Printer Friendly

Family conversation style: its effect on the deaf child's participation.

Family Conversation Style: Its Effect on the Deaf Child's Participation

This research is concerned with documenting the family conversation experience of a small number of deaf children and their hearing parents and siblings in order to understand what features of that experience may influence deaf children to participate in conversation.

Two assumptions have guided the design of the present study. The first is that family conversation is an important aspect of family life to study. From a family-systems perspective (Minuchin, 1974), family conversation represents an important interactive vehicle for developing social skills and a sense of belonging and, thus, for carrying out the socialization and self-identification functions of the family (Turnbull & Turnbull, 1986). Studies of interactions between young deaf children and their hearing mothers reveal that these experiences are missing many of the behaviors believed necessary for developing social skills involved in communicating with peers and adults. Reports of maternal directiveness, intrusiveness, and lack of mutual enjoyment during play with preschool-aged deaf children suggest a conversational environment where the deaf child gets few opportunities to initiate new topics of activities or to try out communicative, turn-taking rules (Meadow, Greenberg, Erting, & Carmichael, 1981; Schlesinger & Meadow, 1972; Weddell-Monnig & Lumley, 1980). In hearing families with deaf children, early communication difficulties may disrupt or prevent the process of conversational interaction. If deaf children are thus limited in participating in their families interactive communication, they are at risk for restricted access to family life.

The second assumption is that the types of conversation directed to deaf children influence the children's participation. One set of factors involves the level of control over the course of conversation that is correlated with participation behavior. Previous research observing conversation between teachers and deaf and hearing students (Wood, D. J., McMahon, & Cranstoun, 1980; Wood, D. J., Wood, Griffiths, Howarth, & Howarth, 1982; Wood, H. A., & Wood, 1984) and mothers and preschool-age hearing and deaf children (Power, Wood, Wood, & MacDougall, 1990) found negative correlations between teacher and maternal control and child measures of initiative and length of turn. Teacher and maternal conversation studies also show that the rate of questioning establishes a tone in the conversation which reduces children's overall willingness to participate in all conversation presented to them. More controlling statements such as questions requiring only a one-word response reduce the initiative to contribute to conversation while less controlling open-ended questions or expressions of opinion elicit greater participation.

A question that has yet to be addressed in the research literature is the extent to which controlling conversational elements influence the deaf child's participation in a family conversation context. The present study was designed to quantify and evaluate deaf children's participation in conversation with their hearing parents and siblings and the relationship of varying levels of control in conversation directed to deaf children and their response. The hypothesis was that the child would encounter quantitatively less control in family conversations than in a classroom, but would respond to controlling and inviting statements and styles in a similar manner as deaf students do to their teachers--or as children respond when they are playing at home alone with their mothers.

The data for the present study were gathered during family dinnertime. The significance of family dinnertime as a context for the study of family conversation is demonstrated by Lewis and Feiring's (1982) study of the family as a social system. Their research of family interaction during mealtime focused on patterns of conversation, that is, on who talked to whom, and on the functions that characterize the family interaction. Tizard and Hughes (1984) used conversations occurring between mother and daughter and teacher and child to compare the home and school as learning environments for the child. They found that out of three time periods in the afternoon, most conversation occurred over lunch.



Ten families from 125 participating in a family-school relationship study (Bodner-Johnson, 1986) were chosen for conversation analysis. Each family had one deaf child with an unaided prelingual hearing loss exceeding 70 dB in the range 250-2000 Hz in the better ear with no other disability. The families were Caucasian, with two hearing parents, and used English as the primary language. Four families used the simultaneous method of communication, and six used only the oral/aural method. The mean number of children was 1.9 (range = 1-3), socioeconomic status was 2.3 (1 = high, 4 = low), and mean age of the deaf child was 11.2 years (range = 10-12.75 years). Four of the deaf children had older siblings (mean age of 13.8 years), and five had younger siblings (mean age of 5.6 years).


Families were videotaped in their homes during their evening meal. Upon arriving at the home, the trained video technician and his assistant set up the equipment in the dining area so that all family members seated at the dinner table were in range of the camera and microphone. They asked if family members had any questions and talked with them about having a typical dinnertime meal. For example, families were told to carry on their regular dinnertime behavior such as serving food, attending to a child's needs, in short, doing whatever they usually do. An effort was made to make the families feel natural and comfortable.

