CHILDES Narrative English Hicks Corpus

Deborah Hicks
Partnership for Appalachian Girls' Education

Participants: 20 1st graders, 18 2nd graders, 5 5th graders
Type of Study: online narration, storytelling
Location: USA
Media type: no longer available
DOI: doi:10.21415/T5GG6H

Browsable transcripts

Download transcripts

Citation information

Publications using these data should cite:

Hicks, D. (1990). Kinds of texts: Narrative genre skills among children from two communities. In A. McCabe (Ed.), Developing narrative structure. Hillsdale, NJ: Erlbaum.

Additional references include:

Berman, R. A. and D. I. Slobin (1994). Relating events in narrative: A crosslinguistic de-velopmental study. Hillsdale, NJ, Lawrence Erlbaum Associates.

Heath, S. (1983). Ways with words: Language, life and work in communities and classrooms. Cambridge, Cambridge University Press.

Quirk, R., S. Greenbaum, et al. (1972). A grammar of contemporary English. London, Longman.

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

The narratives in this directory were collected by Deborah Hicks in the context of a study of primary school children’s narrative genre skills, focusing on their ability to produce a range of kinds of narratives. In the study, children from three primary grade levels — first, second, and fifth —were shown a shortened version of the silent film, “The Red Balloon.” After viewing the film, children were asked to tell the film’s events in three different ways: as a factual news report, an ongoing event case, and as a more embellished story. These three narrative genres are representative of what Heath (1983) terms “key” narratives, or narratives that are found crossculturally in children’s language learning environments. The narrative data were coded by utterances for linguistic forms that might mark genre differences.

This directory contains four subdirectories: 1st, 2nd, 5th, and del. The first three are taken from first, second, and fifth graders in Cambridge, Massachusetts. The fourth is taken from a lower-class SES group in Delaware. For comparison with the Delaware children, these 12 files in the 1st grade directory were used: 4, 5, 12, 16, 27, 29, 30, 35, 38, 39, 40, and 42.

The children with files in the subdirectories called 1st, 2nd, and 5th were first grade, second grade, and fifth grade students in a private elementary school in Cambridge, Massachusetts. The majority of students attending this school were members of middle class families in which one or both parents were working professionals, so that these children could be considered members of mainstream culture. The classrooms were somewhat progressive in nature, so that children were free to choose from a range of activities those that they would work on. Many of the activities that children performed regularly were language activities, such as recounting on tape a story of how the world was created, writing about “what we did in science class,” and recounting personal experiences during sharing time episodes. The narrative genre tasks were thus presented to the children as one of the many options available, and in all but a few cases, children were willing and eager to leave the room for the tasks.

Before performing any of the narrative tasks, children were told that they would watch a film and would then tell what happened in the film in three different ways. In the case of the online narration task, children listened to the experimenter saying “This is [child’s name] and Deborah, sportscasters, and we’re gonna say everything we see happening in the film. I’m gonna start off and then [child’s name] is gonna take over.” The child then watched the 3-minute segment of the film and then the experimenter started the narration by saying “The little boy and the red balloon are going past a church steeple. And they’re coming to a bakery shop. The little boy is looking inside the bakery shop. Now he’s checking in his pocket to see if he has enough money to buy something to eat. Looks good. Now he’s walking into the bakery shop.” Then the experimenter turned to the child and asked, “Can you take over now and be the sportscaster?” The order of the report and event cast tasks was randomly selected within grade levels.

The storytelling task was performed separately from the report and event cast, in a session that took place approximately 1 hour after the completion of the first two tasks. This particular research design was chosen on the grounds that performance of three consecutive tasks would be too demanding for many of the children in the study, particularly the 5-year-old children. For the storytelling task, the leading given by the experimenter was “This is [child’s name] and Deborah, and we’re gonna be storytellers and tell the story of The Red Balloon. I’m gonna start off and then [child’s name] is gonna take over.” During this, the experimenter holds a “storybook” which has on the front cover a picture from the film but which has neither words nor pictures inside. The experimenter then says “The Red Balloon. Once upon a time there was a little boy who lived in Paris, France. One day, on his way to the bus stop, he found this big beautiful red balloon. He wanted the balloon to be his friend.” Then the experimenter turns to the child and says, “Can you take over now and be the storyteller?” At this point, the experimenter passes the storybook to the child.

In an attempt to create some degree of homogeneity in the data, in addition to providing an interaction with the highest possible degree of ecological validity, children were provided with a great deal of contextual support for the tasks. As was noted in the introductory section to this chapter, children were reminded before each task of the particular narrative “voice” they were to assume: that of a news reporter, a sportscaster, or a storyteller. For the storytelling task, children were also given a storybook containing only a single picture on the outside cover, which they were encouraged to hold during the story narration.

The data obtained from the study were transcribed in CHAT and analyzed using the CLAN programs for child language analysis. The entire narration was divided into clauses with one clause on each line of the CHAT transcript. The segmentation of the narrative data was done on the basis of clause units of analysis, following Berman and Slobin (1994). Clause units were defined as any linguistic utterance containing a predicate, so that the following would all be considered separate units of analysis: “he climbed up the stairs”, “when the boy was inside the bakery shop,” and “the boys who stole the balloon.” Segmentation of complement clauses was done so that the utterance “he saw the balloon floating by the door” was segmented into “he saw” and “the balloon floating by the door.” Utterances containing verbs with three arguments such as “he told the balloon to stay by the door” were segmented as “he told the balloon” and “to stay by the door.”

The narrative data were examined in terms of children’s use of specific linguistic forms representing three basic subsystems: a) syntactic constructions, b) temporal and event ex-pressions, and c) indexical clauses. The analysis of syntactic constructions was designed to assess possible genre differences in the syntactic complexity of event casting, reportative, and story narratives. The analysis of expressions of temporality and event relations was designed to examine genre differences in how temporal and logical relations between events were expressed in the discourse. Finally, the analysis of indexical clauses was an attempt to assess genre differences in how children went beyond the basic narration tasks to provide evaluative or descriptive information about events in the narrative.

1.1 Coding

The codes for assessing genre differences in children’s narratives consists of three basic sections representing different linguistic subsystems. The codes for syntax are drawn from Berman and Slobin (1994) and Quirk and Greenbaum (1972). The codes are designed to assess differences in syntactic complexity among the three narrative genres. The codes for temporality and event relationships are drawn primarily from Berman and Slobin (1994) and are designed to assess genre differences in the use of verb forms, aspectual markers, timemarkers and logical connectors. Finally, the measures for indices and intensifiers are designed to assess differences in the marking of nonmainline event clauses or in the embellishment of narrative clauses through the use of highlighters and intensifiers.

   1.1.1 Syntactic Codes
INDindependent clause
SADVsubordinate adverbial clause “holding his balloon”
CMPcomplement clause
RELrelative clause
DIAdirect dialogue, quoted from characters

   1.1.2 Event Codes
T:SQtemporal sequential marker
T:CNtemporal connective
T:ADVtemporal adverb
MODVmodal verbs
HYPhypothetical statement
FUTfuture constructions
L:CONTone event is contingent on another, as in “when X, they Y”
L:TRone event is contrasted with another
C:TEMclauses in temporal sequence
C:ADclauses in adversive relationship
C:C:REAclauses in reason relationship

   1.1.3 Indexical Codes
I:DESindicator of description
I:COMindicator of comparison
I:Sindicator of internal state
I:S:EMOindicator of emotional state
I:S:PHYindicator of physical state
I:S:MENindicator of mental state verb
I:Eindicator of evaluation
I:Oindicator of orientation
I:METindicator of metaphor or simile
I:METAindicator of metacomments
I:TNGindicator of tangential remarks
E:DESevent descriptions
E:HABhabitual events
E:I:EMOinternal state events
E:I:MENinternal mental states
E:I:PHYinternal physical states
C:METcommentary involving metaphors or similes
C:COMcommentary on narration
EX:ADJadjectival expansions
EX:ADVadverbial expansions
Ppitch marking