s thinking, from inception of idea to completed composition. Children are encouraged to engage in the stages of the writing process (planning, drafting, revising, and editing), to write at their own pace about topics of their choice, and to determine whether a composition will be completed and shared with an audience (Calkins, 1994). Few studies in the research literature have shown the success of either of these approaches to writing instruction with students who are deaf (Kluwin & Kelly, 1992; Schleper, 1996; Strassman & D’Amore, 2002).

Given that conversation is the milieu of language acquisition for children (Snow, 1986), written conversations would appear to offer a potentially rich milieu for written language acquisition. Written conversations in the form of dialogue journals and e-mail exchanges have been used by teachers, but there is little evidence in the literature that these strategies have resulted in improved use of written language among deaf students (Johnson, 1997; Lieberth, 1991; Staton, 1985, 1988).

Educators are increasingly using online chat to provide interactive learning experiences, and these environments have the advantage of being highly motivating to students (Holmevik & Haynes, 2000; Kearsley, 2000). Although many observers see only the social aspects of online chats, some note that teachers can use them to extend the educational experience beyond the classroom -- with special benefits perhaps for writing instruction (McGrath, 1997-1998), for students with learning and perceptual disabilities, and for those learning a second language (Ortega, 1997). Relatively few empirical studies have been conducted that actually demonstrate these advantages, however.

If it were possible for written conversations to support written language development, then it would seem logical to imbue written conversations with the features of spoken conversations found to promote the language development of young children. Spoken conversations between parents and young children are characterized by simple and well-formed sentences, much redundancy, clear pauses, exaggerated stress and intonation, many questions, restricted vocabulary, and clarity (DePaulo & Bonvillian, 1978; Garton & Pratt, 1998; Snow, 1986). Investigations into features of conversation that facilitate the language acquisition of deaf children have found the following to be effective:

• Contributions by the adult in the form of personal comments or statements and phatics (phrases such as “Oh, that’s interesting,” or “I see!” spoken with head nodding in response to the child’s contribution, or repeating the child’s comments) (Wood & Wood, 1984; Wood, Wood, Griffiths, Howarth, & Howarth, 1982)
• Expanding and extending what the child has said (Farrar, 1992; Nelson, Welsh, Camarata, & Butkovsky, 1995)
• Adult recasting (Prinz & Masin, 1985)

Compared to e-mail or other asynchronous communications, online chats can provide a closer match to face-to-face conversation because the dialogue can take place in “real time.” When used instructionally, the teacher can monitor these conversations, incorporate conversational features that are particularly nurturing for language acquisition, and apply conversational strategies deliberately in instructional contexts. One such strategy was investigated by Prinz and Masin (1985), who conducted a study of recasting with a group of deaf children to determine if the intervention would improve acquisition of targeted syntactic structures. They defined recasting as a particular kind of adult response to a child’s utterance. It occurs when the adult response is signed or spoken in a different syntactic structure than the child just used, but maintains the central meaning expressed by the child. The participants were six children between the ages of 9 months and 6 years at the outset of the study. Two children had deaf parents who used American Sign Language (ASL) in the home, and the other children had hearing parents whose language with the children could best be described as manually coded English or pidgin sign language. The teachers used pidgin sign, which the researchers characterized as “combining semantic and syntactic aspects of both ASL and signed English” (p. 360).

The children’s parents and teachers were given training on how to recast the children’s utterances into the specific syntactic structures that the researchers had targeted. The study lasted for five months, during which each child received at least 20 hours of intervention. Results showed that the targeted structures did appear in the spontaneous communication of the children. These results leave open the question of whether recasting can be used successfully with written language.

The purpose of the study described here was to design and research an instructional intervention for fostering the writing development of students who are deaf that capitalized on the unique advantages of online chat. The following question guided the study: Can specific features of the written language of students who are deaf be improved through teacher recasting during online discussion?

Back to menu

Method and Results

A single-subject experimental research design was used to establish the effects of the intervention on the written language of deaf students in high school (experiment 1) and middle school (experiment 2). The predominant feature of this research design is its focus on intra-subject comparisons, rather than on the inter-subject comparisons found in other quantitative designs (e.g., experimental and quasi-experimental research). Baseline data are collected prior to the intervention, intervention data are gathered continuously, all data are graphed, and a visual rather than statistical analysis is conducted for each participant’s graphed data. As noted by Cook, Rumrill, Webb, and Tankersley (2001), “Because single subject research is intended to focus on the change in behavior within the subject, the results are interpreted in terms of clinical rather than statistical significance” (p. 133).