The technician turned on the equipment, and he and the assistant left. The videotape ran for an hour; or, when the family finished dinner, they were told they could turn it off and call in the "crew." Afterward, the families were invited to watch the taped, and they all did.

Two important concerns regarding the procedure should be addressed: What effect the equipment would have on the family's behavior and whether one videotaping session would capture a segment of typical dinnertime conversation. Regarding the first issue, discussions with researchers who have studied family conversation under similar circumstances suggested that although an elevation of behavior, or a greater number of interactions, may occur, the quality and pattern of interactions are firmly established and are less susceptible to change in this situation (C. Feiring, personal communication, April 1982). Regarding the second question, the families themselves were asked how representative the recorded dinnertime was of all of their dinnertimes. Although most said they felt a little uncomfortable at first or unnatural, none felt it to be unrepresentative. All in all, the recording was judged to be unobtrusive and effective in light of the precautions taken to record "normal" interaction.

Time-coded, running transcripts of all verbal communication (spoken and signed) were created fromthe videotapes of dinnertime conversation. Transcripts noted which family members addressed conversations to whom. When a family member addressed everyone, each person got counted as the addressee. Transcripts also noted behaviors (e.g., facial expressions, spilling milk) that enhanced understanding of the dinnertime conversation. Hearing graduate students at Gallaudet University who were proficient in sign communication transcribed the tapes. Transcribers were paired to do the task and to complement sign communication skills. (Signs and words were transcribed without distinction as verbal communication. Sign communication proficiency for family members including the deaf child was not considered in this study. The intent in the transcription was to document as carefully as possible what was communicated, regardless of method, to determine the meaning of the conversation occurring among family members.) Conventions for transcripts were developed and are availabe upon request.

Data Base

Data for the conversation analysis derive from a sample of 100 continuous conversation turns per family. A table of random numbers (Roscoe, 1969) was used to mark the start point of the 100 conversation turns out of the family's total conversation that was transcribed. Overall, the 100 turns averaged approximately 7 sequential minutes of conversation per family (range of 4:39 minutes to 8:48 minutes). Each of the 100 turns in the family conversations was coded as a particular type of controlling move. A move is defined as the last statement in a turn. The coding system used was derived from work done by Wood, D.J., et al. (1982) and is presented in Table 1. All coding was done by the author.

The measures used in this study are as follows (from Wood, D.J., et al., 1982):

Child Initiative. How often the deaf child's responses contain contributions, elaborations, or questions not directly elicited or required by his or her family member's preceding move.

a. Overall: All contributions and questions from the child as a proportion of all his or her response moves.

b. After two-choice questions: The proportion of ansers to two-choice questions or tags which are followed by an elaborating contribution or question from the child.

c. After wh-how questions: The proportion of answers to wh-how-type questions which are followed by an elaborating contribution or question from the child.

d. After contributions: The proportion of responses to contributions which include contributions or questions from the child.

e. After phatics: The proportion of responses to phatics which include contributions or questions from the child.

f. After all questions: The proportion of all appropriate answers to all questions and tags which are followed by elaborating contributions from the child.

Child Loquacity. The mean number of words per turn in a child's response to another's moves (MLT or mean length of turn).

Child Spontaneity. Turns initiated by the child which are not directly elicited by another family member expressed as a proportion of all turns.

The extent of control parent and sibling conversations exert over the deaf child's response is reflected in a summary index.

Tenor. The control structure of the conversation. The percentage of controlling moves in all mother, father, and sibling moves directed to the deaf child. All requests for repetitions, enforced repetitions, two-choice and wh-how type questions (categories 5.1, 1, 1, 3) expressed as a proportion of all moves.


Reliability checks were carried out before coding of the transcripts began. The agreement ration (agreement/[agreement + disagreement]) computed on 1 minute (average of 23 turns) from each of three family videotapes for the language used and for "speaker" and "listener" ranged from 87% to 100% for language and was 100% for who was conversing to whom.

