Conti-Ramsden 2 Corpus

Gina Conti-Ramsden
Human Communication and Deafness
The University of Manchester

Participants: 6
Type of Study: clinical
Location: UK
Media type: no longer available
DOI: doi:10.21415/T55S3P

Browsable transcripts

Download transcripts

Citation information

Publications using these data should cite:

In accordance with TalkBank rules, any use of data from this corpus must be accompanied by at least one of the above references.

Additional references include:

Project Description

This corpus contains transcripts from three children with specific language impairment (SLI) and their three younger normal siblings.

The Families of the Children with SLI

The families involved in this longitudinal study were part of a larger project investigating the language development of children with specific language impairment (SLI), and of their younger nonimpaired siblings (Conti-Ramsden & Dykins, 1991; Conti-Ramsden, Hutcheson, & Grove, 1995). Families were informed of the research project through the speech and language therapy services in the northwest of England, and asked if they would be willing for the research workers to visit them and discuss their possible involvement in more detail. During an initial visit, the research project was explained and parents were given the opportunity to opt for a longer longitudinal involvement of approximately 2 years. In addition, the researchers collected language samples, by means of an audio recording, from the child with SLI and, at a separate session, from the younger sibling. The first 50 child utterances were transcribed from the recordings in order to ascertain the mean length of utterance of the children, using Brown’s (1973) criteria, with the modifications suggested by Miller (1981). From the outset, it was made clear to the parents that no identifying information would be revealed except to the research workers, and that the family could terminate their longitudinal involvement in the research project at any time. Accordingly, any data collected from the family at that point would be destroyed if desired. Three families agreed to participate in the longitudinal phase of the project. We examine the data obtained from these three children with SLI and their younger normal language learning siblings. The children with SLI were named Colin, Andrew, and Mark and the younger siblings were named Chris, Nina and Adam.

Characteristics of the Children

The characteristics of the children with SLI and their younger siblings at the beginning of the study are presented in the table that follows in terms of age and psychometric results. It can be seen that the participants with SLI were three expressively impaired children, all male, with severe problems (as can be seen from the discrepancy between their age and their MLU obtained on the language sample). The three children performed within one standard deviation of the mean in the Leiter International Performance Scale, which provided a measure of IQ. In addition, they were tested in a number of comprehension measures. These children had varying comprehension profiles with below-average vocabulary comprehension (as measured by the British Picture Vocabulary Scale (BPVS); 18 to 26 percentile rank), poor comprehension of grammar (as measured by the TROG; 20 to 40 percentile rank), but better overall auditory comprehension abilities (as measured by the Preschool Language Scale; results in Table 1). The younger siblings were two males and one female who ranged in age from 1;11 to 2;2 at the beginning of the study. The three siblings performed within one standard deviation of the mean in the measure of IQ. They also had expressive language and auditory comprehension skills well within normal limits. The siblings were too young to be tested for comprehension of grammar (TROG) or vocabulary comprehension (BPVS).

In addition, all six children had adequate hearing sensitivity as determined by pure-tone audiometry screening bilaterally at 500, 1000, and 2000 Hz at 25 dB (equivalent to pure tone thresholds of 25 dB HL, re: ANSI, 1989). The three children with SLI had eventful birth histories with the three children being anoxic at birth. Developmental histories ascertained by a questionnaire to parents revealed all developmental language milestones to be delayed in the three children with SLI. In addition, motor milestones appeared delayed for Colin and Andrew.

All six children spoke English in monolingual homes and came from intact (two parent) families. In all three families, the mothers remained at home as housewives while the fathers went out to work; all the parents had secondary education. All children with SLI were receiving speech therapy in a clinic or were enrolled in language-based classrooms for children with SLI (called “language units” in England).

Video Recordings

The video recording sessions lasted approximately 15 to 20 minutes and were conducted in the homes of the families using the play materials available there. In order to keep the parents as unconcerned as possible about the nature of their own speech, they were told that the research was primarily about the children’s communicative development. The instructions given to the parents were “play as you normally do.” The three families participated in a number of dyadic interactions including mother, father, and sibling. The present study mainly concerns itself with the mother–child play interactions although some father–child interactions were occasionally also used.

All the children were videotaped every 6 weeks, but illness and cancellations meant that video samples were, on average, once every 3 months over a 15 month period. A further sample was taken after approximately 10 to 16 months, completing a 2-year observation period. In the present study, we examined seven sessions over the two-year period for each of the three families. As the aim of the study was to examine the development of expressive language (in particular the early stages of verb use), MLU in words was thought to be a better indication of expressive language than MLU in morphemes. This was also a more appropriate measure for comparisons with the nonimpaired younger siblings as we were interested in what the children were doing at the point at which they were just starting to use multiword speech.


The first 10 minutes of each of the seven mother–child sessions were transcribed. The transcriptions contained information about verbal and nonverbal interactions, and the context in which these events occurred. This was carried out in accordance with the CHILDES guidelines for CHAT. The computerized transcripts were then compared with the original videotaped data by an independent transcriber in order to verify their accuracy. This process resulted in 97.0% inter-transcriber reliability. Any disagreements concerning the transcription were resolved by re-examination until consensus was reached. The data from the present study are available in the CHILDES database.

The number of child utterances was noted for each MLU point for each child. We were able to include in the analysis 100 child utterances for each of the seven sessions for each child. Transcripts from the father–child interaction sessions (carried out on the same day) were used in some cases to supplement those mother–child sessions containing too few child utterances.

Children at Beginning of Study
Colin (SLI)5;81.45895;491
Chris (Sibling)2;21.28912;1131
Andrew (SLI)5;31.771155;6105
Nina (Sibling)1;111.451012;7131
Mark (SLI)3;91.281053;10103
Adam (Sibling)1;111.51962;0104
PLS-AC = Preschool Language Scale Auditory Comprehension
ACQ = Auditory Comprehension Quotient