In addition, a qualitative content analysis of interviews with the students at the conclusion of each experiment was conducted.

A convenience sample of intact classrooms was used from which students were selected based on teacher nomination. Each student used an Internet-ready computer on which The Palace client software had been uploaded. Unlike many ways available for implementing an online chat environment, The Palace provides a two-dimensional graphical environment rather than a purely textual one (Ingram, Hathorn, & Evans, 2000). Participants see on screen a representation of a room or other setting in the background, and avatars (visual representations) of all participants in the foreground. The avatars can be customized with props of various kinds, and the participants’ typed comments appear both in dialogue balloons attached to the appropriate avatar and in a log window that maintains a record of the entire conversation. Participants can move from room to room or within a room by clicking the mouse, and they are involved in conversations only in the room they currently occupy.

In this study, the participants met in a virtual representation of the parking lot of Kent State University’s White Hall, and moved indoors to a virtual lounge for their discussions. In this room, the discussions could take place privately because the door could be locked once all avatars were inside. Thus, other visitors to The White Hall Palace site were unable to observe or join the discussion.

Experiment 1

Participants. The participants were three deaf students and three hearing students. The former were attending a state program for the deaf at a school where all students are deaf and all teachers are certified to teach deaf students. The three students, who were nominated by their science/social studies teacher to participate in the study because of their academic level and keyboarding skills, communicated in ASL and were identified audiologically with profound hearing losses. The hearing students were attending a special technology-rich classroom at Kent State University in Kent, Ohio, USA, for a period of several weeks. The classroom teacher nominated three students whose independence and maturity indicated they could successfully participate in the study.

Three pairs were formed randomly so that each pair consisted of a deaf student and hearing student. One of the investigators, an experienced teacher of the deaf and a teacher educator, served as online teacher. One investigator and a research assistant served as onsite facilitators, at the Kent State classroom and the school for the deaf classroom, respectively.

Procedure. The software was installed at each site and the students were given an overview of the project. The schedule of days and times for online chat was coordinated between the two sites, which were 135 miles (220 km) apart. Each pair was given ten minutes of time for online discussion each day, for a period of three weeks.

The first day, the students were encouraged by the online teacher to tell about themselves, ask questions of each other, and explore the features of The Palace. Baseline data on the targeted language structure, descriptors (words used to modify or describe a noun or verb), were collected beginning on the second day. The students’ classroom teacher chose descriptors as the dependent variable from a list (provided by the investigator/online teacher) of structures commonly found in early language acquisition (Brown, 1973) but that did not appear with consistency in the students’ written language.

On the first day, the students were told they would work together to develop a report on astronomy. The online teacher gave them a list of ten topics, each accompanied by a question. The teacher asked the students first to choose the three topics they liked best, and then to decide on one topic for the project. Pair A-X chose stars -- Why do stars twinkle? Pair B-Y chose sun -- Will the sun explode when it runs out of fuel? Pair C-Z chose comets -- Why do comets have tails? Each day the students discussed their topic and what they had learned through Internet research during the 20 minutes each day they worked outside The Palace, as well as in the evening at home.

The students were not told that the online teacher’s contributions to conversations in The Palace would be a recasting of the language of the deaf students, because the intervention was designed to replicate in a written language milieu the conversational features that facilitate language acquisition of hearing children in an oral language milieu. Just as the adult does not point out to the child that he or she is recasting the child’s language in oral conversation, we did not point out to the pairs that the online teacher was recasting the deaf participant’s written language in the online chats. For example, in oral conversation between an adult and child, the child might say, “I see a bird.” The adult might respond by saying, “Yes, I see a pretty blue bird.” This is typical of the kind of exchange between adults and young children that has been found to facilitate language acquisition. The adult never says, “I will be adding descriptors to your sentence so that when you say something like ‘I see a bird,’ I’ll respond by adding ‘pretty’ and ‘blue’ to make the sentence, ‘I see a pretty blue bird.’”