The reliability of coding was frequently checked against the videotapes by a research assistant who was familiar with the coding scheme. Disagreements with the author's original coding were discussed, and final decisions about changing the codes were then made.


Contorl of Conversation

Determining which family conversation styles invite deaf children's participation begins with the analysis of the control found in family conversation--that is, how and to what extent family members assume the responsibility for controlling the course of conversation. For example, is what is said generally predictive of what will follow? Consider such statements to the deaf child: "Say hello to your sister" or "did you see that film?" These are more controlling statements. The child will probably say "Hello" and the response regarding the film will likely be "yes" or "no." But consider the statement. "Next week is our family vacation." The response is fairly unpredictable, and considerably less control is exerted over the conversation. The deaf child may follow up by offering a suggestion for a good vacation spot for the family or by asking if the family dog will be going along.

The distribution of the move and response types for both "speaker" and "listener" in family conversations is given in Table 2. The table shows, for example, the percentage of their total conversation that mothers, fathers, and siblings (family) spent directing questions to the deaf children and the percentage of the time the children added on contributions to their answers (v4).

Family members tend to balance their conversation time with their deaf children between making personal contributions and stating ideas (47% of the time) and in asking them questions (39% of the time). When the deaf children answered questions, they also added a contribution onto the answers 42% of the time. However, deaf children almost neveer responded to any family moves with a question of their own. When they did, the greates proportion of their questions asked for repetitions of what was said.

Table 3 presents data on how deal children responded to different conversation moves directed to them by their parents, brothers, and sisters.

Generally, deaf children negotiated their end of family conversation. They correctly interpreted the basic meaning of their parents' and siblings' questions and other moves and made relatively few mistakes. Those questions they did miss tended to be wh- type, often suggested as posing a particular differnt linguistic challenge for deaf children. The only time the children responded with their own questions was whey they encountered less controlling conversations from their family, that is following personal contributions. The deaf children in these families followed up statements or comments made by their parents or siblings with a comment or idea of their own, which carried the conversation forward only 25% of the time they had opportunity to do so.

Three differences can be observed thus far between deaf children's conversation patterns with their families and those reported in the classroom studies (e.g., Wood, D. J., et al., 1982) and the mother-child study (Power et al., in press). First, families question their deaf children less frequently than teachers question their deaf students in school (39% compared with 58%) but about as often as mothers question their 8-9-year-old deaf children during play at home (36%). Instead of questioning, families tend to present their children with ideas and personal observations (47% of the time) which essentially invite them to take up the conversation. Teachers did similar inviting in only 27% of their conversations with their deaf students, while mothers did so in 44% of their conversation.

A second difference among family, school, and maternal conversations involves the frequency with which the deaf children elaborated on their answers to questions. The deaf children in conversation with their teachers contributed additional information to only 14% of their answers (Wood, D.J., et al., 1982), whereas children elaborated 35% of their answers to their mothers (Power et al., in press) and elaborated 42% of the time to their families.

Third, the deaf children in this study contributed their own ideas and statements far less frequently (average = 25%) in their responses to their families' less controlling moves than Wood, D. J., et al. (1982) observed in classroom interaction between teachers and deaf students (average = 72%) and than deaf children contributed in their responses to their mothers (74%) (Power et al., in press).

Relationship Between Family Conversation

Style and the Deaf Child's Participation

This section describes how differences in family conversational style effect the deaf children's level of participation.

The first question in the analysis is whether or not the deaf children's participation in conversation, measured by their initiative and loquacity, varied given different degrees of control in conversation directed to them, that is, after questions, contributions, and phatics. No significant differences were observed in the percentage of the children's responses showing initiative. The deaf children consistently initiated regardless of the type of move family members directed to them. Second, regarding loquacity, the children did no show a pattern of responding such that their highest MLTs occurred in response to families' phatics and contributions with lowest MLTs to questions. In fact, wh-how type questions (and to some extent two-choice questions) met with significantly longer utterances from the children than did contributions and phatics (Wilcoxon matched-pairs signed-ranks test, p's < .05 and .10).

