2003 IFA Congress: Montreal, Canada

Negotiation Patterns of Families with Stuttering and Non-Stuttering Children

Christa L. Shuman, M. A., and A. Richard Mallard, Ph. D./CCC-SLP

Southwest Texas State University, San Marcos, Texas


The purpose of this study was to determine if families with stuttering children and families with non-stuttering children differed for the variables of number of turns, number of turns with questions, number of interruptions, or total negotiation time "during a negotiation task. The experimental (S) group included ten families with a child who stuttered and the control (NS) group included ten families with children who did not stutter matched to children in the experimental group. Significant differences were not found for turns, interruptions, turns with questions, or total negotiation time. A significant interaction difference existed between the groups for number of interruptions.

  1. Introduction
Since 1986, Southwest Texas State University (SWT) has conducted a family program for the treatment of stuttering. Modeled after Rustin’s program in England (1987), the SWT program (Mallard, 1998) works with the entire family and not just the individual who stutters. Whereas traditional therapy programs emphasize speech modification only, the SWT program includes social skills training, negotiation, and problem solving.

Negotiation is an important aspect of the SWT program. A family’s negotiation style impacts how they arrive at solutions that are acceptable to all the family members. The program emphasizes negotiation as a way for a family to decide how they can best solve their problem of stuttering. A routine procedure conducted with all families during the initial interview is to observe how they communicate during a negotiation task. An example of a negotiation task is: “Your family has decided to get a new pet. How would you decide what kind of pet to get?” The family is allowed to discuss the issue and arrive at a solution. This interaction is videotaped. Information about how the family interacts (i.e., number of interruptions, number of questions, and turn taking) is gathered by the clinicians for future use in therapy.

Over the years, it has been observed that turn taking, questions, interruptions, and parental domination of talk time in family negotiations are frequently addressed during the course of treatment. For example, in a negotiation of one family in the program, the mother took 55% of the turns, and 67% of her turns included questions. Three of the father’s five turns were interruptions, two of which were commands about behavior directed to the child who stuttered. No other family member interrupted during the negotiation. It has been noted also in the program that some family members took disproportionately more turns, asked more questions, and interrupted more often than other family members. Negotiation patterns differed, with some families spending several minutes negotiating while other families talked for little more than a minute. It would be important to the assessment and management of stuttering in a farnily-oriented, problem-solving program to determine the roles of these variables and how their occurrences differ between families with children who stuttered and families with children who did not stutter.

No previous research was identified for communication behaviors within the context of family negotiations. Such studies would be important in a farnily-oriented, problem-solving therapy program, leading to an increased understanding of how families with children who stutter interact in negotiation. The purpose of this study was to identify the differences that may exist between families with children who stutter and families with children who do not stutter relative to number of turns, 266 Theory, research and therapy in fluency disorders number of turns with questions, number of interruptions, and total talk time in a negotiation setting. If differences are found to exist between the groups, improved procedures for assessment and management can be developed to help families with children who stutter modify their interactions to be similar to those of families with children who do not stutter.

  1. Method

Ten families were in the S group, each with a child who stuttered. These families were referred to the stuttering program at the SWT clinic. The families responded to the same prompt and had a recording suitable for transcription. The ages of the children who stuttered ranged from 5 years, 7 months to 12 years, 1 month of age, with an average age of 9 years, 2 months. There were four females and six males. The ethnicities of the families included nine Caucasian and one Hispanic.

The NS group consisted of ten families who had no children who stuttered. One of the children in each family was matched to a stuttering child from the S group according to age (within six months), ethnicity, and gender. These families were found primarily through personal contacts throughout the Central Texas area. The ages of the children matched to the children who stuttered ranged from 5 years, 5 months to 12 years, 5 months of age, with an average age of 9 years, 4 months. Data for the NS group were collected within a span of two months to ten years later than data for the S group.


The case history holds great importance for family-oriented, problem-solving based stuttering programs. All families referred for therapy complete a verbal case history. This case history is obtained by the clinician who will conduct the program. As part of the initial session, a standard procedure is to videotape a negotiation session.

To participate in this study, the negotiations had to occur in a clinical setting during the initial meeting with the family. This meant that all the families in the NS group had to agree to meet with the researcher at the Southwest Texas State University Speech-Language-Hearing Center. All of the families were given the same prompt for discussion and the interactions for the families in the S group occurred prior to the initiation of the therapeutic program. The entire family was required to participate in the interaction and the children who did not stutter had to be free of speech, language, or hearing problems, as determined by a screening immediately prior to the interaction.


S Group. The families with children who stuttered were referred to the SWT clinic due to a suspected fluency disorder. Each of the children who stuttered received a standard speech, language, and hearing screening presented by faculty and/or graduate students supervised by certified speech- language pathologists. All the target children in the S group did not exhibit additional speech, language, or hearing problems.