In single-subject research design, changes in the participant’s response are investigated by comparing pre-intervention performance with intervention performance. Pre-intervention performance is gathered as baseline data. During baseline, the online teacher engaged the students in the same type of discussion that they would have in the intervention phase, except that no recasting was carried out. Thus, the instructional conditions were controlled during baseline so that the only difference between the two phases was the intervention strategy of recasting. Baseline was maintained for three days because that was the point at which the data fell within a narrow range (see Kazdin, 1982; Neuman & McCormick, 1995).

During intervention, the online teacher recast the deaf student’s written language using the targeted language structure. In recasting, the online teacher attempted to include a descriptor not used in the student’s written comment and to frame responses to be

• Close in time and direct
• Well-formed syntactically
• Understandable by the student
• Closely related to the meaning and structure of the student’s comment
• Not a grammatical correction of the student’s comment

In the following examples, the target language structure is underlined. Except for this underlining, the language of the child and teacher are reproduced exactly as written in the chat. Just as in oral conversations, the pattern of talk did not follow from person to person evenly, and comments were not always responsive to the preceding statement. Also, students were reading the chat, typing, and reading their research notes during the online chat. Some comments went unanswered, and we could not know for certain whether they were read and ignored or missed while students’ attention was directed at writing or reading research notes.

Two weeks after the conclusion of the study, the two available deaf students and the three hearing students were interviewed. The research assistant who interviewed the deaf students is proficient in ASL, an experienced teacher of the deaf, and has a severe hearing loss. Each interview was taped (audiotape for the hearing students and videotape for the deaf students) and transcribed by the onsite facilitator. The interview protocol, which can be found in the appendix, was designed to elicit responses about the students’ attitude toward the activity and awareness of the writing strategies they were using.

Analysis. The Palace server software can be configured to record logs of all conversations that take place on it. These server logs were the main source of quantitative data. The investigators counted the number of descriptors appearing in the deaf students’ written language on a daily basis. The data were graphed, and visual analysis was used to determine if a change had occurred; if a change was noted, its magnitude, trend, and latency were recorded, along with a determination of whether the change appeared reliable (see Kazdin, 1982; Neuman & McCormick, 1995).

The dependent variable was the number of descriptors. The independent variable was teacher recasting of each student’s written language through use of descriptors within a well-formed and understandable, though not necessarily grammatically complete, written response.

For the dependent variable, interobserver agreement was determined by comparing the number of descriptors noted by each observer for a random selection of 10 percent of student logs during baseline and intervention. Agreement was found to be 85.4 percent. For the independent variable, the steps of the recasting procedure were listed, 10 percent of baseline and intervention logs were randomly selected, and each observer checked “yes” or “no” to indicate whether the online teacher followed each step. Agreement was found to be 100 percent. The 10 percent figure was chosen as being in line with single-subject research design in which “agreement needs to be assessed within each phase of the investigation, preferably at least a few times within each phase” (Kazdin, 1982, p. 51).

Interviews with the deaf students were transcribed by the research assistant, who had also conducted the interviews. A direct word-for-word transcription was not possible because ASL and English do not share a common syntactic structure; the transcripts thus represented translations that retained the meaning expressed by the students. Interviews with the hearing students were transcribed word for word by the investigator who had conducted them. The transcribed interviews were analyzed using content analysis (LeCompte, Preissle, & Tesch, 1993; Patton, 1990). The transcripts were read several times, and notes related to student attitude toward the activity and awareness of strategies were made directly on the copies.

Results. Due to absences of students A and C, only pair B-Y completed the study. As shown in Figure 1, data points included three days of baseline and seven days of intervention. (Note that in all figures, nonacademic days include sessions devoted to introductions, reintroductions after holidays, and absences.) Visual analysis of the graphed data indicate that intervention was effective in promoting significant increases in Student B’s use of descriptors, the targeted language structure. Student B used a maximum of four descriptors during baseline. The first day of intervention, she used 13 descriptors; her use of this language structure ranged from 5 to 30 occurrences over the seven days of intervention. For pairs A-X and C-Z, data points included three days of baseline and two days of intervention. Therefore, no conclusions could be drawn about the effectiveness of the intervention for these two pairs.

