CHILDES English Post Corpus

Kathy Post
Graduate School of Education
Harvard University

Participants: 3
Type of Study: naturalistic
Location: USA
Media type: audio
DOI: doi:10.21415/T5DP46

Browsable transcripts

Download transcripts

Citation information

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 study examined children from families living in Suwannee County, Florida. This area was chosen because it fits the criterion of being a rural, Southern community that has a predominantly white, working-class population. It is also the family home of this inves-tigator’s parents and numerous aunts, uncles, and cousins. Through work on the qualifying paper, contacts had already been established at the Suwannee County Health Department to obtain access to the immunization records. In addition, the investigator’s shared regional accent and shared cultural experiences allowed easy access to and ready acceptance by the study families.


The participants were three girls and their families. Two of the girls were approximately 19 months of age at the time of selection and the third was 22 months. One participant was the second-born, one was the third-born, and the third participant was the fourth-born child in the family. The next-oldest sibling in each family was approximately 18 months older than the third- and fourth-born participants and the older sibling was 2 years older than the second-born participant. All of the participants were the products of normal pregnancies, were healthy at the time of taping and had no apparent hearing, speech or mental deficiencies. They were located through immunization records obtained from the Suwannee County Health Department and the records of a pediatrician in Live Oak. The mothers were contacted by telephone and a meeting arranged to ascertain if the family met the criteria for the study.


At the initial meeting the mothers were told that the investigator was interested in ob-serving how children change or develop over time. Because the families were selected ac-cording to the age of the younger child, the younger child was called the target child. It was emphasized, however, that we were interested in how both children developed over time, not just the younger one. It was explained to the families that they were entering a long-term commitment to the study. It was agreed that the sum of $100.00 would be paid to the families as an incentive to complete the study. Fifty dollars was paid to each family at the end of the first taping session and the remaining $50.00 was paid at the end of the final tap-ing session. In addition, at the end of the fifth taping session, the investigator took the fam-ilies out to lunch. The mothers also signed a consent form. The mothers were interviewed periodically and asked their views on child rearing, how their children spend the day, meth-ods of discipline, goals for their children, and other related issues. They were asked more specific questions on how they thought their children learn language, how the parents affect their children’s language learning, how important language learning is, how they thought their children will do in school, what they thought their children should be able to do by the time they enter kindergarten, and what they perceived as the long-term and short-term ef-fects of their children’s education. In addition, the mothers were asked how they felt about their children watching television and if it should be regulated or not. In the ethnographic tradition, detailed descriptions of the community and the children and their families are provided. The descriptions help the reader understand the environment in which these chil-dren are learning language. The purpose of these portraits is to add a richness and completeness to the data.

Data Collection

Within one week of the initial meeting a taping session was conducted in each partici-pant’s home. Taping was accomplished with a Sony™ Video 8 AF portable videocassette recorder. Prior to the taping the recording equipment was brought into the participants’ homes to allow the children to examine the equipment to lessen the disruptive influence on the normal interaction. Occasionally, during the taping sessions, one of the children would come over to examine the camera, but generally, the camera was ignored once the toys were introduced. Each taping session lasted approximately 60 minutes. A toy bag provided by the investigator was introduced at the beginning of the session and the family played with whatever interested them. The toys were appropriate for the ages of the children and includ-ed such items as books, blocks, action figures, puzzles, pull-toys, and one playset. New toys and books were added over the taping sessions to maintain interest. Occasionally the participants would bring out a favorite toy or book of their own. Taping sessions were scheduled at times which the mothers deemed most convenient and most likely to have the children in a receptive mood. Taping took place approximately every four weeks. Because of illness or vacations, occasionally the sessions were scheduled a little farther apart. A total of 10 sessions was recorded for each family. Thus the language development of these three children was followed over a period of about 9 months during the latter part of their second year and the early part of their third year. This age was chosen because during this time the child’s language is expanding greatly in complexity of syntax and size of vocabulary. The investigator was present for all of the tapings and made contextual notes to aid in transcription. The presence of an outsider and recording equipment no doubt affects the in-teraction among the family members. For this reason one “practice” taping session was made with each family, the data from which was not included in later analyses. It was ob-served that with repeated exposure to the taping situation the disruptive effects of the data collection methods were minimal.

Data Coding

Transcripts of the tapes were made as soon as possible after the session by the investi-gator. The data was transcribed in CHAT. In determining how to divide up maternal utter-ances, it was decided to use breath groups. That is, what the mother said on one breath was considered to be one utterance. One of the mothers tended to have longer utterances than the other two. On occasion, this mother (Darla) might take a quick supplemental breath, but the pause would be less than one second and it was clear that she was continuing the same utterance. The feedback data were obtained using the modified coding scheme described in Demetras, Post, and Snow (1986).