NS Group. The lead researcher screened the target children in the families with children who did not stutter to rule out speech, language, and hearing problems. Each child had his/her hearing screened at an intensity of 20 dB HL for the frequencies of 1000, 2000, and 4000 Hz. The clinician initiated a brief, informal conversation with the child to screen speech and language. The child passed the speech and language screening if responses were judged to be age-appropriate in the areas of pragmatics, receptive and expressive language, and articulation. As in the S group, the parents were asked about their child’s speech, language, and hearing functions in addition to the screening procedures.

Recording Procedures. As each family entered the clinical setting, they were asked to sit together to accommodate videotaping for adequate video and audio recordings of each participant. Typically, each family sat on a sofa, but they were not instructed where to sit relative to one another. The testing procedures did not differ for the S group and the NS group. The clinician provided the families with the following instructions before the initiation of the negotiation session:
When I start the videotape, I will introduce your family by last name and today’s date. Then, I will give you a topic to discuss. Whenever you have finished discussing, I will stop the videotape.

A camcorder [JVC Compact Super VHS camcorder (GR-SX960U)] with a zoom lens and built-in microphone was positioned two meters from the family at a height of 1.5 meters. The clinician stood behind or to the side of the camcorder during the interactions. The prompt used for the family negotiation was:
You have been given two thousand dollars to take a vacation. You may spend less than the two thousand dollars but you may not spend more than the two thousand dollars. How would you decide what to do?

When the family finished discussing the topic, the camcorder was stopped. This was determined usually by a verbal response (i.e., “Okay, we’re done.”) from the family. At the end of the session, the family was given the opportunity to watch their negotiation interaction and ask questions.


All interactions were transferred to master videotapes. The interactions for the S group were on one videotape and the interactions for the NS group were on another videotape. Each interaction was transcribed orthographically and then analyzed for the components to answer the research questions. Conversational turns, interruptions, questions, and total talk time were computed for each negotiation.


The first rater for this study was the lead researcher, a second-year graduate student with graduate-level coursework in fluency and research in communication disorders and a year of clinical practice. The second rater was a certified speech-language pathologist with over thirty years of experience in stuttering therapy who held the Certificate of Specialty Recognition in Fluency Disorders from the American Speech-Language-Hearing Association (ASHA). Interrater and intrarater reliability measures were determined for turns, interruptions, and turns with questions. A complete negotiation was randomly selected from the NS group for reassessment. The video of negotiation was reviewed with the aid of an orthographic transcript of the interaction provided to each rater. Each rater counted the number of turns during the interaction and marked turns with questions and interruptions on the transcript. These numbers were compared to the initial analysis and percentages were derived. Interrater reliabilities were 96% for turns, 95% for turns with questions, and 88% for interruptions. Intrarater reliabilities were 98% for turns, 95% for turns with questions, and 91% for interruptions.

  1. Results
The independent variable was the videotaped negotiation interaction conducted with each family. This study investigated four dependent variables: turns, interruptions, turns with questions, and total talk time. Analyses for turns, interruptions, and turns with questions were done with the percentage data, while analysis for total talk time was done with the raw data.

Analyses were conducted using SPSS 9.0 for Windows (SPSS, 1998) and consisted of univariate ANOVAS. Preliminary testing of the data consisted of testing for correlations among the variables. The highest level of significance was .56 between turns and questions, indicating that the variables were all measuring different behaviors. For the statistical analyses, siblings 1 and siblings 2 were collapsed into a single category because only four families had a second sibling.

No significant differences were found between the S group and the NS group for turns, interruptions, turns with questions, or total talk time during family negotiations.

Although there was not a significant difference between the S and NS groups for the variable of interruptions, the groups-persons interaction was significant. For the S group, the mothers and fathers both produced high percentages of interruptions with mothers producing slightly fewer interruptions than fathers. Children and siblings, however, produced considerably fewer interruptions than either mothers or fathers. In the NS group, like the S group, mothers produced fewer interruptions than fathers, but both were lower than the mothers and fathers in the S group. The target children in the NS group had a higher percentage of interruptions than the mothers and fathers. They also had a higher percentage of interruptions than the target children in the S group. The siblings in the NS group produced fewer interruptions than the target children, falling between the means for the mothers and fathers.

  1. Discussion

Unlike previous studies that observed the interactions of one parent interacting with one pre- school child (Meyers, 1990; Kelly & Conture, 1992; Cherry & Lewis, 1976), this study observed the negotiation interactions of school-aged children with their entire families. The task of negotiation as related to stuttering therapy has not been studied previously. Interactions in past research have included free-play sessions (Cherry & Lewis, 1976; Golinkoff & Ames, 1979; Kelly, 1994; Kelly & Conture, 1992; Langlois et al., 1986; Meyers, 1990; Meyers & Freeman, 1985a, 1985b, 1985c; Rondal, 1980;) and spontaneous conversation between parents and children (Kasprisin-Burelli et al., 1972; Rondal, 1980). This differs from negotiations in that a negotiation specifies a topic and has an end result. The topic channels the course of conversation and influences the way in which families may approach the task. No time limit is predetermined and the task is complete when the family reaches an agreement. This is not to say that all families approach the task in the same manner. For example, the task for the current study was deciding how to spend $2000 for a family vacation. The way a family chose where they would go for a vacation may have been partially based. on the different life experiences and circumstances of each family. Some families chose to use the money to visit relatives or go to a nearby city whereas other families chose more elaborate destinations, which involved details such as travel arrangements and accommodations. One of the most interesting findings of this study was the variability that existed between families regardless of the presence or absence of children who stuttered, such as the variability found for total talk time.

Total Talk (Negotiation) Time

A wide range existed within the S and NS groups for total talk time during the negotiation. For example, one family in the S group negotiated for 542 seconds and two families negotiated for only 55 seconds. Within the NS group, one family negotiated for 735 seconds and one family negotiated for only 57 seconds. Significant differences were not found between the groups, possibly due to large standard deviations, a small sample size, and only one observation per family. In most of the previous studies of parent-child interactions as they relate to stuttering, the length of time spent in an interaction was previously determined by the researchers'(i.e., Kasprisin-Burelli et al., 1972; Meyers & Freeman, 1985a, 1985b, 1985c; Ratner, 1992).


The results of this study agreed with the findings of Kelly and Conture (1992) who found no significant difference for turns between mothers of children who stuttered and mothers of children who did not stutter. These studies differed, however, with respect to number of subjects, age, and task. Kelly and Conture studied 26 mother-child dyads with children ranging from 3 years, 2 months to 4 years, 10 months of age as they engaged in free-play. The ten families in each group of the current study ranged in age from 5 years, 5 months to 12 years, 5 months and engaged in negotiation.


Although significant differences in interruptions were not found between the S and NS groups, a significant difference did exist between the two groups for the pattern of interruptions. Mothers and fathers in the S group had higher percentages of interruptions than the target children and siblings. The target children in the NS group had more interruptions than any other family member. In both groups, fathers had a higher percentage of interruptions than mothers. N 0 post-hoc testing was done to determine the level of significance between fathers’ and mothers’ interruptions, but the difference was not likely to be significant due to the small variance between the two means.

A significant difference was not likely for siblings between the two groups due to the small difference between the means, but significant differences may exist between mothers, fathers, and children for the S and NS groups. Such differences would support the findings of Kasprisin- Burelli et al. (1972) and Mordecai (1979). The studies found that parents of children who stuttered interrupted more often than parents of children who did not stutter. Kasprisin-Burelli et al. investigated conversations between school-aged children and one of their parents. Mordecai (1979) investigated the communication behaviors of twenty mothers and fathers during triadic interactions with their preschool children. Ten of the children stuttered and ten of the children did not stutter. Kelly (1994) investigated the verbal behaviors of father-child dyads and reported that fathers produced many interruptions, regardless of whether the child stuttered or not. The current study did not support Kelly’s ï¬ ndings. Fathers in the S group produced more interruptions than any other family member. In the NS group, however, the target children produced more interruptions than any other family member.

Unlike the findings of Meyers and Freeman (l985a) or Kelly and Conture (1992), this study suggested that significant differences do exist between families with children who stutter and families with children who do not stutter. Meyers and Freeman studied 12 boys who stuttered and 12 boys who did not stutter (ages 4 years to 5 years, 11 months) as they interacted with their mother, an unfamiliar mother of a child who stuttered, and an unfamiliar mother of a child who did not stutter. Kelly and Conture studied 13 boys who stuttered and 13 boys who did not stutter (ages 3 years, 3 months to 4 years, 8 months) interacting with their own mothers.

Turns with Questions

The results of this study did not agree with Langlois et al. (1986) who found that mothers of children who stuttered asked more questions than mothers of children who did not stutter. Langlois et al. studied eight children who stuttered and eight children who did not stutter interacting with their mothers within their own homes. The children ranged in age from 5 to 9 years.

Meyers (1990) observed children who stuttered (ages 2 years, 3 months to 6 years, 4 months) interacting separately with their mothers, fathers, and peers. The results of this study suggested that parents asked significantly more questions than peers. The current study is similar to the Meyers study in that it occurred in a clinical setting and it included siblings, which provides a similar interaction to the children as that of peers. The differences, however, outweigh the similarities of these two studies. The current study included older children who stuttered and children who did not stutter and observed the interactions of these children with their entire family (instead of each conversational partner individually).

Implications for Clinical Assessment and Management

This study was based on the methods of a specific clinical program and its purpose was to investigate family negotiation patterns. The results are applicable to the Rustin model (1987) for stuttering assessment and intervention like the one at SWT. Negotiation and problem Solving are integral to the program. The results of this study, however, revealed no significant differences between families with children who stuttered and families with children who did not stutter. Given the problem-solving approach inherent to the program, however, it is important to continue making these pre-therapy observations due to the wide variances that can be expected between families. 270 Theory, research and therapy in fluency disorders

The clinical usefulness of the negotiation session has proven valuable because parents can view and analyze their pre-treatment video. Parents are often able to determine strengths and strategies to facilitate more fluent behavior without prompting from a clinician. It should be noted that all families judged their negotiations in this study to be fairly typical to what would occur in their own homes. It is hoped that this study will lead to further research within the area of family negotiation patterns as they relate to stuttering and stuttering therapy.

Cherry, L., & Lewis, M. (1976). Mothers and two-year-olds: A study of sex-differentiated aspects of verbal interaction. Developmental Psychology, 12 (4), 278-282.
Duncan, S. (1972). Some signals and rules for taking speaking turns in conversations. Journal of Personality and Social Psychology, 283-292.
Golinkoff, R. M., & Ames, G. J. (1979). A comparison of fathers’ and mothers’ speech with their young children. Child Development, 50, 28-32.
Guitar, B., Schaefer, H. K., Donahue-Kilburg, G., & Bond, L. (1992). Parent verbal interactions and speech rate: A case study in stuttering. Journal of Speech and Hearing Research, 35, 742- 754.
Kasprisin-Burelli, A., Egolf D. B., & Shames, G. H. (1972). A comparison of parental verbal behavior with stuttering and nonstuttering children. Journal of Communication Disorders, 5, 335-346.
Kelly, E. M. (1994). Speech rates and turn-taking behaviors of children who stutter and their fathers. Journal of Speech and Hearing Research, 37, 1284-1294.
Kelly, E. M., & Conture, E. G. (1992). Speaking rates, response time latencies, and interrupting behaviors of young stutterers, nonstutterers, and their mothers. Journal of Speech and Hearing Research, 35, 1256-1267.
Langlois, A., Hanrahan, L. L., & Inouye, L. L. (1986). A comparison of interactions between stuttering children, nonstuttering children, and their mothers. Journal of Fluency Disorders, 11, 263-270. '
Mallard, A. R. (l998a). Encouraging a broader perspective in judging the effectiveness of stuttering therapy. Journal of Fluency Disorders, 23, 123-125.
Meyers, S. C. (1990). Verbal behaviors of preschool stutterers and conversational partners: Observing reciprocal relationships. Journal of Speech and Hearing Disorders, 55, 706-712.
Meyers, S. C., & Freeman, F. J. (198521). Interruptions as a variable in stuttering and disfluency. Journal of Speech and Hearing Research, 28, 428-435.
Meyers, S. C., & Freeman, F. J. (l985b). Are mothers of stutterers different: An investigation of social-communicative interaction. Journal of Fluency Disorders, 10, 193-209.
Meyers, S. C., & Freeman, F. J. (1985c). Mother and child speech rate as a variable in stuttering and disfluency. Journal of Speech. and Hearing Research, 28, 436-444.
Mordecai, D. R. (1979). An investigation of the communicative styles of mothers and fathers of stuttering versus nonstuttering preschool children. Dissertation Abstracts International, 40, 4759-B.
Ratner, N. B. (1992). Measurable outcomes of instructions to modify normal parent-child verbal interactions: implications for indirect stuttering therapy. Journal of Speech and Hearing Research, 35, 14-20.
Rondal, J. A. (1980). Fathers’ and mothers’ speech in early language development. Child Language, 7, 353-369.
Rustin, L. (1987). The treatment of childhood dysfluency through active parental involvement. In L. Rustin, H. Purser, & D. Rowley (Eds.), Progress in the treatment of fluency disorders (pp. 166-180). London: Taylor and Francis Ltd. SPSS Inc. (1998).
SPSS Base 9.0 for Windows Users Guide. SPSS Inc., Chicago IL.

join button


The IFA implemented Japanese translations of some pages on the site for the 2018 Joint World Congress. Choosing Japanese below to see these translations.

Not all pages are translated, but you can use Google translate to see a machine translation using the switch below

Google Translate


Journal of Fluency DisordersBrowse the current issue

The official journal of the International Fluency Association
IFA Members receive online access to JFD as a member benefit.

Read more: JFD

Twitter Feed @official_IFA

IFA on Facebook