Content analysis of the interviews indicated that both the deaf and hearing students considered the topic (astronomy) to be fun and interesting, The Palace to be a “neat” way to communicate, and the activity to be a nice change from usual classroom activities. For example, Student B commented, “Involve friend thought that neat and fun.” Student A noted, “I think it’s pretty cool and fun to learn. It was different. Class is always the same. Think math, science etc. etc. Learning stars is fun.” Hearing Student Y said, “It was neat. I liked it. It was different.” However, the deaf students found it difficult to find information on the Internet that they could share with their partner online. For example, Student A said, “I enjoyed it but it’s hard searching the Internet for information.

In terms of awareness of strategy use, the deaf students realized that the online teacher’s role was to encourage discussion about the topic and to provide assistance in explaining vocabulary and concepts used by the hearing students. For example, Student B said, “If my partner’s talk type is an impossible word I didn’t understand, then I asked [the online teacher] what that means. Then I understand what it means and it helps me.” Both the hearing and deaf students understood that their role was to share ideas with their partner and that their partner was supposed to share ideas with them. For example, deaf Student A commented, “I liked my partner giving me some information. It helped me.” Hearing Student Z said, “I learned stuff from him, and I’m pretty sure he learned stuff from me.... I was trying to do my best and find out good information that would let him know, you know, give him some idea of what I know and see if he could run with it.”

Experiment 2

Participants. The participants were four deaf students and four hearing students, ranging in age from 10 to 12 years. The deaf students attended a program at public middle school in Oregon, in which all of the students in the class are deaf and the classroom teacher is a certified teacher of deaf students; however, the deaf students are integrated into some subject areas with hearing peers, depending on each deaf student’s Individualized Education Program. All students communicated in manually coded English (a sign language system that represents English in a visual-gestural modality) and were identified with severe to profound hearing losses. The classroom teacher nominated four students whose academic level and keyboarding skills indicated they could participate in the study successfully.

The hearing students were attending a special technology-rich classroom at Kent State University for a period of several weeks. The classroom teacher nominated four students whose independence and maturity indicated they could participate in the study successfully.

Four pairs were formed randomly so that each pair consisted of a deaf student and hearing student. The investigator who served as the online teacher in the first experiment also served as the online teacher in this experiment. One investigator served as the onsite facilitator at the Kent State classroom; the classroom teacher served as the online facilitator in the deaf students’ classroom.

Procedure. The software was installed at each site and the students were given an overview of the project. The schedule of days and times for online work was coordinated between the two sites, which were 2000 miles (3200 km) and three time zones apart. Four weeks were scheduled for the study: two weeks before winter break and two weeks after. It was possible to schedule two 30-minutes segments daily, so each pair was given 15 minutes of time for online discussion each day.

As in the first experiment, on the first day, the students were encouraged by the online teacher to tell about themselves, ask questions of each other, and explore the features of The Palace. Baseline data on the targeted language structure, conjunctions, were collected beginning on the second day. Conjunctions were chosen as the dependent variable by the teacher from a list of structures commonly found in early language acquisition (Brown, 1973) but that did not appear with consistency in the students’ written language.

The students were told they would work together to develop a report on astronomy for a younger child to read. As in the first experiment, the online teacher gave them a list of ten topics, each accompanied by a question. The teacher asked the students first to choose three topics they liked, and then to decide on one topic for the project. Pair D-T chose galaxies -- Can humans travel to other galaxies? Pairs E-U and G-W chose Mars -- Was there ever water on Mars? Pair F-V chose Mars and Venus -- Was there ever water on Mars and could humans live on Venus? As in the first experiment, each day the students discussed their topic and what they had learned about it through Internet research. They were not told that the online teacher would be recasting the language of the deaf students.

Baseline was maintained for three days because that was the point at which the data fell within a narrow range (see Kazdin, 1982; Neuman & McCormick, 1995). As in the first experiment, the intervention consisted of the online teacher recasting the deaf student’s written language using the targeted language structure. In the following examples, the target language structure is underlined. Except for the underlining, the language of the children and teacher are exactly as written in the chat. These excerpts are briefer than those presented in the description of the first experiment. Just as the high school students had done, these middle school students were reading the chat, typing, and reading their research notes while conversing online. We surmised that their reading and writing abilities were weaker than those of the high school students, with the result that these younger children made fewer written comments. Just as with the high school students, some comments went unanswered and we could not know whether they had been read and ignored or simply missed while the students’ attention was directed at writing or reading research notes.

