Discussing stories: On how a dialogic reading intervention improves kindergartners’ oral narrative construction

https://doi.org/10.1016/j.jecp.2010.07.002Get rights and content

Abstract

Oral narrative skills are assumed to develop through parent–child interactive routines. One such routine is shared reading. A causal link between shared reading and narrative knowledge, however, has not been clearly established. The current research tested whether an 8-week shared reading intervention enhanced the fictional narrative skills of children entering formal education. Dialogic reading, a shared reading activity that involves elaborative questioning techniques, was used to engage children in oral interaction during reading and to emphasize elements of story knowledge. Participants were 40 English-speaking 5- and 6-year-olds who were assigned to either the dialogic reading group or an alternative treatment group. Analysis of covariance results found that the dialogic reading children’s posttest narratives were significantly better on structure and context measures than those for the alternative treatment children, but results differed for produced or retold narratives. The dialogic reading children also showed expressive vocabulary gains. Overall, this study concretely determined that aspects of fictional narrative construction knowledge can be learned from interactive book reading.

Introduction

A young child’s language development includes the ability to construct oral stories. In fact, gaining the ability to discuss abstract ideas socially in narrative conversations may be a fundamental motivation to learn language altogether (Donald, 1991). Constructing oral stories allows young children to verbalize real or imagined events in ways that not only communicate social messages to others but also help themselves to derive meaning from experiences (Nelson, 2007). It is essential for children to gain narrative knowledge so as to discuss and organize their lives into meaningful episodes. To be understood by naive conversational partners, however, it is suggested that children’s stories need to be structured chronologically, include causal links to the goals and motivations of characters, and provide sufficient background information (Peterson, 1994). This story construction knowledge is usually assumed to develop through parent–child routines. One routine is shared book reading.

Interestingly, correlational studies suggest that early narrative skills are related to children’s later literacy development, with moderate to strong relations existing between the production of fictional narratives and concurrent, as well as future, reading comprehension (Griffin et al., 2004, O’Neill et al., 2004, Snow et al., 2007). However, although correlational research has found links between story construction skills and literacy, there is a lack of concrete evidence supporting a causal relationship between shared book reading and storytelling skills. In the current study, we examined whether shared book reading was causally linked to fictional narrative construction abilities.

Narrative construction skills undergo extensive development between 2 and 5 years of age, beginning with toddlers uttering single-phrase two-word utterances and ending with complex multiepisodic stories. At 5 years of age, typically developing children from middle-income homes are able to produce narratives that are chronologically structured and sequential, although these stories often end abruptly without proper conclusions (Peterson & McCabe, 1983). In addition, 5-year-olds are able to include some words, such as and and then, that connect sentences together cohesively (Peterson & McCabe, 1991) and are able to include enough details about characters to allow a naive listener to understand who are the protagonists and what are some of their thoughts and motivations (Stein & Glenn, 1979).

Narrative skills do not develop at the same rate in all children; children from low-income households, for example, have been found to have deficits in storytelling skills when compared with their peers from middle-income households (Peterson, 1994). Given the discrepancy in narrative skills across economic populations, the development of narrative skills cannot be governed entirely by maturation. Instead, it is theorized that children acquire storytelling skills through social interaction. Vygotskian theory suggests that parents “scaffold” their children into developing better narrative skills (Vygotsky, 1978). More simply, during dialogues with learned others, children encounter narratives that are more sophisticated than their own and internalize the skills that allow them to improve their own constructions.

Support for this notion is found in intervention studies that focused on the development of autobiographical stories. These intervention studies showed that training mothers to ask more elaborative, open-ended “wh-” questions while engaging in autobiographical narrative dialogues with children improved the quality of narratives produced by 3- to 5-year-olds (Peterson, Jesso, & McCabe, 1999). In addition, parents can tailor questions asked to improve specific aspects of narrative skills (Peterson & McCabe, 2004). For example, parents can scaffold children to better contextualize their narratives by asking for more context detail in their questioning (e.g., who, what, when, where). Thus, through feedback from their conversation partners, children encode the relevant information on autobiographical storytelling that is emphasized in joint discussion (Nelson, 2007).

In addition to the study of autobiographical narrative knowledge as described above, researchers have investigated children’s acquisition of fictional narrative knowledge (e.g., Mandler, 1984, Stein and Glenn, 1979). Telling fictional stories seems to be an important skill linked to later literacy development given that 12 of the 15 investigations on the relation between narratives and literacy that were found required children to tell fictional stories unassisted by a conversational partner (for critical analyses of six of these articles, see O’Neill et al., 2004, Roth et al., 1996). Moreover, fictional storytelling, prompted by wordless picture books or other stimuli, is often used by speech language pathologists to assess oral language delays (Melzi & Caspe, 2008). Given the relationship between fictional narratives and later literacy, it is important to understand how and in what contexts children might acquire information on fictional narrative construction.

One context that might promote fictional narrative knowledge is book reading. The illustrations themselves in picture books may help to foster the acquisition of fictional narrative knowledge. Paris and Paris (2003) suggested that young children learn to “read” the pictures in a picture book to ascertain the meanings of narrative components and add these understandings to their narrative schemas. However, young children’s main exposure to books typically occurs during adult–child shared book reading. In these activities, adults may give information about the plot, supporting the narrative elements found in the illustrations and, thus, emphasizing narrative knowledge (Graham, 1990). For example, a picture of a boy crying may help children to understand the character’s emotional response to an event, and a parent may explain the picture to the child in terms of the events causing the child’s tears, thereby highlighting elements of narrative plot.

However, during shared book reading, parents may be supplying narrative knowledge in other ways than simply reading the text and explaining pictures. Shared reading provides a context where dialogues about the fictional stories in storybooks can be created, and these dialogues could emphasize narrative elements that are present both in pictures and in the text. Parents may ask and answer questions during shared reading, thereby creating a dialogic environment that contains many opportunities for adults to scaffold narrative elements in a manner similar to adult–child dialogues about past events. In the current study, the focus was on the possibility of children learning fictional narrative skills from shared adult–child reading.

Past research has not found a definitive answer as to whether or not shared reading influences children’s narrative knowledge. Sénéchal, Pagan, Lever, and Ouellette (2008) did not find an association between the frequency of shared reading at home and children’s ability to produce narratives, either fictional or autobiographical. That study, however, did not assess the quality of the parent–child interactions during shared reading. Investigation of the natural reading behaviors of parents with their preschoolers found that parents do not typically engage in interactive reading with their children (Whitehurst et al., 1988). In fact, parents’ reading styles typically include few dialogues, and these dialogues involve mainly yes/no questions or directives. Most often, parents read the text directly without engaging their children in the story or the discourse (Huebner & Meltzoff, 2005). Is it the case that shared reading enhances children’s narrative skills only when adults adopt a dialogic interaction style during shared reading?

Recognizing that a more active role in storybook reading may be beneficial to children’s literacy development, Whitehurst et al. (1988) designed an intervention called dialogic reading that is meant to encourage adults to create dialogues during story time (see Mol, Bus, de Jong, & Smeets, 2008, for an up-to-date meta-analysis of the dialogic reading literature). During reading sessions, the adult reader is required to encourage children’s oral contributions using elaborative “wh-” and open-ended questions, repetition of good responses, and expansion of incomplete responses to illustrate the difference between what was said and what could have been said. This reading technique has been found to positively impact literacy skills in young children; in particular, most of the research conducted on this type of intervention has focused on the impact of dialogic reading on expressive vocabulary. In fact, positive gains in expressive vocabulary were found in children between 3 and 5 years of age (Crain-Thoreson and Dale, 1999, Lonigan and Whitehurst, 1998, Wasik and Bond, 2001, Whitehurst, Arnold, et al., 1994, Whitehurst, Epstein, et al., 1994) when conducted in groups of up to eight children (Hargrave and Sénéchal, 2000, Lonigan and Whitehurst, 1998, Whitehurst, Arnold, et al., 1994) and when compared with an alternative treatment comparison group (Arnold et al., 1994, Lonigan and Whitehurst, 1998, Valdez-Menchaca and Whitehurst, 1992, Whitehurst, Arnold, et al., 1994). Young children are clearly internalizing language knowledge highlighted by interactive shared reading.

Although the dialogic reading literature focuses on improving vocabulary skills, the types of “wh-” questions used to create dialogues about storybooks are not unlike the elaborative questions that have been found to improve children’s narrative skills. Therefore, it is hypothesized that using this dialogic reading procedure, in the context of books that highlight story elements, would impact narrative skills. A search of the published literature yielded a single study testing the impact of dialogic reading on narrative knowledge. Zevenbergen, Whitehurst, and Zevenebergen (2003) determined that using a dialogic reading intervention could enhance children’s inclusion of some important aspects of narratives within their retellings of heard stories such as children’s inclusion of more details about the characters’ mental states, motivation, and dialogues. However, that study was limited to investigating only a few aspects of narrative knowledge in a retelling paradigm. The current study was conducted to determine whether a dialogic reading intervention could enhance a wider range of narrative components in both a retelling and production paradigm. These components are meant to represent the different areas of narrative knowledge that children would need to grasp so as to create chronological, complex, cohesive, and contextual narratives.

To measure a child’s structural knowledge, the narratives are analyzed for the story elements that make up a story grammar. A story grammar is a sequence of elements that are essential to a structured story plot, and as such they include introductions, setting details, character descriptions and emotional/cognitive responses of the characters, initiating events, plans to solve conflicts, attempts to solve conflicts, reactions to events, and conclusions (Stein & Glenn, 1979). Story grammar units are the basic elements of a story that organize story events in a sequential and meaningful way. In fact, the structure of a narrative is considered to be equivalent to the gist or central meaning of a story (Mandler, 1984). Internal responses and internal plan story grammar units were assessed separately as a subset of narrative structure called mental state references. This was done to confirm findings by Zevenbergen et al. (2003).

Measures of language complexity that demonstrate a child’s ability to produce linguistic components within the narratives were used. These measures include the total number of words, the ratio of number of different words spoken over total number of words, and the mean length of utterances. The ratio of different words to total words reflects the amount of varied language used and expresses the complexity of the child’s narrative language. Furthermore, the mean length of utterances was calculated to determine whether the child was more likely to have longer and more complex syntax. This measure was first described as a measure of language complexity by Brown (1973) and subsequently was used as a measure of syntactic knowledge for children up to 5 years of age (Bishop and Adams, 1990, Miller and Chapman, 1981).

A child’s use of decontextualized language such as anaphora can serve as an index of contextual knowledge. Anaphora is a linguistic term for an expression referring to a previously mentioned idea. When a character is introduced, it is appropriate to first mention this character by his or her name or the title of the being preceded by an indefinite article. For example, the proper way to introduce a rabbit character is to call him or her a rabbit or Sam. Once a character has been introduced within the focus of the discourse, it is acceptable to refer the character by a pronoun or another appropriate label such as he or the rabbit in the case of the bunny character. Young children, however, have difficulty with this grammar rule and often assume shared knowledge with a naive listener and introduce characters in their narratives as he or she (Hickmann & Hendriks, 1999).

The integration of structure, content knowledge, and linguistic knowledge can be assessed by counting the number and variety of connectives produced (Peterson & McCabe, 1991). Connectives are said to tie the story together by semantically relating clauses together. It may be the story grammar units that structure the overall narrative, but the connectives act to link the story together at the local clausal level (Cain, 2003). Higher quality narratives might differ not only in the number of connectives but also in the variety of connectives used. Therefore, the appropriate use of cohesive ties was measured in the current study.

Within the field of fictional narrative research, no consensus has been determined on the narrative paradigm that best assesses a child’s narrative construction ability. Two main types of tasks dominate the field: narrative retelling tasks and narrative production tasks. Narrative retelling tasks are so named because participants are first told an oral story and then asked to retell this story at some point thereafter; however, the prompt, procedure, listener, time period, and medium vary across the literature. In any given retelling task, a participant may be asked to retell a story in the presence of a familiar listener (Merritt & Liles, 1989) or a naive listener (Botting, 2002) and based on a series of pictures or a picture book (Hesketh, 2004), a video (Merritt & Liles, 1989), a film strip (Coelho, 2002), a purely oral narrative with no props (Ukrainetz et al., 2005), and so on.

Narrative production tasks, alternatively, are tasks in which participants are asked to produce a novel fictional narrative based on an initial prompt. Similar to retelling tasks, production tasks have a variety of contexts and methods that are not standardized. For example, participants may be asked to produce a fictional story based on story stems (the first sentence of a story [Merritt & Liles, 1989]), one picture (Coelho, 2002), several pictures (Hickmann & Hendriks, 1999), a wordless storybook (Botting, 2002), a video (Eaton, Collis, & Lewis, 1999), and so on.

Because no consensus on the type of task, even within the retelling or production genres, can be achieved, it seems ill advised to treat retelling and production tasks as tasks that elicit equivalent narrative skills. In fact, comparison studies have suggested that the retelling task may be easier for children, resulting in children demonstrating a more logically structured, linguistically complex, connected and contextualized story than when they are asked to produce a novel story (Merritt & Liles, 1989). Furthermore, it is possible that the retelling task is more a measure of children’s comprehension of story elements than a measure of their ability to construct a narrative (Nelson, 2007). In the current study, both narrative production and retelling tasks were assessed to distinguish any differing effects of shared book reading on these two paradigms.

This study tested whether a shared reading intervention would improve kindergartners’ narrative ability as compared with children receiving an alternative treatment. The intervention was administered in small groups twice a week for 8 weeks. It was hypothesized that dialogic reading should enhance the four key aspects of narrative ability—story structure, language complexity, cohesion, and decontextualized language—through the appropriate use of anaphora. Furthermore, the retelling task should be easier for children to complete; therefore, the effects of the intervention should be more discernible in the children’s retelling stories.

Section snippets

Participants

Participants were 40 English-speaking 5-year-olds who were recruited from kindergarten programs in a large city in central Canada. The kindergarten classrooms were in schools identified by their school board as located in neighborhoods that have high concentrations of low-income households and greater mobility of residents than neighborhoods for other schools in the school board district. However, a letter sent home through teachers to parents asking for consent for their children’s involvement

Preliminary analysis

Preliminary analyses of children’s receptive vocabulary were conducted using the PPVT-III (Dunn & Dunn, 1997). No differences in receptive vocabulary scores were found between groups prior to intervention (p = .54). The mean PPVT-III standard scores were 94.86 (SD = 14.11) for the dialogic reading children and 97.47 (SD = 12.63) for the alternative treatment children. PPVT-III scores were used as a covariate in subsequent analyses if these scores were related to child outcomes.

Although all children

Discussion

Oral narrative skills are considered to be essential to young children’s social communication (Donald, 1991) and their ability to comprehend and organize event knowledge (Nelson, 2007). Moreover, autobiographical oral narrative skills are assumed to develop in the context of parent–child conversations (Peterson et al., 1999). The goal of the current study was to test whether adult–child conversations while reading storybooks would enhance fictional narrative skills. The main hypothesis was that

References (53)

  • K. Cain

    Text comprehension and its relation to coherence and cohesion in children’s fictional narratives

    British Journal of Developmental Psychology

    (2003)
  • R. Case

    General and specific views of the mind, its structure, and its development

  • C.A. Coelho

    Story narratives of adults with closed head injury and non-brain injured adults: Influence of socioeconomic status, elicitation task, and executive functioning

    Journal of Speech, Language, and Hearing Research

    (2002)
  • C. Crain-Thoreson et al.

    Enhancing linguistic performance: Parents and teachers as book reading partners for children with language delays

    Topics in Early Childhood Special Education

    (1999)
  • M. Donald

    Origins of the modern mind

    (1991)
  • L.M. Dunn et al.

    Manual for the Peabody Picture Vocabulary Test—third edition (PPVT-III)

    (1997)
  • J.H. Eaton et al.

    Evaluative explanations in children’s narratives of a video sequence without dialogue

    Journal of Child Language

    (1999)
  • D.B. Elkonin

    U.S.S.R

  • J. Graham

    Pictures on the page

    (1990)
  • T.M. Griffin et al.

    Oral discourse in the preschool years and later literacy skills

    First Language

    (2004)
  • A. Hesketh

    Grammatical performance of children with language disorder on structured elicitation and narrative tasks

    Clinical Linguistics & Phonetics

    (2004)
  • M. Hickmann et al.

    Cohesion and anaphora in children’s narratives: A comparison of English, French, German, and Mandarin Chinese

    Journal of Child Language

    (1999)
  • B. MacWhinney

    The CHILDES project: Tools for analyzing talk

    (2000)
  • B. MacWhinney et al.

    The child language data exchange system

    Journal of Child Language

    (1985)
  • J.M. Mandler

    Stories, scripts, and scenes: Aspects of schema theory

    (1984)
  • K. McCartney et al.

    Best practices in quantitative methods for developmentalists

    Monographs of the Society for Research in Child Development

    (2006)
  • Cited by (211)

    View all citing articles on Scopus
    View full text