Second, the analysis assesses the effect of the mother, father, and sibling's rate of questioning or the tenor of their conversation on the child's participation by looking at the relationships between tenor and the child's initiative, loquacity, and spontaneity.

The relationship of tenor and overall child initiative was not found to be significant. However, correlations between tenor and child initiative after wh-questions (r = -.48, p < .10) and after phatics (r = > .44, p < .101) provide evidence to suggest that the deaf child is less likely to go beyond what is required in response to these moves under the condition that his or her family's general conversation tone is one of questioning. Similarly, the data show a tendency for deaf children to be spontaneous conversationalists with their families if family members ask them fewer questions (r = - .43). In contrast, a fairly high and positive relationship (r = .59, p < .05) was found between tenor and child initiative after two-choice questions.

The tenor of the conversation plays an important role in the number of words or signs the children use to express themselves. While the correlation between tenor and the child's overall MLT is low and not significant (r = - .14), the correlation between tenor and MLT after phatics (r = - .58, p < .05) indicates that children are more loquacious to phatic moves in low-control families. The data also show a tendency for deaf children to be more loquacious after wh- questions in low control families (r = - .43). The correlations regarding MLT generally support the concept of control in conversation; that is, more questions meet with less participation from children. However, as was found with child initiative, a high, positive, and significant correlation was found between tenor and MLT in response to two-choice questions (r = .56, p < .05). This suggests that, with two-choice questions, children say more, the more often they encounter questions.


The amount of control family members exerted over the course of conversation with the deaf child, as hypothesized, is more in line with that of mothers with their deaf children and quite low compared with teachers interacting with their deaf students.

The conversation environment established by the families was fairly balanced in terms of levels of control and invitation. That is, family members integrated about as many questions into their conversations to the deaf child as they did ideas and phatics. Thus, these families set up relatively inviting conversational environments in which control did not function as a depressant on the child's willingness to participate. This is supported by the finding that parents' and siblings' more controlling questions met with relatively greater initiative from deaf children than did ideas or comments. In fact, deaf children in this study were most loquacious after wh- questions.

However, this study provides evidence that family conversation style with the deaf child affects the child's participation. As predicted from the concept of "tenor" taken from earlier studies, the children in family conversation did respond differentially to overall family conversation style. They were more spontaneous and elaborating to wh- questions and after statements when there were fewer questions overall to be answered. This line of analysis also showed that children showed more loquacity and initiative to two-choice questions when there were more questions put to them.

This last finding is in contrast with information reported in both the maternal and classroom studies. One possible explanation for this finding is that two-choice questions are generally easier to answer and that the child is choosing them to gain entry to conversation participation. The two-choice question contains more specific content that helps the respondent and there is less opportunity to be wrong. Further, families may be taking extra care in the way they asked these questions. An examination of the two-choice questions showed that very often the question is set up to focus the question for the child -- to make it more easily understood. The most frequent set-up was the use of repetition, as in: "Did you like it? Did you like the street?' and "Are you going to tell Bill about this? Are you going to tell him about it?" Also, an elaborated answer to a two-choice question showed up when parents tagged it onto a wh-question; they used the wh-question part of the turn to set up the two-choice question, as in: "What was the favorite thing you did on your trip? Did you ride the roller coaster?' and "So what time is that, about 1:30?" The two-choice question then may be more accessible to the child. Even under a flood of questioning in family conversation, the child, rather than becoming inhibited, takes advantage of the two-choice question as an entry point to participation.

Deaf children in this study were generally responsive in conversation with their hearing parents and siblings, but were least likely to carry conversations forward to develop an idea or topic suggested by another family member. Thus, although deaf children are gaining access to family conversation either on their own initiative or because their families facilitate it, they are not successfully taking up opportunities to become engaged in the content. This is an issue of central importance to families and to professionals who work with families with deaf children. Whether this harks back to deaf children's early fairly unsuccessful attempts at engagement in dyadic interactions with their mother or whether it is what might be expected to ensue when methods of communication may not match is a question for further study.

Several practical implications of these findings for teachers and families are the following:

* Conversation is a functional mechanism for facilitating the deaf child's access to an important aspect of family life. Families have their child's attention during dinnertime conversation. Children are at the ready for participation. Parents should make the most of that initiative by being responsive receptors, by paying attention to the flow of conversation, and by expressing interest and having fun.