The day after intervention ended, the students were interviewed online. This format was used because neither investigator could travel to the deaf students’ classroom within two weeks of the investigation’s conclusion. The original interview protocol was not followed because the number of questions was beyond the scope of a 15-minute session. Instead, the online teacher asked the students what they liked and didn’t like about the online chats, the astronomy project, and The Palace.

Analysis. As in the first experiment, The Palace server logs were the main source of quantitative data. The investigators counted the number of conjunctions that appeared in the deaf students’ written language on a daily basis. The data were graphed and visual analysis was used to determine if a change had occurred; if a change was noted, its magnitude, trend, and latency were recorded, along with a determination of whether the change appeared reliable (see Kazdin, 1982; Neuman & McCormick, 1995).

The dependent variable was the number of conjunctions. The independent variable was teacher recasting of each student’s written language through use of conjunctions within a well-formed and understandable, though not necessarily grammatically complete, written response.

For the dependent variable, interobserver agreement was determined by comparing each observer’s determination of the number of conjunctions for a random selection of 10 percent of the student logs during baseline and intervention. Agreement was found to be 88.6 percent. For the independent variable, the steps of the recasting procedure were listed, 10 percent of randomly selected baseline and intervention logs were selected, and each observer checked “yes” or “no” to indicate whether the online teacher followed each step. Agreement was found to be 100 percent.

As in the first experiment, the interviews were analyzed using content analysis (LeCompte, Preissle, & Tesch, 1993; Patton, 1990).

Results. As shown in Figures 2 to 5, data points included three days of baseline for all pairs. Pairs D-T and E-U had 11 days of intervention, pair F-V had six days, and pair G-W had eight days. Visual analysis of graphed data indicated that the use of conjunctions in the deaf students’ written language showed sporadic increases from baseline. For example, Student F showed an increase from one to three conjunctions the first day of intervention. However, she was absent the next day and, on the day she returned to school, she used no conjunctions. Student G used two conjunctions on one day of baseline, but none on the other two days. She showed consistent use of one or two conjunctions each day of the intervention and six conjunctions on two different days, but no pattern was discerned in trend or latency. Student D used conjunctions on only 5 of the 11 days of intervention, and Student E used conjunctions on only 3 of the 11 days. Given the lack of reliable change, no effects of the intervention could be established.

Given the absence of positive findings for the intervention, we examined the data to determine if we could establish the reasons for variability in the number of conjunctions used by the participants. In single-subject research design, visual analysis of graphed data provides a more conservative view of the strength and stability of the relationship between the intervention and student performance than measures of statistical significance might show; thus, studies that follow this design are more likely to result in the screening out of weak variables than statistical manipulation of data (Heward, 1987). We found no relation between the number of recastings by the online teacher and participants’ use of conjunctions. We also found no relationship between mean length of utterance (i.e., the average length of each participant’s sentences) and the number of conjunctions used. We considered the possibility that the hearing partner’s use of conjunctions might influence the deaf partner’s use of conjunctions; however, we found no such relation.

Content analysis of the interviews indicated that in terms of attitude, both the deaf and hearing students enjoyed being online and liked the astronomy project. For example, Student E noted, “i have fun working to together.” Student V wrote, “I think that Mars was a good subject!!!” Student E noted, “I like talk about Juiter.” For awareness of strategy use, they realized they needed the online teacher and each other. For example, Student T wrote, “The chat thing had to be the best. I’m glad you were here because we wouldn’t have gotten anything done.” Student D wrote, “and I like [Student T]’s what he know.” They also realized that they had difficulty typing as fast as they were thinking. Student D commented, “i dont like is typing when I want to say.”

Back to menu

Discussion

The study was designed to determine if the instructional intervention of teacher recasting during online discussion could foster the writing development of deaf students by increasing their use of a targeted language structure. Results indicated that the intervention appeared to be effective with the one high school deaf participant who completed the study, but not effective with the middle school deaf participants. Results do not disclose whether student age was the discriminating factor in effectiveness. The intervention may have been more effective in the first experiment because certain language features (e.g., descriptors) are more salient or more amenable to teacher recasting than others (e.g., conjunctions). Results also indicated that the deaf and hearing participants had positive attitudes toward the online chat and the astronomy activity. They also recognized that the online teacher provided support and assistance for their discussions.

