Rondal Corpus: Normal

Jean-Adolphe Rondal
Laboratoire de Psychologie
University of Liège

Participants: 21
Type of Study: naturalistic
Location: USA
Media type: no longer available
DOI: doi:10.21415/T5RC93

Browsable transcripts

Download transcripts

Citation information

Publications using these data should cite:

Additional relevant publications are:

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

Project Description

This corpus contains files from 21 English-speaking Down syndrome children, along with a set of files from 21 children in the normal control group. The data from the control group are in the directory of English normal files and the data from the Down syndrome children are in the directory with the clinical subjects. The data were collected from chil-dren in Minnesota by Jean Rondal. A fairly full report on the project is given in Rondal (1978). The samples are matched for mean length of utterance. The original study was designed to examine differences in maternal speech directed to normal and Down syndrome children.


The participants of this study were 21 Down syndrome children and their natural mothers and 21 normal children and their natural mothers. As a condition for participating in the study it was required that none of the mothers in the two groups had been or were currently engaged in any early education curriculum for parents with special emphasis on promoting early language abilities in children. All of the normal children and their mothers and 14 of the identified Down syndrome children lived in the Minneapolis-St. Paul area. The seven remaining Down syndrome children and their mothers lived in other towns in Minnesota. Karyotypes were obtained for all the Down syndrome children and all were reported to be Trisomy 21s. There were 12 girls and 9 boys among the Down syndrome children, and 8 girls and 13 boys among the normal children. No effort was made to balance the two groups of children for gender, as it was thought not to be an important variable.

In order to participate in the study, the children could not have any debilitating heart condition, obvious sensory impairment, or more generally any medical condition (other than Down syndrome for the Down syndrome children) that might seriously limit their development, and their speech had to be reasonably intelligible. The normal and Down syndrome children were matched on linguistic development as measured by MLU. The children’s MLU was computed using the criterion given in Brown (1973). The only exception to Brown’s criterion was that MLU was based on the total sample (i.e., one–hour speech recording) rather than the first 100 utterances. On the basis of the children’s MLU, the mother–child pairs were divided into three language-level categories for each population of normal and Down syndrome children. Specified MLU ranges for the three language levels were 1.00 to 1.50, 1.75 to 2.25, and 2.50 to 3.00. The following table lists child and mother MLU along with the child’s age in months at the three language levels.

LevelNo.GroupChildSexAgeMother MLUChild MLU

There are two CHAT files from each child with the exception of Martin and Stella. The files for these two children were lost in 1976 and cannot be recovered. All of the names given are pseudonyms. Participants are identified by first names and with a three-digit code. In this code the first number is for the group level with “1” indicating Down syndrome and “2” indicating normal; the second number is for the language level, and the third number is for the participant number within the particular cell.


The mothers of normal children and the mothers of Down syndrome children were matched on the following criteria: ethnic group (white), familial monolingualism (English), familial structure (both husband and wife living at home), mother free of any major sensory handicap, maternal intelligence not obviously outside of the normal range (no intelligence test given), and socioeconomic status (the families selected for the study were predominantly drawn from the middle class). Perhaps more important than socioeconomic status (usually based on occupational and educational level of the head of the household) for research of this type, is the mother’s educational level. The mothers selected for this study were matched on the Educational Scale supplied by Hollingshead in his two factor Index of Social Position. The overall means of the mothers of Down syndrome children on the Hollingshead’s Educational Scale was 2.67 (SD = 1.02) versus 2.71 (SD = 0.90) for the mothers of normal children. This difference was found to be not significant.

No effort was made to match mothers of normal and retarded children for age, nor to match normal and retarded children for birth order, number of siblings, and age differences between the children in the family as it is known that in the cases of Trisomy 21, the mean age of the mother at the birth of the child is significantly older than in control populations. This, in turn affects birth order and family composition for Down syndrome children as they are more likely to be later-born children than are normal children. The average age of the mothers of Down syndrome children in this study was 514.86 months (SD 100.84 months) versus 338.29 months (SD 49.42 months for the mothers of normal children. The average birth order was 3.76 (SD 2.30) for the Down syndrome children and 1.76 (SD 0.89) for the normal children. From language level 1 to language level 3 respectively, approximately 12% to 40% of the siblings of the Down syndrome children were no longer living in the family home at the time of the study.

Data Collection

The verbal interaction between mother and child was tape recorded at home in a free-play situation. The investigator was present in the home during the tape recording and made every effort to keep his presence as discreet as possible. The mothers were told that the study was primarily about child language development in a plausible attempt to keep them as unconcerned as possible about their own speech. Moreover, the mothers were asked not to engage the investigator in conversation during actual recording.

It is possible that mother–child interactions in the presence of an observer are somewhat different from what they are “behind closed doors.” Even if mothers modified their behavior toward the children in the observer’s presence, it is improbable that they would be able to invent, at once, new and different mother–child interaction patterns (Moerk, 1972). Besides, there is no reason to expect the observer’s presence to affect differentially the verbal behavior of mothers of normal and Down syndrome children.

In order to preserve as much spontaneity and naturalness in the mother–child interac-tions as possible, no specific instruction other than “do what you usually do when you play and talk with the child and use whatever kind of toys or material you want to use, only avoid recitations” was given to the mothers as to what they should do with the children during the free-play situation. It turned out the free-play situations and the material used by the mothers were surprisingly similar from home to home, particularly for those normal and Down syndrome children at language levels 1 and 2. The sessions alternated the use of Play-doh games; shape-matching or shape-folding games; play-action games such as the farm game, the airport, the village, the school, Sesame Street, McDonald’s, or PlaySkool; and looking at picture and storybooks. The contents of the free-play situations were somewhat more heterogeneous for the two groups of children at the third language level, with several mothers of normal and Down syndrome children spending part or all of the two recording sessions in conversation with the child using toys and pictures as a support for conversation. There were two recording sessions each lasting half an hour for each mother and child pair. The two recording sessions took place on two different days at approximately a 1-week interval. They were preceded, on another day, by a 20-minute “get acquainted session” during which the investigator familiarized himself with the mother and the child, obtained first-hand information on the child’s language level, and gave the child an opportunity to extinguish most of his or her orientation reactions to the tape recorder by having it displayed and functioning in the room, which additionally supplied information on the effects of the acoustics of the room on the tape recording.