Wordless Picture Books: Assessing Narrative Comprehension Skills and Expressive Language Ability in Children with Language Disorders and their Typical Peers : Wordless Picture Books: Assessing Narrative Comprehension Skills and Expressive Language Ability in Children with Language Disorders and their Typical Peers Jill Markon
Minneapolis Public Schools Background : Background In recent years, speech-language pathologists and classroom teachers have collaborated to improve instruction for children struggling to comprehend and produce narratives, and to identify children who may have language disorders based on a narrative sample (as a component of a pre-referral process). This collaboration is due in part to the correlation between narrative ability and reading achievement (see Paris & Paris, 2001 for a review) as well as evidence suggesting relationships between receptive language disorder and narrative comprehension (Miles & Chapman, 2002), and expressive language disorder and narrative production (Kaderavek & Sulzby, 2000). Slide 3: Several methods have been used to assess narrative comprehension and production, one method including a comprehension-only wordless picture book task (Westby, 1999). This task is relatively short (within 15 minutes), easy to administer, literature based, and fits well within a general reading/writing curriculum. Compared to a retelling task, in which children are asked to retell a story told to them, it demands less memory and contains a high level of permanent, visual support. These may be important variables for children with language disorders. Due to the nature of the task, in which participants tell the story as they go through the pages, a wordless picture book may serve a dual purpose, to assess narrative comprehension and assess expressive language ability. Purpose of the Study : Purpose of the Study The purpose of the present study was to establish a small set of local norms for narrative comprehension skills and expressive language ability using a wordless picture book. The target population was kindergarten and first grade children. Participants included children with combined receptive-expressive language disorders and their grade level peers. Each narrative was scored for two measures of comprehension (number of story lines expressed, number of details expressed) and three measures of expressive language (length, complexity, and correct grammar of utterances). The dependent variables were based on Berman’s (1988) and Miles and Chapman’s (2002) research as well as Minneapolis Public School’s expressive protocol. Research Questions : Research Questions First, given a wordless picture book task, what is the average number of story plot lines and story details expressed by typically developing kindergarten and first grade students? Compared to their peers, do children with language disorders differ in the average number of story plot lines and story details expressed? Second, within the picture book task, what are the average length, complexity, and grammatical correctness of utterances expressed by typical kindergarten and first grade students? Compared to their peers, do children with language disorders differ in these variables? Finally, using a factor analysis, which of these variables best measures underlying language ability? Participants : Participants Data from 28 participants were collected. Twenty of the 28 were typically developing; eight of the 28 were children with combined receptive-expressive language disorders based on Minnesota State Eligibility (-2.0 standard deviations below the mean on two standardized tests). The language disorder sample represented an actual caseload, which included Caucasian and African American children. According to annual IEP(s), hearing and vision were normal. Measures for general intelligence were not collected, since this is not district protocol until age 7. There were no standardized tests of language at the time of data collection. Slide 7: While kindergarten and first grade students were the targeted populations, limited data from preschool and second were also collected in order to interpret developmental trends. Data were collected in the Jan-Feb of 2002 and again in 2003 (a two-year span). Due to school transfer, data for some participants were collected in one year only. Over the course of two years, the number of typically developing participants and those with language disorders are listed below: Task : Task The wordless picture book, Froggie Goes to Dinner (Mercer Mayer), was used as the narrative stimulus. Mayer’s collection of frog books has been used extensively in the literature. Each participant was given a variation of the directions, “Do you know this book? It does not have any words. Your job is to tell me the story.” Before telling the story, participants completed a picture walk by looking at each page four seconds. After viewing the whole book, participants began their stories. Each story was audiotaped, transcribed, and scored for narrative comprehension and expressive language ability. The total task time varied between 8 and 12 minutes. The scoring procedure is listed below. Narrative Comprehension : Narrative Comprehension Each narrative was scored one point for mention of each the six story plot lines:
1. Onset of Problem: Frog jumps in boy’s pocket
2. Discovery of Problem: Boy notices frog at restaurant
3. Problem: Frog causes trouble at restaurant
4. Solution: Waiter tries to catch frog
5. Consequence: Boy’s family must leave restaurant
6. Conclusion: Boy and frog are punished Each narrative was scored one point for mention of each the nine story details:
1. Frog jumps into the saxophone
2. Musician has reaction
3. Frog jumps into woman’s salad
4. Woman has reaction
5. Frog jumps into man’s drink
6. Frog kisses man
7. Boy has reaction to waiter
8. Family has reaction to leaving
9. Family has reaction to boy/frog’s behavior Expressive Language : Expressive Language Utterances were scored for expressive language ability according to Minneapolis Public School’s standard protocol. See protocol manual for exact scoring procedures. General procedures are described below.