Limitations of the study stem largely from scheduling and technology difficulties. The first limitation was the brief time available for the online chats. For the first experiment, only 30 minutes each day for a three-week period were found to be convenient for the students at the two sites. This window of time allowed each pair to be online for only 10 minutes daily, and inevitably a few minutes were lost as students signed on and off, particularly during the first few days of the study when the students were not familiar with The Palace. For the second experiment, a 60-minute convenient time was identified, enabling each pair to be online for 15 minutes, but the experiment was bifurcated by the winter break. Thus, a day for reintroductions was needed the first day online after vacation.

The second limitation was the number of online sessions. When one partner was absent, whether because of illness or a school event, the pair missed a day of intervention. With only 15 possible days in experiment 1 and 20 possible days in experiment 2, each absence was an important loss. Although the research literature offered no benchmark for number of days of intervention, we anticipated a minimum of 10 continuous days for effectiveness. None of the pairs met this minimum. For the high school pair, the seven total days of intervention were interrupted by two days of absence. For two of the middle school pairs, nine of the intervention days were continuous after winter break. For the third middle school pair, all of the eight intervention days took place consecutively after winter break. For the fourth middle school pair, five of the six intervention days took place consecutively, with an absence occurring the day after the first intervention.

The following excerpts show the sorts of discussions that took place about absences, and consumed portions of time allotted for online chat.

The third limitation was getting and staying connected. One issue concerned school and state department of education firewalls that limited public school access to a list of acceptable Internet sites. The Palace was an unknown site, and on the first scheduled day of experiment 1, the school for the deaf was unable to connect to it. (The Kent State University network had no such firewall.) Due to the intercession of the school superintendent, the firewall was taken down two days later. In the interim, the onsite facilitator brought in her personal laptop computer and connected via modem to her Internet provider. Thus, one day was lost in the first experiment. For the second experiment, the onsite teacher had the students use her laptop computer from the outset, in order to avoid needing to request removal of the firewall.

A second technical issue was server uptime. In both experiments, one day was lost to server or network downtime at the school site or at Kent State. The following excerpts from two sessions shows discussion among the two investigators (online teacher and investigator on site at Kent State) and research assistant onsite at the school for the deaf about Internet connectivity; the first excerpt is from experiment 1 and the second from experiment 2.

The fourth limitation was the keyboarding skills of the students. Real-time chats require keyboarding competence. Although we did not assess the students’ keyboarding skills, we assumed that the hearing students attending the technology-rich classroom at Kent State would have keyboarding competence, and keyboarding skill was one of the criteria for teacher nomination of the deaf students. Yet the middle school students’ comments during the interview indicated that they felt challenged in trying to type to keep pace with their ideas, and the contributions of the deaf students indicated that they may have had more difficulty than their hearing peers since they were dealing not only with keyboarding skills but with written language skills as well.

The following excerpt illustrates the limited number of written responses typical within each session for the students in the second experiment.

Several implications emerge from the study. The narrowest relate to the use of the intervention in writing instruction with deaf students. Recasting during online chats appears to have potential as an intervention with deaf students at high school age, but not with younger deaf students. The intervention assumes that the student knows the targeted language structure and is able to use it in speaking (or signing) but is not using the structure at all, or not using it consistently, in writing. The younger the student, the more likely it is that this assumption is erroneous. However, if the intervention was effective in the first experiment because descriptors are more salient or amenable to recasting than conjunctions, then increasing the salience of targeted language features and selecting features that lend themselves to recasting may be crucial to the success of the intervention.

The intervention also assumes that the time online offers equal access for the two students and online teacher to participate in the chat. However, written conversations online may not reflect the conversational turn taking of spoken conversations. Thus, the deaf students may not have had sufficient opportunity to demonstrate their ability to incorporate the targeted language structures into their written comments, particularly in light of their difficulty with written English. Future research should involve deaf students at the upper middle school and lower high school levels to determine if there is an optimal age and developmental level for the intervention; it should also examine the transfer to offline writing as well as the maintenance of changes over time.

The broadest implications relate to the use of online chats for teaching and learning. Most synchronous online discussions in educational settings are focused on the content of the course at hand, at least after the initial period of play and exploration are complete. The procedures researched in this study open the door to using online chats to modify specific skills. For example, most courses in higher education and other settings in which chats might be used demand extensive writing. Conceivably, instructors could begin to modify writing behavior through online strategies such as recasting. In addition, the current emphasis on collaborative teaching and learning methods suggests that teachers might increase the frequency and effectiveness of truly collaborative behaviors by using a variety of methods to encourage them in online chats.

The point is that teachers can use computer-mediated communications for a variety of purposes. Teachers at many levels may have to learn new technical abilities in order to affect their students’ behaviors and skills.

Back to menu

References

Brown, R. (1973). A first language: The early stages. Cambridge, MA: Harvard University Press.
Back

Calkins, L.M. (1994). The art of teaching writing (2nd ed.). Portsmouth, NH: Heinemann.
Back

Cook, B.G., Rumrill, P.D., Jr., Webb, J., & Tankersley, M. (2001). Quantitative research designs. In P.D. Rumrill, Jr. & B.G. Cook (Eds.), Research in special education: Designs, methods, and applications (pp. 121-158). Springfield, IL: Charles C. Thomas.
Back

DePaulo, B.M., & Bonvillian, J.D. (1978). The effect on language development of the special characteristics of speech addressed to children. Journal of Psycholinguistic Research, 7, 189-211.
Back

Farrar, J. (1992). Negative evidence and grammatical morpheme acquisition. Developmental Psychology, 28, 90-98.
Back

Garton, A., & Pratt, C. (1998). Learning to be literate: The development of spoken and written language. Oxford, UK: Basil Blackwell.
Back

Heward, W.L. (1987). Production and display of graphed data displays. In J.O. Cooper, T.E. Heron, & W.L. Heward (Eds.), Applied behavior analysis (pp. 106-141). Columbus, OH: Merrill.
Back

Holmevik, J.R., & Haynes, C. (2000). MOOniversity: A student’s guide to online learning environments. Boston: Allyn & Bacon.
Back

Holt, J. (1993). Stanford Achievement Test -- 8th edition: Reading comprehension subgroup results. American Annals of the Deaf, 138, 172-175.
Back

Ingram, A.L., Hathorn, L.G., & Evans, A.D. (2000). Beyond chat on the Internet. Computers and Education, 35(1), 21-35.
Back

Johnson, H.A. (1997, June). Internet use within deaf education: Current status, trends and applications. Paper presented at the Convention of American Instructors of the Deaf/Council of Educational Administrators of Schools for the Deaf, Hartford, Connecticut.
Back

Kazdin, A.E. (1982). Single-case research designs: Methods for clinical and applied settings. New York: Oxford University Press.
Back

Kearsley, G. (2000). Online education: Learning and teaching in cyberspace. Belmont, CA: Wadsworth.
Back

Kluwin, T.N., & Kelly, A.B. (1992). Implementing a successful writing program in public schools for students who are deaf. Exceptional Children, 59, 41-53.
Back

LaSasso, C.J., & Mobley, R.T. (1997). National survey of reading instruction for deaf or hard-of-hearing students in the U.S. The Volta Review, 99, 31-58.
Back

LeCompte, M.D., Preissle, J., & Tesch, R. (1993). Ethnography and qualitative design in educational research (2nd ed.). San Diego, CA: Academic.
Back

Lieberth, A.K. (1991). Use of scaffolded dialogue journals to teach writing to deaf students. Teaching English to Deaf and Second-Language Students, 9(1), 10-13.
Back

McGrath, C. (1997-1998). A new voice on interchange: Is it talking or writing? Implications for the teaching of literature. Journal of Educational Technology Systems, 26, 294-297.
Back

Nelson, K.E., Welsh, J., Camarata, S., & Butkovsky, N. (1995). Available input for language-impaired children and younger children of matched language levels. First Language, 15, 1-17.
Back

Neuman, S.B., & McCormick, S. (1995). Single-subject experimental research: Applications for literacy. Newark, DE: International Reading Association.
Back

Ortega, L. (1997). Processes and outcomes in networked classroom interaction: Defining the research agenda for L2 computer-assisted classroom discussion. Language Learning and Technology, 1(1), 82-93.
Back

Patton, M.Q. (1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage.
Back

Prinz, P.M., & Masin, L. (1985). Lending a helping hand: Linguistic input and sign language acquisition in deaf children. Applied Psycholinguistics, 6, 357-370.
Back

Rottenberg, C.J., & Searfoss, L.W. (1992). Becoming literate in a preschool class: Literacy development of hearing-impaired children. Journal of Reading Behavior, 24, 463-479.
Back

Schleper, D.R. (1996). Write that one down! Using anecdotal records to inform our teaching. The Volta Review, 98, 201-208.
Back

Snow, C.E. (1986). Conversations with children. In P. Fletcher & M. Garman (Eds.), Language acquisition: Studies in first language development (pp. 69-89). Cambridge, UK: Cambridge University Press.
Back

Staton, J. (1985). Using dialogue journals for developing thinking, reading, and writing with hearing-impaired students. The Volta Review, 87, 127-154.
Back

Staton, J. (1988). Dialogue journals. Language Arts, 65, 198-201.
Back

Strassman, B.K., & D’Amore, M. (2002). The write technology. Teaching Exceptional Children, 34, 28-31.
Back

Tompkins, G.E. (2000). Teaching writing: Balancing process and product (3rd ed.). Upper Saddle River, NJ: Prentice Hall.
Back

Williams, C.L. (1994). The language and literacy worlds of three profoundly deaf children. Reading Research Quarterly, 29, 125-155.
Back

Williams, C.L., & McLean, M.M. (1997). Young deaf children’s response to picture book reading in a preschool setting. Research in the Teaching of English, 31, 377-366.
Back

Wood, H.A., & Wood, D.J. (1984). An experimental evaluation of the effects of five styles of teacher conversation on the language of hearing-impaired children. Journal of Child Psychology, Psychiatry, and Allied Disciplines, 25, 45-62.
Back

Wood, D.J., Wood, H.A., Griffiths, A.J., Howarth, S.P., & Howarth, C.I. (1982). The structure of conversations with 6- to 10-year-old deaf children. Journal of Child Psychology, Psychiatry, and Allied Disciplines, 23, 295-308.
Back

About the Authors

Barbara Schirmer is dean of the School of Education and Allied Professions and a professor of special education at Miami University (Oxford, OH, USA). She received her Ed.D. in reading and language and B.S. in elementary education from the University at Buffalo, State University of New York, and her M.Ed. in teaching deaf and hard-of-hearing children from the University of Pittsburgh. She has taught at the West Virginia School for the Deaf and Boston School for the Deaf, been on the faculty of the University of Wisconsin-Milwaukee and Lewis & Clark College, and was department chair at Kent State University. Her research into the literacy development of deaf children has been published in numerous professional journals, and she is the author of two books published by Allyn and Bacon. Contact her by e-mail at schirmb@muohio.edu.

Albert Ingram is an associate professor of instructional technology at Kent State University (Kent, OH, USA), where he teaches courses in instructional design and technology. He received his Ph.D. in educational technology from Arizona State University. He has taught at Governors State University and Kent State University and has worked at a variety of other organizations, including Digital Equipment Corporation, the American College, the Software Engineering Institute, and the University of Medicine and Dentistry of New Jersey. He is coauthor of Exploring Current Issues in Educational Technology and FrontPage 2002: An Introduction to Web Design for Educators and Trainers. He has published papers in numerous journals, including Educational Technology, the Journal of Educational Technology Systems, Educational Technology Research and Development, the Journal of Educational Computing Research, and Computers in the Schools.

Back to top

Authors’ note: Thanks are extended to Sarah McGough, University of Illinois at Urbana-Champaign, USA, Tina Harrison, University of Science and Arts of Oklahoma, Chickasa, USA, and M. Lynn Woolsey, University of North Florida, Jacksonville, USA, for their assistance with the study.

For a printer-ready version of this article, click here.

Citation: Schirmer, B.R., & Ingram, A.L. (2003, July/August). Using online chat to foster the written language development of students who are deaf. Reading Online, 7(1). Available: http://www.readingonline.org/articles/art_index.asp?HREF=schirmer/index.html

Reading Online, www.readingonline.org
Posted July 2003
© 2003 International Reading Association, Inc.   ISSN 1096-1232