Abstract
van Bysterveldt, A. K., & Westerveld, M. F. (2017). Children with Down syndrome sharing past personal event narratives with their teacher aides: A pilot study. International Journal of Disability, Development and Education, 64, 249–269.
In several recent issues of Word of Mouth, I discussed the importance of personal narratives and strategies for assessing and developing personal narratives. In this article, the authors investigate how children with Down syndrome in New Zealand share personal narratives with their teacher aides.
Competence in producing personal narratives has been linked to socioemotional well-being and the development of personal identity (Reed & Spicer, 2003). The NZ (New Zealand) curriculum emphasizes the importance of personal narratives in a range of communicative situations, such as “developing and conveying personal voice” (New Zealand Ministry of Education, 2007, p. 7), and “achieving a sense of coherence when constructing texts . . . through deliberate choice of content, language, and text form” (New Zealand Ministry of Education, p. 13). Considering the importance of personal narrative competence to classroom participation and access to the curriculum, it is of concern that children with Down syndrome demonstrate significant difficulties narrating past personal experiences despite the emphasis on social participation in inclusive education (Dolva, Gustavsson, Borell, & Hemmingsson, 2011; van Bysterveldt, Westerveld, Gillon, & Foster-Cohen, 2012).
Research has shown that contextual support influences children’s oral narrative performance, including children with Down syndrome (Kover, McDuffie, Abbeduto, & Brown, 2012). In general, children tell longer stories, containing a greater variety or range of words, but with greater fluency (i.e., using fewer mazes such as false starts or reformulations) when allowed to refer to pictures (Westerveld & Heilmann, 2012). Miles, Chapman, and Sindberg (2006) reported that picture support in a narrative task particularly benefited the mean length of utterance (MLU) of adolescents with Down syndrome. As a consequence, Miles et al. strongly recommended the use of pictures for intervention purposes to reduce cognitive demands and to facilitate “access to knowledge not expressed without such support.”
Narrative production is also influenced by examiner discourse characteristics (Dickinson & Porche, 2011). Although equal participation might be the aim of a conversational interaction, the elicitation of personal narratives requires a different approach from the examiner. Their role is to provide sufficient support and scaffolding for the narrator to continue and to successfully get their story across but not too much input to interrupt the flow of the story or to prescribe certain responses at the expense of the child’s own personal voice (Peterson & McCabe, 1983). The amount of adult questioning, the examiner’s MLU, or the examiner’s number of utterances per minute all influence children’s oral narrative production (Kover et al., 2012). For example, negative correlations have been found between the amount of examiner talk and the MLU produced by participants with Down syndrome in a fictional narrative condition. In this study, the authors asked the following questions:
How do children with Down syndrome perform on microstructure and macrostructure measures of personal narrative ability, using their own and standard protocol photo prompts?
Are there differences in personal narrative performance in their own and standard protocol photo prompts conditions?
What is the relationship between teacher aide language and child language during personal narratives?
The Study
Participants/Methods
Ten children with Down syndrome (five boys and five girls) ranging in age from 6.9 to 13.0 years participated. Personal narratives were elicited across two conditions: (a) protocol photos and (b) own photos. In the Protocol photos condition, children were shown 10 photos of events that are generally familiar to children (e.g., outing to the beach; a school trip; a doctor’s visit; see items in Box 1), using a standard language sampling protocol developed by Westerveld and Gillon (Westerveld, Gillon, & Miller, 2004), based on the Conversational Map Procedure that was originally devised by Peterson and McCabe (1983). The Protocol is available for free download at: http://www.education.canterbury.ac.nz/documents/gillon/languageprotocol.pdf
Box 1. Personal Narrative Prompts
Oh look who’s this? (Ronald McDonald). I went to a birthday party at McDonald’s last year. Have you ever been to McDonald’s?
We went to the beach in the holidays. These children dug a big hole in the sand and waited for the sea to fill it up. Have you been to the beach? What happened last time you went to the beach?
This little girl had to go to the Doctor, cause she had a bad cough. Have you ever been to the Doctor’s?
These friends are watching somebody arriving on a big plane. Have you ever been on a plane? Have you ever been out to the airport to watch the planes?
Oh look, this girl fell off the bars and hurt her knee. She had to go to the sick bay and they put a plaster on. Have you ever broken anything? Did you ever hurt yourself in the playground?
These children went on a school trip. They all went on a bus to a museum. Have you ever been on school trip?
The examiner read a scripted introduction before asking the child “Did anything like that ever happen to you?” If the child said yes, the examiner asked “Can you tell me about it?” In the Own Photos condition, children were asked to bring their own photos and asked to share what happened in the photos. The intent of using children’s own photos was to reduce the impact of potential difficulties in retrieving event knowledge from memory (and hence reducing cognitive demands of the task). The teacher aide was reminded to try and avoid leading questions and to use open-ended prompts (e.g., “Can you tell me more?”).
Analysis
Microstructure analysis
The following measures were calculated using the Systematic Analysis of Language Transcripts–New Zealand (SALT-NZ).
Children with Down syndrome
Mean length of utterance in words (MLU-W) as a measure of grammatical competence.
Number of different words (NDW) was used as a measure of semantic diversity.
Percent maze words (PCMZ) was calculated as a measure of verbal fluency and is sensitive to task complexity for children with language-learning difficulties.
Percent intelligible utterances (PCIntellUtt) was calculated to provide an impression of the level of intelligibility the participants demonstrated in connected speech.
Mean turn length in utterances (MnTurnUtt). This measure was used to calculate the child’s level of contribution to or control of the discourse.
Number of words used first: 1st mention. The number of words used first was calculated to obtain an overall impression of the children’s spontaneous language use.
Teacher aides
Teacher aide MLU in words (TA MLU-W).
Number of utterances produced by the teacher aide
Percentage of teacher aide utterances containing a question
MeanTurnUtt.
Macrostructure analysis
It was accomplished using high-point analysis (McCabe & Rollins, 1994).
One past event: The child mentions one past tense event.
Two past events: The narrative contains two past tense events.
Three past events: The narrative contains three past tense events related to the topic.
Leapfrog: The narrated order of at least three past tense events does not mirror a logical sequence.
Chronological: The narrative contains a sequential narration of more than three past tense events but without evaluation.
End-at-high-point: The narrative builds to high point, but there is no resolution.
Classic: The narrative builds to a high point with a resolution.
Results
Child intelligibility ranged from 69% to 98% intelligible utterances (mean around 90%) despite eligibility criteria requiring participants to be intelligible to unfamiliar listeners.
At microstructure level, language measures improved with age, and there was considerable variability across the group. MLU in words ranged from 1.7 to 5.9 revealing significant delays in grammatical ability.
At macrostructure level, there was no clear observable trend associated with age and measures of narrative quality, with most children able to narrate at least three past tense events, but only one participant producing a personal narrative that approximated a classic narrative structure.
There were no differences in personal narrative performance in response to the child’s own photos versus a set of standard photo prompts.
In general, teacher aides closely matched the participants with Down syndrome in terms of number and mean turn length of utterances.
There were negative correlations between the amount of teacher aide talk and the children’s verbal output in utterances, MLU in words, and mean turn length—the more the aides talked, the less the children talked.
Discussion
The children’s intelligible on these personal stories was less than the intelligibility of children with Down syndrome who told fictional stories from wordless picture books (Miles et al., 2006). The authors suggest that this difference in intelligibility might be due to the greater cognitive demands posed by the personal narrative task. Personal narratives require the retrieval of events from memory as well as sequencing of these events in a logical order and recounting them orally using specific words and grammar. Children with Down syndrome are known to have motor speech programming difficulties (Kumin, 2006), which may be more prominent in spontaneous language contexts where cognitive demands of the task are particularly high.
Although we did not investigate causal connections between teacher aide talk and children’s language output, we tentatively suggest that increased examiner talk may present an additional challenge, rather than support their students to tell their story. Professional development may need to address the importance of verbal scaffolding to assist the child in narrating within their linguistic zone of proximal development, where they are supported to bridge the gap from assisted narrating to a more independent performance to tell their story successfully.