1. Mean Length of Utterance (MLU). Each word received one point, with the exception of exact repetitions, fillers, proper names, and contractions (predicate contractions counted as two words).
2. Mean Complexity of Utterance. Each utterance received a score from 0 to 10, with 0 for a single word and 10 for an embedded subordinate clause.
3. Percentage of Grammatically Correct Utterances. With consideration of dialect, each utterance was scored as grammatical or ungrammatical based on any element of grammar (e.g., tense, agreement, articles, word order, etc.). The percentage was then calculated. Results: Group Differences : Results: Group Differences A Welch Two Sample T-Test was used to analyze group differences. The scores below represent the average score within each participant group across each grade level. Significance *p = .01 level. LD = Language Disorder Results: Factor Analysis : Results: Factor Analysis A factor analysis was completed to determine which of the five measures (plot, details, MLU, complexity, grammar) was best correlated with underlying language ability. The factor analysis was completed for kindergarten and first grade only, due to the low number of children with language disorders in preschool and second grade. Key Findings : Key Findings With the exception of MLU in first grade, children with language disorders were significantly different from their typical peers in all narrative comprehension and expressive language measures
Story plot and story details:
1) skills increased for all participants, regardless of group
2) story plot was near 100% by second grade for typical peers
3) participants with language disorders did not show a relative strength for story plot or story details
1) MLU, complexity, and grammar increased more for children with language disorders than for typical peers
2) grammar increased for typical children from preschool to first grade (90% by first grade) while children with language disorders were still behind (69%) Key Findings : Key Findings Children with language disorders in kindergarten and first grade were at least one year behind in all measures
Complexity of utterances had the highest correlation with underlying language ability in both kindergarten and first grade, followed by story plot and mean length of utterance
As children progressed to first grade, correlations with underlying language ability decreased with the exception of complexity of utterances, possibly suggesting task ceiling effects
Grammar had an especially low correlation in first grade, which may be explained by participants’ type of disability (combined receptive-expressive disorder) and high grammar scores in for typical peers in kindergarten, first, and second grade. While grammar is an important clinical marker for Specific Language Impairment, SLI participants often have expressive disorder rather than combined receptive-expressive disorder. Implications : Implications The purpose of the present study was to explore assessment of narrative comprehension and expressive language ability using a wordless picture book. With the exception of MLU in first grade, children with language disorders were significantly discrepant from their typical peers in both narrative comprehension and expressive language ability. Although this was a small study, these findings suggest a wordless picture book task may be useful for indicating the presence of language disorder (thus, leading to a more thorough evaluation) as well as planning a dynamic treatment progression for improving narrative skills. In my school district, it has been particularly useful in the pre-referral stage of assessment. Given six new referrals since initial data collection, the results of this task were aligned with standardized tests of language and other language samples. References : References Berman, R.A. (1988). On the ability to relate events in narrative. Discourse Processes, 11, 469-497.
Kaderavek, J.N., & Sulzby, E. (2000). Narrative production by children with and without specific language impairment: Oral narratives and emergent readings. Journal of Speech, Language, and Hearing Research, 43, 34- 49.
Miles, S., & Chapman, R.S. (2002). Narrative content as described by individuals with down syndrome and typically developing children. Journal of Speech, Language, and Hearing Research, 45, 175-189.
Paris, A.H., & Paris, S.G. (2001). Assessing narrative comprehension in young children. Reading Research Quarterly, 38, 36-76.
Westby, C.E. (1999). Assessing and facilitating text comprehension problems. In H.W. Catts & A.G. Kamhi (Eds), Language and reading disabilities (pp. 154-222). Needham Heights, MA: Allyn & Bacon. Kindergarten Samples: Typical Peer : Kindergarten Samples: Typical Peer Kindergarten Sample: Language Disorder : Kindergarten Sample: Language Disorder Contact Info : Contact Info Jill Markon
Bryn Mawr Elementary
252 Upton Ave.
Minneapolis MN 55405