* Mine the possibilities of questions but don't let go of a good idea. Questions seem to invite the greatest participation from the child--perhaps because their structure makes meaning clearer and, therefore, easier to respond to. Families should be encouraged to be their own researchers and observe how their children respond to comments/ideas they make versus questions they ask. Perhaps family members should rephrase or simplify the language of these comments to be sure the deaf child is understanding the content. The goal is to engage the child into more conversation on the same topic.

* "Mix it up" so that questions and ideas are integrated more evenly in conversation. The best strategy for encouraging participation in family conversation is to avoid a high rate of questioning relative to other conversation moves. Questions strung together seem to have a cumulative effect that diminishes the child's participation.

This study considers a relatively small number of selected families with deaf children. How far the conversation behaviors the families demonstrated may generalize to all families with deaf children requires further study, as does the question of whether these findings will illuminate any differences between "oral/aural" and "simultaneous communication" families.

In the beginning of this article, I indicated that participation in family conversation signals access to an important function of family life. Bossard and Boll wrote in their 1948 text, The Sociology of Child Development, that the dinner table was the focal point for most family interaction and was the time the family was most apt to reveal its true self. Dinner thus becomes one of the few times remaining in our busy lives for developing family relationships. The deaf child's participation in this opportunity is demonstrated by the findings in this study.


Bodner-Johnson, B. A. (1986). The family environment and achievement of deaf students: A discriminate analysis. Exceptional Children, 52, 443-449.

Bossard, J., & Boll, E. (1948). The sociology of child development. New York: Harper & Row.

Lewis, M. & Feiring, C. (1982). Some American families at dinner. In L. M. Laosa & I. E. Sigel (Eds.), Families as learning environments for children (pp. 115-145). New York: Plenum Press.

Meadow, K. P., Greenberg, M. T., Erting, C., & Carmichael, H. (1981). Interactions of deaf mothers and deaf preschool children: Comparisons with three other groups of deaf and hearing dyads. American Annals of the Deaf, 126, 454-468.

Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard University Press.

Power, D. J., Wood, D. J., Wood, H. A., & MacDougall, J. (1990). Maternal control over conversations with hearing and deaf infants and young children. First Language, 10, 19-35.

Roscoe, J. T. (1969). Fundamental research statistics. New York: Holt, Rinehart and Winston, Inc.

Schlesinger, H., & Meadow, K. (1972). Sound and sign: Childhood deafness and mental health. Berkeley: University of California Press.

Tizard, B., & Hughes, M. (1984). Young children learning. Cambridge, MA: Harvard University Press.

Turnbull, A. P., & Turnbull, H. (1986). Families, professionals, and exceptionality: Columbus, OH: Merrill.

Wedell-Monnig, J., & Lumley, J. (1980). Child deafness and mother-child interaction. Child development, 51, 766-774.

Wood, D. J., McMahon, L., & Cranstoun, Y. (1980). Working with under-fives. London: Grant McIntyre.

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 and Psychiatry, 23(3), 295-308.

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 and Psychiatry, 25(1), 45-62.

BARBARA BODNER-JOHNSON is Professor and Chair in the Department of Education, Gallaudet University, Washington, D.C.

The work reported in this article was carried out in part with funds from the Gallaudet University Small Grants Program in Graduate Studies.

Manuscript received June 1989; revision accepted October 1989.
COPYRIGHT 1991 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bodner-Johnson, Barbara
Publication:Exceptional Children
Date:May 1, 1991
Previous Article:Paradigmatic distinctions between instructionally relevant measurement models.
Next Article:Direct instruction in math word problems: students with learning disabilities.

Related Articles
School for the Deaf welcomes Aboriginal students.
DB-link: National Information Clearinghouse on Children Who Are Deaf-Blind. (Better Speech and Hearing Month)(Organizational Spotlight).
Deaf children: the right to be heard.
Early intervention collaboration-deaf role models.
Beyond hearing: cochlear implants work best when given early.
Groundbreaking exhibition charts "History Through Deaf Eyes".

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters