Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Apr 16.
Published in final edited form as: Lang Speech Hear Serv Sch. 2013 Apr;44(2):211–223. doi: 10.1044/0161-1461(2012/12-0021)

Structural and Dialectal Characteristics of the Fictional and Personal Narratives of School-age African American Children

Monique T Mills 1, Ruth V Watkins 2, Julie A Washington 3
PMCID: PMC3988833  NIHMSID: NIHMS567492  PMID: 23633645

Abstract

Purpose

To report preliminary comparisons of developing structural characteristics associated with fictional and personal narratives in school-age African American children.

Method

Forty-three children, grades two through five, generated a fictional and a personal narrative in response to a wordless-book elicitation task and a story-prompt task, respectively. Narratives produced in these two contexts were characterized for macrostructure, microstructure, and dialect density. Differences across narrative type and grade level were examined.

Results

Statistically significant differences between the two types of narratives were found for both macrostructure and microstructure but not for dialect density. There were no grade-related differences in macrostructure, microstructure, or dialect density.

Conclusion

The results demonstrate the complementary role of fictional and personal narratives for describing young children's narrative skills. Use of both types of narrative tasks and descriptions of both macrostructure and macrostructure may be particularly useful for characterizing the narrative abilities of young school-age African American children, for whom culture-fair methods are scarce. Further study of additional dialect groups is warranted.

Keywords: narrative, assessment, African American English

Spoken narration is an important window into language ability, because it provides a naturalistic and internally complex context for examining children's language use and reveals the procedures by which children organize, comprehend, and produce language (Champion, Seymour, & Camarata, 1995). In addition, narration has practical significance for school-age children. A large part of school curricula involves tasks that are narrative in nature, such as storytelling, summarizing, retelling, and reporting (Hughes, McGillivray, & Schmidek, 1997). Further, narration is often included in state educational benchmarks, like the Common Core Standards adopted by 45 of 50 states. For example, by grade two in the Common Core Standards, students are expected to “tell a story or recount an experience with appropriate facts and relevant, descriptive details, speaking audibly in coherent sentences” (National Governors Association Center for Best Practices & Council of Chief State School Officers [NGA Center & CCSSO], 2010, p. 23).

When children tell stories with well-formed macro- and microstructure, they demonstrate key language growth and are positioned to meet state benchmarks. Narrative macrostructure refers to the hierarchical structure that moves the listener from the beginning of the story, when the setting and characters are introduced, through a plot that develops to its resolution at the end of the story (Hughes et al., 1997). Macrostructural narrative language includes statements regarding time, place, and characters and temporal sequencing of narrative events. Narrative microstructure refers to how words and utterances work together to build a cohesive story (Hughes et al., 1997) and so depends on syntactic and semantic productivity, complexity, and accuracy. Microstructural narrative language is assessed by such measures as total number of utterances, number of different words, clause density, and use of cohesive devices that children produce in their narratives.

Children's narratives may also be affected by their dialect use. In fact, in a study of 4- and 6-year-old children, Ross, Oetting, and Stapleton (2004) found that those who produced greater levels of AAE also produced better narrative structure than those with lower levels of dialect density. Given the importance of macrostructure and microstructure in descriptions of narrative development, knowing how dialect use influences narrative structure may facilitate more accurate characterizations of the narrative development of young African American English (AAE) speakers. However, the relationship between dialect use and narrative structure has not been examined for school-age AAE speakers. Knowing more about how the structural characteristics of narrative and dialect use are shaped by two commonly used but rarely compared narrative contexts (fictional and personal) may add critical information about how narratives develop and, therefore, how they should be assessed. This may be particularly crucial for AAE-speaking children, for whom culture-fair assessments are lacking.

Fictional and Personal Narrative Differences

Children across cultures produce both fictional (Berman & Slobin, 1994) and personal narratives (Bliss & McCabe, 2008). The two narrative types differ in the manner or style in which they are presented. Fictional narratives are accounts that are presented about fabricated events and characteristics (e.g., an alien who visits a family in a park). Conversely, personal narratives are presented as accounts of specific, real-world events that have been experienced by the narrator or someone known to the narrator (e.g., a cousin who got in trouble last week for arriving home late). It is not surprising that fictional and personal narratives place different cognitive demands on the narrator (Hudson & Shapiro, 1991). For instance, content knowledge may be more easily accessed in personal than in fictional narration because of the narrator's familiarity with the events and characters in the story. The relative ease of access to content helps to explain observations suggesting that personal narratives are the earliest acquired and most prevalent narrative type produced by school-age children (Hudson & Shapiro, 1991).

A child's ability to produce a well-formed fictional or personal narrative may vary as a function of his or her age, language ability, and cultural practices. In the remainder of this introduction, we review the major ways these factors influence fictional and personal narration. We then review existing literature on narrative assessment of school-age African American children and present a synopsis of the development of macrostructural, microstructural, and dialectal narrative language in African American children.

Age

In a study comparing fictional and personal narration across three grade groups (preschool, first, third), Hudson and Shapiro (1991) found that, as they developed, children included a greater number of macrostructural and microstructural elements in fictional and personal narratives. Students across all three grade groups produced one macrostructural element more frequently in personal than in fictional narratives: descriptions, or statements about what the characters or the story scene looked like. The authors concluded that children in preschool through third grade understand the importance of orienting listeners by providing setting information in personal narratives. At present, there is a dearth of information about developmental differences between fictional and personal narrative structure in older school-age children.

Language Ability

Differences between fictional and personal narration have been explored in school-age children with and without typical language ability (Allen, Kertoy, Sherblom, & Pettit, 1994; Hudson & Shapiro, 1991; McCabe, Bliss, Barra & Bennett, 2008; Shiro, 2003). McCabe et al. (2008) used high point analysis to compare macrostructural completeness of fictional and personal narratives produced by children ages 7 to 9 years with language impairment. For these children, macrostructural completeness was greater in personal than in fictional narratives. The authors suggested that “children with language impairment [were] more capable of personal versus fictional narrative discourse” (p. 200).

In contrast, among typically-developing low- and middle-income school-age children narrative macrostructure has been found to be better developed in fictional narratives than in personal narratives. For example, Shiro (2003) appraised evaluative language (i.e., statements including information about emotion, cognition, perception, physical state, intention, relation, and reported speech) of fictional and personal narratives in 113 typically-developing first- and fourth-grade Venezuelan children, ages 6 and 10 years. The author found that evaluative language was produced more in fictional than in personal narratives and, therefore, suggested that narrative competence must be assessed in more than one context, given the impact that task-related factors may have on narrative abilities.

Allen et al. (1994) showed that fictional and personal narratives follow different developmental trajectories in a study of narrative type differences among 36 4- to 8-year-old European American children with high and low language ability. Results indicated that narrative structure of children in both language ability groups varied depending on the narrative type: specifically, children produced personal stories with lower levels of complexity (e.g., reactive sequence and complete episode) and fictional stories with higher levels of complexity (e.g., action sequence and multiple-episode). In terms of microstructural differences (e.g., number of utterances), Allen et al. found a language ability by narrative type interaction, such that children with low language ability produced a greater number of utterances in personal than in fictional narratives while children with high-language ability produced a greater number of utterances in fictional than in personal narratives.

These studies suggest that, among typically developing children, the personal narrative advantage that exists in grades preschool through three (Hudson & Shapiro, 1991) shifts to a fictional narrative advantage as children mature through later school-age years (Allen et al., 1994; Shiro, 2003). Conversely, there is a personal narrative advantage for language-impaired children (McCabe et al., 2008), who seem to follow the trajectory of younger school-age children (Hudson & Shapiro, 1991).

Cultural Practices and Narrative Assessment of School-age African American Children

In addition to influences related to age and language ability, fictional and personal narration is also influenced by language socialization, because children may have more or less experience with each narrative type depending on their cultural background. Personal narratives have been reported as the earliest-acquired and most prevalent type of narrative produced in children from the U.S. (Hudson & Shapiro, 1991). However, although children from European American communities are likely to experience narratives as factual renditions of personally experienced events and as fictional accounts read from storybooks, African American children are likely to experience narratives as personal accounts told amid spirited conversation (Heath, 1982, 1983), which are often told in fictionalized ways. That is, personal narratives may in fact be “tall tales” in which the events are embellished.

Additionally, the spoken narration of African American children may be influenced by aspects of an oral tradition that values audience participation (e.g., call and response), nonlinearity (e.g., linking events thematically rather than temporally), and stylistic embellishment (e.g., verbal dueling) (Michaels, 1981; Smitherman, 2000; Van Keulen, DeBose, & Weddington, 1998). Given that narrative practices differ cross-culturally, it is imperative that cultural bias in narrative assessment be minimized, which can only occur with greater knowledge of narrative development in children from different cultural backgrounds.

Although narrative features such as those described in the preceding section are central to the African American oral tradition and thus may be regarded as strengths, traditional narrative assessments may undervalue or even misinterpret these features as weaknesses. Indeed, differences in narrative discourse expectations between home and school may place African American children at a disadvantage academically (Heath, 1983; Weddington, 2010), especially when only limited discourse samples and measures are included in assessments. African American children are also at risk for misdiagnosis with language impairment when features of their narratives are perceived as errors reflecting language disorder rather than language difference stemming from cultural storytelling patterns (Curenton, 2006). Therefore, it is imperative to identify culture-fair metrics to assess narrative skills in academic and clinical contexts.

Narrative Development in African American Children

Because studies of narrative development can provide a basis for normative comparisons of fictional and personal narratives, we provide an overview of macrostructural, microstructural, and dialectal development of spoken narration in African American children from toddlerhood through the school-age years.

Macrostructural Development

Most of the work on the development of narrative macrostructure in African American children has mainly been done in the fictional narrative context. To date, only one study has examined differences between personal and fictional narratives of African American children, and it focused on toddlers (Sperry & Sperry, 1996). Two- to 3-year-old African American children included more events and episodes in narratives produced in fictional contexts (e.g., fantasy, pretend play) than in personal ones (e.g., relating past event). Taken together with language ability literature suggesting fictional narrative advantages in typically-developing school-age children and the cultural literature suggesting that African American children have more exposure to personal narratives, Sperry and Sperry's finding may indicate that very young African American children are more attuned to the artful aspects of personal narratives that keep the audience captivated than to the referential aspects that keep the story moving forward linearly. Further, the personal narratives to which these children are exposed may be highly embellished and performative (Heath, 1983).

With age, African American children demonstrate greater macrostructural completeness in their fictional narratives. For example, Price, Roberts, and Jackson (2006) showed that while 4-year-olds mainly produced story endings and attempts to solve problems, 5-year-olds included more types of macrostructural elements (e.g., setting, initiating event, internal response, attempts). Young narrators also demonstrate cognitive advancement through the use of mental- or internal-state words (e.g., think, feel, know). For example, 4- and 5-year-old African American and European American children included more such words than 3-year-olds in their fictional narratives (Curenton, 2004; Curenton & Justice, 2004). By age seven, African American children produce more embellishment and fantasy than their Latin American and European American peers (Gorman, Fiestas, Peña, & Clark, 2011). Taken together, results of these studies indicate that African American children gain skill in telling why the story is of interest to the audience. Child narrators do this by evaluating the character's motivations and intentions as the character attempts to resolve a conflict or achieve a goal (e.g., finding a missing pet).

Studies of personal narration indicate that African American children produce a repertoire of styles that includes sequential, moral-centered stories as well as the performative narratives that are characteristic of African American oral traditions (Bloome, Champion, Katz, Morton, & Muldrow, 2001; Champion et al., 1995; Champion, 1998). Champion suggested that elicitation context and culture influenced the macrostructural narrative language of school-age African American children. At present, direct comparisons of macrostructural narrative language development between fictional and personal narratives in African American children are critically lacking.

In this study, we made use of an expressive elaboration analysis as the measure of macrostructure, because it includes artful elements of narration that are highly esteemed by African American communities (Van Keulen et al., 1998), and it is sensitive to developmental differences in school-age children from ethnically diverse backgrounds (Ukrainetz, Justice, Kaderavek, Eisenberg & Gillam, 2005). Artfulness is measured in terms of the specific types of expressive elaboration, as follows: Appendages, which signal the beginning, middle and end of the story; Orientations, which provide setting and character information; and Evaluations, which make the story events increasingly lucid (Ukrainetz et al., 2005). Appendix A provides examples of each of these macrostructural elements.

Microstructural Development

As with macrostructural narrative development of African American children, we have derived much of our knowledge of microstructural narrative language from the fictional narrative context. Syntactic productivity, complexity, and accuracy have been examined in preschool and school-age African American children (Curenton, 2004; Curenton & Justice, 2004; Horton-Ikard, 2009). In terms of syntactic productivity, Curenton found, in a sample of African American and European American children, that 3-year-olds produced fewer utterances and had a shorter average sentence length than 4- and 5-year-olds. Curenton and Justice found that 3-year-old African American and European American children also produce fewer conjunctions than 4- and 5-year-olds. One study examining the microstructural narrative language of school-age African American children focused on syntactic accuracy (Horton-Ikard, 2009). Findings indicated that cohesive adequacy improved with age, such that 9- and 11-year-olds produced more referential markers such as pronouns than 7-year-olds to discuss previously mentioned information in their fictional narratives. Collectively, these findings indicate that African American children increase syntactic productivity in their spoken narratives with age.

While studies of the development of microstructural narrative language in African American children have focused primarily on syntax, studies of child narrators of other ethnicities have focused on semantics as well as syntax. For example, bilingual (Spanish/English) children produced a higher number of different words and complex syntax in first grade than in kindergarten in their fictional narratives (Uccelli & Páez, 2007). Similarly, Fey, Catts, Proctor-Williams, Tomblin and Zhang (2004) found that the number of different words, clause density, and percentage of grammatical utterances produced in fictional narratives increased from age 7 to age 9 in an ethnically mixed sample of children (4% African American). Currently, there is a dearth of information on microstructural development, both syntactic and semantic, in the personal narratives of African American children.

Dialectal Development

Dialect density—the extent to which speakers produce features of AAE—has been examined in the fictional narratives of preschool and school-age children (Connor & Craig, 2006; Craig et al., 2009; Ivy & Masterson, 2011; Ross et al., 2004). Connor and Craig found that preschoolers who produced fictional narratives with high or low dialect densities demonstrated stronger early literacy skills (e.g., letter-word recognition, phonological awareness) than children who narrated with moderate levels of dialect density. Craig et al. demonstrated that between first and second grades, dialect density of spoken narratives was inversely related to reading achievement scores. Dialect density was not measured across grade level in either of these studies, so it is unclear whether dialect use increased, decreased, or plateaued with age.

AAE-speaking child narrators demonstrated an increase in dialect density from age four to six (Ross et al., 2004). In their study, Ross et al. (2004) found that dialect use, specifically use of the preterite had + V-ed construction (e.g., they had looked for the frog), was associated with narrative skill, such that children who produced more complex narrative macrostructure also produced greater dialect density. Ross et al. (2004) collected samples that may have been personal or fictional narrations, but they analyzed them jointly so it is unknown whether this finding holds across the two narrative types. By school-age, it appears that African American children's dialect use in fictional narration may level off. Ivy and Masterson (2011) showed that spoken dialect density between third and eighth grade students did not differ. Currently, there is a dearth of developmental data on dialect use in the personal narratives of AAE-speaking children.

In summary, African American children produce greater macrostructural completeness and microstructural productivity and accuracy as they age. While there is an increase in dialect density during the preschool years, dialect density stabilizes during the school-age years. At present, no studies have compared structural and dialectal narrative characteristics between fictional and personal narrative of African American children.

The Current Study

The current study expands our knowledge of the developing structural differences between fictional and personal narratives among African American children. To compare the two narrative types across different language abilities, we included children across several grade levels. We posed the following research questions:

  1. How do fictional and personal narratives differ in macrostructure (expressive elaboration), microstructure (lexical diversity, syntactic complexity), and dialect density?

  2. How do narratives from children with typical language development differ in narrative macrostructure, microstructure, and dialect density across four grade levels (2-5)?

It was hypothesized that there would be an overall fictional narrative advantage for expressive elaboration, because fictional narratives would have a higher rate of Appendages and Evaluation, while personal narratives would only have a higher rate of Orientations. We predicted a fictional narrative advantage in microstructural narrative language. We expected dialect density to be higher in the personal context given that the story content would be familiar. We hypothesized that macrostructural and microstructural narrative language would increase with grade level and that dialect density would remain stable across grade level.

Method

Participants

Fifty school-age, typically-developing African American students from three school districts (four elementary schools) in Central Illinois participated in the current study. Seven children were excluded from analyses because they were not African American (n = 5) or because they had language impairment (n = 2). Classroom records were used to determine race/ethnicity and parent reports were used to determine history of speech and language services. Data from the remaining 43 children are presented here. These children were in self-contained general education or self-contained gifted education classrooms in grades 2 through 5. Sixty-eight percent (n = 27) of the children were from low socioeconomic status (SES) backgrounds. SES was determined by free or reduced lunch status as reported by parents. Parent report was unavailable for three students; for these students, SES was determined by the percentage of children in their school receiving free or reduced lunch. Since this number was above 70%, those three students were considered low SES.

Children's spoken dialect use was measured with Part I of the Diagnostic Evaluation of Language Variation–Screening Test (DELV–S) (Seymour, Roeper, de Villiers, & de Villiers, 2003). The DELV–S has criterion scores that categorize children as speaking with strong, some, or no variation from Mainstream American English (MAE). In this sample, 34.9% (n = 15) of the children produced no variation from MAE, 20.9% (n = 9) produced some variation from MAE, and 44.2% (n = 19) produced strong variation from MAE.

Receptive and expressive language skills were assessed with the Peabody Picture Vocabulary Test, Fourth Edition (PPVT-4) (Dunn & Dunn, 2007) and the Expressive Vocabulary Test-Second Edition (EVT-2) (Williams, 2007), respectively. The PPVT-4 is a test of single-word receptive vocabulary, and the EVT-2 is a test of single-word vocabulary production. Language tests were administered in separate sessions from narrative elicitation. Table 1 displays language characteristics of the child narrators in this study. The Behavioral and Social Sciences Institutional Review Board at The Ohio State University approved the present study.

Table 1.

Characteristics of Child Narrators

Grade
2nd (n = 13) 3rd (n = 11) 4th (n = 9) 5th (n = 10)
Measure M SD M SD M SD M SD
Age (years; months) 7;3 .43 8;4 .44 9;5 .47 10;2 .31
PPVT-4 (standard score) 103 16.1 114 22.5 104 11.6 101 8.9
EVT-2 (standard score) 112 18.5 121 22.1 121 17.0 114 15.2
Open in a new tab

Note. PPVT-4 = Peabody Picture Vocabulary Test 4th ed.; EVT-2 = Expressive Vocabulary Test 2nd ed.

Narrative Elicitation Procedures

The first author, a bidialectal (AAE/MAE) African American, elicited one fictional and one personal narrative from each participant. The examiner spoke the dialect produced by the child to elicit the most naturalistic narrative. Fictional narratives were elicited following the protocol of Berman and Slobin (1994), in which the examiner asked individual participants to think of a story to tell as they looked through all the pictures in the wordless book, Frog, Where Are You? (Mayer, 1969). The examiner directed the participant back to the beginning of the book when the student had completed reviewing all pictures. The child then generated a story while looking at the pictures in the book. The examiner intermittently prompted participants to continue as needed (e.g., “Tell me more,” “Mhm”). If the child made no ending statement, then the examiner asked “Is that the end?”

Personal narratives were elicited using a conversational map procedure in which individual participants generated a story after listening to a story prompt (Peterson & McCabe, 1983). Consistent with procedures described by Mainess, Champion, and McCabe (2002), the examiner asked students to teach her a card game. The card games ranged in type (e.g., go fish, crazy eights, a game the child made up) and in complexity. The purpose of this task was to prepare the child for a narrative context by having her or him practice talking about a child-friendly topic. During the ongoing card game, the examiner presented one of five story prompts until a narrative was successfully elicited. The child-friendly prompts are presented in Appendix B. In most cases (n = 40), only one story prompt was needed to elicit a narrative. The examiner elicited more than one personal narrative from a few students (n = 3) who did not respond to an earlier prompt. In these cases, only the longest personal narrative was included in the dataset for analysis. All story samples were audio recorded using a Marantz PDP 201 (Itasca, IL) compact disc recorder with an external table microphone. Fictional and personal narratives were elicited in the same session. Narrative order was counterbalanced to reduce sequencing effects.

Narrative Analysis

Oral narratives were orthographically transcribed using Systematic Analysis of Language Transcripts (SALT) (Miller & Iglesias, 2010). Graduate research assistants in communication sciences and disorders segmented narrative transcripts into C-units, using the scoring criteria developed by Loban (1976). C-units are independent clauses plus their modifiers, including one main clause along with accompanying subordinate clauses. C-unit segmentation has been established as an appropriate procedure for examining oral language samples (Loban, 1976) and has been utilized in previous studies of narrative (Hester, 2010) and discourse (Craig, Washington, & Thompson-Porter, 1998; Thompson, Craig, & Washington, 2004) abilities of African American children. Children in this study produced fictional stories that ranged from 19 to 84 C-units in length (M = 38.30, SD = 12.04) and personal stories that ranged from three to 45 C-units (M = 11.91, SD = 7.75).

Descriptions of dependent measures

The four dependent measures in the study were used to assess narrative macrostructure (expressive elaboration), narrative microstructure (lexical diversity, syntactic complexity), and dialect density.

Expressive elaboration

The expressive elaboration analysis of macrostructural narrative language follows procedures outlined by Ukrainetz and Gillam (2009). This analysis consisted of the following 14 elements within three categories: Appendages (introducer, abstract, theme, coda, ender), Orientations (name, relation, external, personality), and Evaluations (modifier, expression, repetition, internal state, dialogue). We examined each C-unit for presence of a codable word or phrase. Words and phrases could receive only one within-category code but multiple between-category codes. For instance, “It's about a missing frog named Anthony” would be coded as It's about [abstract] a missing [modifier, personality] frog named Anthony [name]. The word missing receives two codes—as modifier [Evaluation] and personality [Orientation]—because these are between-group, rather than within-group category codes. For efficiency, each occurrence of a specific element was coded only on the first mention.

Appendages (except theme) and Orientations were awarded 0 or 1 point. Unlike Appendages (except theme) and Orientations, Evaluations and theme were expected to occur multiple times throughout a story. Therefore, Evaluations and theme received 0 or 1 point for up to two occurrences. The 20 possible points for expressive elaboration are derived from 6 possible points from Appendages, 4 from Orientation, and 10 from Evaluation. Appendix A displays the expressive elaboration scoring system.

Lexical diversity

Number of different word roots rate (NDW rate)

The number of different word roots (NDW), a measure of lexical diversity, is the total number of different word roots (i.e., words without inflection) in a narrative. Children in the current study produced fictional stories that contained 49 to 151 different word roots (M = 100.07, SD = 24.65) and personal stories that contained 15 to 142 different word roots (M = 49.33, SD = 24.58). Following the procedures outlined in Greenhalgh and Strong (2001), NDW was divided by total number of C-units to control for differences in fictional and personal narrative length, yielding a measure of NDW rate.

Syntactic complexity

Clause density (C-Density), also referred to as a subordination index, is a measure of syntactic complexity. We used SALT's definition of a clause (Miller & Iglesias, 2010), which excludes non-finite forms. C-Density is calculated by dividing the total number of clauses by the total number of C-units. Clauses, main or subordinate, were statements consisting of a subject (They) and a finite predicate (looked out the window).

Dialect density

To determine the extent of AAE produced by child narrators, the Dialect Density Measure (DDM; Craig & Washington, 2006) was calculated by dividing the total number of morphosyntactic tokens by the total number of words in each narrative. DDM has been used to identify systematic differences in dialect production rates (Craig & Washington, 2006).

Interrater Agreement

Blind interrater agreement was conducted by a second scorer on 17 of the 86 narratives (20%) consisting of a random selection of 8 personal and 9 fictional narratives. Point-to-point comparisons were made to determine concordance on C-unit segmentation, bound morpheme marking, word transcription, DDM, C-Density, and expressive elaboration. These comparisons were calculated by dividing the number of agreements by the total number of possible agreements and disagreements.

Narrative transcription

Concordance on occurrences of C-units—independent clauses plus their modifiers—resulted in agreement rates of 99.7%. Concordance on morpheme marking (e.g., frog/s = plural vs. frog/z = possession) resulted in agreement rates of 98.5%. Concordance on the presence of a word (e.g., in vs. on) resulted in perceptual agreement rates of 98%.

C-Density

Concordance on number of clauses within each C-unit resulted in agreement rates of 95.3% for fictional narratives and 86.7% for personal narratives.

DDM

Concordance on occurrences of a morphosyntactic feature of AAE within each narrative resulted in agreement rates of 94% for fictional narratives and 100% for personal narratives.

Expressive elaboration

Concordance on the presence of expressive elaboration elements within each narrative resulted in agreement rates of 94.8% for fictional and 88.8% for personal narratives. Expressive elaboration categories showed the following agreement rates for fictional narratives: Appendages was 86.8%, Orientations was 97.5% and Evaluations was 90.4%. Expressive elaboration categories showed the following agreement rates for personal narratives: Appendages 73.2%, Orientations 93.3%, and Evaluations 89.7%.

Results

The aim of this study was to examine context- and grade-related differences in structural and dialectal narrative characteristics used by school-age AAE speakers. Means and standard deviations for variables related to macrostructure (expressive elaboration scores), microstructure (NDW rate, C-Density), and dialect (DDM) for fictional versus personal narratives are presented in Table 2. Means for DDM did not differ by narrative type. Means for NDW rate and C-Density were higher in the personal than in the fictional context. Means were higher for overall expressive elaboration scores in fictional than in personal narratives. Figure 1 displays the percentage of maximum possible points for each expressive elaboration category, i.e., Appendages, Orientations, and Evaluations. In the fictional context, child narrators provided 38%–48% of possible points for each category. In the personal context, child narrators provided fewer possible points for each category (28%–45%).

Table 2.

Descriptive Statistics of Macrostructural, Microstructural, and Dialectal Narrative Language Variables

Narrative Context
Variable Fictional Personal
M SD M SD
Macrostructural Variable
EE 8.28 2.19 6.40 3.04
Microstructural Variables
NDW rate 2.68 0.45 4.55 1.16
C-Density 1.12 0.08 1.20 0.24
Dialect Density
DDM 0.02 0.01 0.02 0.01
Open in a new tab

Note. EE = expressive elaboration; NDW rate = Number of different word roots per communication-unit; C-Density = clause density; DDM = dialect density measure.

Figure 1.

Figure 1

Open in a new tab

Mean percentage of maximum possible points for three categories of expressive elaboration (Appendage = 6 points, Orientation = 4 points, and Evaluation = 10 points) across fictional and personal narratives.

Given that children were in either general or gifted classrooms, we first established whether differences in narrative language existed based on classroom type (general, gifted). A repeated measures multivariate analysis of variance (MANOVA) was conducted to examine the classroom differences in macrostructural (expressive elaboration scores), microstructural (NDW rate, C-Density), and dialectal (DDM) narrative language across narrative context. The dependent variables were expressive elaboration scores, NDW rate, C-Density, and DDM. The within-subjects variable was narrative context (fictional, personal). The independent variable was classroom type (general, gifted). There was a main effect of classroom on the combined dependent variables, F(4, 38) = 3.07, p < .05, Wilks’ lambda = .75, ηp2= .24. In follow-up univariate tests using a Bonferroni adjusted alpha level of .012, no classroom type differences reached statistical significance: expressive elaboration, F(1, 41) = 1.73, p = .19, ηp2= .04; NDW rate, F(1, 41) = 3.57, p = .06, ηp2 = .08; C-Density, F(1, 41) = .02, p = .88, ηp2 = .00. Classroom differences in DDM did not reach, but trended toward, statistical significance, F(1, 41) = 6.13, p = .017, ηp2 = .13. The narrative context by classroom type interaction did not reach statistical significance, F(4, 38) = 24.32, p = .94, Wilks’ lambda = .28, ηp2= .01. Narratives from children in general and gifted classrooms were included together in all subsequent analyses, because there were no classroom type differences in any of the study variables. Table 3 displays the number of children in general and gifted classrooms across the second, third, fourth, and fifth grades.

Table 3.

Number of Child Narrators in General and Gifted Classrooms

Classroom Grade
2nd 3rd 4th 5th Total
General 8 6 2 7 23
Gifted 5 5 7 3 20
Total 13 11 9 10 43
Open in a new tab

Narrative Context Differences

A one-way between-groups multivariate analysis of variance (MANOVA) was performed to examine narrative context differences in macrostructural (expressive elaboration scores), microstructural (NDW rate, C-Density), and dialectal (DDM) narrative language. The dependent variables were: expressive elaboration scores, NDW rate, C-Density, and DDM. The independent variable was narrative type (fictional, personal). There was a main effect of narrative context on the combined dependent variables, F(4, 81) = 25.47, p < .001, Wilks’ lambda = .44, ηp2= .55. In follow-up univariate tests using a Bonferroni adjusted alpha level of .012, the only context-related differences to reach statistical significance were the macrostructural variable, expressive elaboration scores, F(1, 85) = 10.80, p < .012, ηp2 = .11, and one microstructural variable, NDW rate, F(1, 85) = 96.00, p < .012, ηp2 = .53. There were no context-related differences in C-Density, F(1, 85) = 3.83, p = .053, ηp2 = .04 or in DDM, F(1, 85) = .16, p = .68, ηp2 = .00. An inspection of the mean scores indicated that child narrators produced higher NDW rates in personal narratives than in fictional narratives. Conversely, child narrators showed greater expressive elaboration in fictional narratives than in personal narratives.

Given that expressive elaboration scores were comprised of three categories—Appendages, Orientations, and Evaluations—a follow-up MANOVA was conducted to assess the effect of narrative context on expressive elaboration category scores. There was a main effect of narrative context on the combined dependent variables, F(3, 82) = 4.37, p = .007, Wilks’ lambda = .86, ηp2= .14. In the univariate tests using a Bonferroni adjusted alpha level of .016, the only context-related differences to reach statistical significance were Appendages, F(1, 85) = 11.13, p < .016, ηp2 = .12, and Evaluations F(1, 85) = 6.66, p = .012, ηp2 = .07. Appendages, such as introducers and enders, were produced more in fictional narratives (M = 2.53, SD = .82) than in personal narratives (M = 1.79, SD = 1.20). Similarly, Evaluations, such as interesting modifiers and internal state words, were produced more in fictional narratives (M = 3.84, SD = 1.47) than in personal narratives (M = 2.79, SD = 2.21). There were no context-related differences in the production of Orientations (e.g., name and relation), F(1, 85) = .45, p = .501, ηp2 = .01.

Grade and Age Differences

A one-way between-groups ANOVA was conducted to examine the impact of grade level on each dependent variable. Scores on each variable were summed across fictional and personal narratives. The dependent variables were: expressive elaboration scores, NDW rate, C-Density, and DDM. The independent variable was grade level. Participants were divided into four groups based on their grade level, second through fifth. Grade level differences in narrative macrostructure did not reach statistical significance: expressive elaboration scores, F(3, 42) = 2.44, p = .07, η2 = .16. Similarly, grade differences in narrative microstructure variables did not reach statistical significance: NDW rate, F(3, 42) = 1.39, p = .25, η2 = .09; C-Density, F(3, 42) = .05, p = .98, η2 = .00. Grade differences in dialectal narrative language did not reach statistical significance, DDM, F(3, 42) = 1.38, p = .26, η2 = .08.

We divided participants into grade groups, but this method of assessing developmental growth in narrative structure did not take advantage of the continuous age spread of participants. We therefore investigated the relationship between age and the study variables using the Pearson product-moment correlation coefficient. There were no significant correlations between age and any of the study variables. Correlation coefficients were small, ranging from r = −.08 to r = .13. Table 4 shows the intercorrelations of age and the study variables.

Table 4.

Summary of Intercorrelations, Means, and Standard Deviations for Age and Narrative Language Scores

Measure 1 2 3 4 5 M SD
Intercorrelations
1. Age - .13 −.12 −.10 −.08 8.68 1.21
2. EE - −.09 .19 −.19 14.67 4.09
3. NDW rate - .44* −.32 7.23 1.28
4. C-Density - −.24 2.33 .26
5. DDM - .04 .02
Open in a new tab

Note. Macrostructural (EE), microstructural (NDW rate, C-Density), and dialectal (DDM) measures were summed across fictional and personal narratives. EE = expressive elaboration; NDW rate = number of different word roots per communication unit; C-Density = clause density; DDM = dialect density measure.

*

p < .01.

Narrative ability of children in this study did not differ based on classroom, grade level, or age. However, it was critical to rule out language ability as a spurious variable. Therefore, we tested the relationship between language ability (as measured by EVT-2 and PPVT-4 scores) and the study variables using the Pearson product-moment correlation coefficient. There was a negative correlation of medium strength between EVT-2 scores and DDM, r = −.34, n = 43, p < .05. There was a negative correlation of medium strength between PPVT-2 scores and DDM, r = −.31, n = 43, p <.05. DDM was the sole study variable related to PPVT-4 and EVT-2 scores, as shown in Table 5.

Table 5.

Summary of Intercorrelations for Narrative Language and EVT-2 and PPVT-4 Scores

Measure EVT-2 PPVT-4
1. EE .03 −.11
2. NDW rate .27 .21
3. C-Density .01 −.19
4. DDM −.34* −.31*
Open in a new tab

Note. Macrostructural (EE), microstructural (NDW rate, C-Density), and dialectal (DDM) measures were summed across fictional and personal narratives. EE = expressive elaboration; NDW rate = number of different word roots per communication unit; C-Density = clause density; DDM = dialect density measure. EVT-2 = Expressive Vocabulary Test, 2nd ed.; PPVT-4 = Peabody Picture Vocabulary Test, 4th ed.

*

p < .01.

Discussion

We examined context- and grade-related differences in the structural and dialectal characteristics of the spoken narratives of school-age AAE speakers. Results indicated structural differences between fictional and personal narratives. Fictional narratives were more artful, containing more expressive elaboration, than were personal narratives. While fictional narratives contained more expressive elaboration, personal narratives contained more lexical diversity. Narrative types did not differ in their dialectal characteristics. There were no grade-level-related differences in expressive elaboration, lexical diversity, syntactic complexity or dialect density of African American children in grades two through five.

Context-Related Differences in Narrative Macrostructure

Because narrative macrostructure tends to be better developed in fictional than in personal contexts for typically-developing school-age children, we expected expressive elaboration scores to be higher in fictional than in personal narratives for our sample of African American children. Consequently, our finding of a fictional narrative advantage in expressive elaboration is congruent with previous studies showing a fictional narrative advantage over personal narrative in evaluative language (Shiro, 2003) and in episodic structure (Allen et al., 1994) among typically-developing school-age children.

While children in the current study produced more Appendages and Evaluations in the fictional narrative context than in the personal narrative context, performance on Orientations did not differ across narrative context. This findings is incongruent with previous findings suggesting that young school-age children produce more Orientations (e.g., setting) in personal than in fictional narratives (Hudson & Shapiro, 1991). This incongruence may be due to study design differences in elicitation task and in analytic strategy. Children in both studies were given a verbal prompt in the personal context. However, children in the present study produced a fictional narrative based on a wordless picture book while children in Hudson and Shapiro's (1991) study produced a fictional narrative based on a verbal prompt, “Can you tell me a make-believe story about [a birthday party, the doctor's office, Halloween, and a trip]?” Thus, children in the present study had the visual advantage of seeing important setting and background information to support the production of Orientations in the fictional context to a greater extent than did children in Hudson and Shapiro's (1991) study. In terms of analytic strategy differences, narrative context was a within-subjects variable in the present study but a between-subjects variable in Hudson and Shapiro's (1991) study. It may be that personal narrators in their study understood the importance of providing background information while fictional narrators did not.

In this study, expressive elaboration scores were higher in fictional than in personal narratives, but there was substantial room for growth in the two story types. That is, students in the current study did not reach the highest available scores on any expressive elaboration element (see Figure 1). Ukrainetz and Gillam (2009) reported expressive elaboration scores in fictional narratives of typically-developing 8-year-olds that were higher than those of 8-year-olds in the present study. However, the present study included a smaller sample of children than Ukrainetz and Gillam (2009). Scores reported by Ukrainetz and Gillam were based on a sample of 26 children, while expressive scores of 8-year-olds in the present study were based on a sample of 11 children. In addition, children in Ukrainetz and Gillam's (2009) study received a verbal model as well as visual support in their two fictional narrative contexts (picture sequence and single picture). In the current study, children also had visual support in the fictional context (wordless book); however, they did not hear a sample fictional narrative before generating a story of their own. Since Ukrainetz and Gillam (2009) pooled expressive elaboration scores across two elicitation contexts, it is difficult to determine whether expressive elaboration of fictional narratives differed as a function of elicitation context. Therefore, further investigation is needed to determine the best context within which to elicit fictional narratives that are rich in expressive elaboration.

Lower than previously reported scores on expressive elaboration obtained across narrative type in this study may have stemmed from the narrative elicitation procedures used. Although the wordless-book and child-friendly story prompts were selected to elicit school-style and home-style narratives, respectively, the examiner did not vary her elicitation style between narrative conditions. She remained in a teacher-like mode when presenting story prompts for personal narrative elicitation, which may have reduced overall performativity and artfulness of child narrations.

Context-Related Differences in Narrative Microstructure

Narrative microstructure has been shown to depend on language ability and narrative type (Allen et al., 1994). That is, children with low language ability tend to produce more utterances in personal than in fictional narratives, while children with high language ability tend to produce more utterances in fictional than in personal narratives. Similarly, children in the current study also produced more utterances in fictional than in personal contexts.

In this study, narrative microstructure was measured in children with typical language ability using syntactic complexity and lexical diversity. Findings from the current study indicated that syntax was equally complex across narrative contexts; however, lexical diversity differed by narrative context. That is, when narrative length was controlled, child narrators produced more different word roots in personal stories than in fictional stories. Perhaps child narrators are best able to produce diverse vocabulary in the context that is most familiar to them. Because personal narratives represent past experiences rather than imagined ones, they may facilitate access to content knowledge to a greater extent than fictional narratives; however, this microstructural language advantage in the personal context may come at the expense of macrostructural narrative language.

Limitations

This study was primarily limited by a small sample size, which hampered our ability to detect significant grade-level and age differences. Small sample size was only an issue for between-subjects comparisons but not for within-subjects comparisons, where narrative type differences were detected. This sample of African American children was not only small, but it was also diverse. Children represented different ability levels, because they were being educated in either general or gifted education classrooms. Because children from these two types of classrooms were not equally distributed across each grade (see Table 4), we speculated that classroom differences accounted for the lack of significant grade-level differences in narrative structure. However, classroom differences did not reach statistical significance for any of the study variables.

We speculated that grade-level effects may have been attenuated by the inclusion of a precocious group of third graders who achieved higher PPVT-4 scores than second, fourth, and fifth graders and who outperformed second and fifth graders on the EVT-2 (see Table 1). However, we did not find that age, grade level, or language ability was related to structural characteristics in this small sample. In terms of dialectal characteristics, we found no association between DDM and age or grade level, which aligns with previous research indicating that dialect density of spoken narratives plateaus during the school-age years (Ivy & Masterson, 2011). However, we found a medium negative association between DDM and three language measures: NDW rate, PPVT-4 scores, and EVT-2 scores. This finding should be interpreted with caution and warrants further investigation of the specific morphosyntactic features of AAE used by children with higher scores and children with lower scores on the three language measures. It is encouraging that vocabulary is an area of language that is malleable and responsive to direct instruction in school-age African American children from low-SES backgrounds (Steele, Willoughby, & Mills, 2012).

Children in this study did not undergo a comprehensive language assessment. Instead, they were only tested on vocabulary, which has been previously associated with narrative ability (Uccelli & Páez, 2009). The data collection protocol was necessarily limited by time constraints of the school setting, in which there was a critical need to protect instructional time.

Narrative elicitation procedures also presented challenges. Personal narratives were elicited by a conversational map procedure that included a somewhat distracting card game. The card game may have encouraged conversation, rather than narration.

Interrater agreement was higher for fictional narratives than for personal narratives. Disparate levels of interrater agreement may have existed because coders had different degrees of familiarity with the fictional and personal stories that the child participants were attempting to tell. It may have been easier to achieve agreement on expressive elaboration coding for fictional narratives because the fictional stories were all based on the same wordless book. Thus, the structure for fictional narratives was confined to the characters and events in the stimuli. In contrast, personal narratives included characters and events that were both inconsistent across participants and unfamiliar to coders.

Future Direction

The next step in this work is to examine structural and dialectal characteristics in the fictional and personal narratives of a larger set of school-age African American children to determine whether grade-level effects are present. Personal narratives, which were elicited during a card game, tended to be shorter than fictional narratives. Future studies should elicit personal narratives from school-age children using child-friendly story prompts in the absence of such a potentially distracting routine.

Expressive elaboration coding may need to be modified in future studies when children produce narratives with multiple episodes. The expressive elaboration coding scheme was originally developed by Ukrainetz and Gillam (2009) to assess macrostructural narrative language of single-episode narratives; however, Appendages (theme, coda) and Orientations (external, personality) may occur more than once in multiple-episode narratives. For instance, the setting of a story may begin in a bedroom and end in a forest. In this case, up to two points should be awarded for external orientations. Similarly, a character's personality may shift from selfish to selfless as she or he encounters and resolves one or more conflicts. Future studies should modify expressive elaboration scoring by increasing the maximum possible points available when assessing multiple-episode narratives. The expressive elaboration scoring scheme provided by Ukrainetz and Gillam (2009) will be appropriate for comparing single-episode narratives with multiple-episode narratives.

It will be important to assess written as well as spoken narratives in African American school-age children. State benchmarks for writing are inclusive of fictional and personal narratives, requiring children in grades one through five to “write narratives to develop real or imagined experiences or events using effective technique, well-chosen details, and well-structured event sequences” (NGA Center & CCSSO, 2010, p. 18). Results of this study suggest that fictional narratives offer a promising context for assessing narrative writing, because they contain more macrostructural completeness than personal narratives.

Additional comparisons of the structural and dialectal characteristics of fictional and personal contexts are needed to determine whether these narrative types develop along different paths. Examinations comparing narrative types across spoken and written contexts will also be valuable. Which structural and dialectal elements are robust across spoken and written contexts? An answer to this question will improve our ability to identify points of intervention for macrostructural and microstructural narrative language as well as points of instruction in AAE–MAE code-switching.

Clinical Implications

In our efforts to distinguish African American children with language differences from those with language impairment, it is critical to develop culture-fair methods of assessing their language ability. Expressive elaboration analysis holds promise as a culture-fair method of assessing the macrostructural narrative language skills of young school-age African American children, because it includes categories that may be differentially valued across home (Evaluations) and school (Appendages, Orientations) contexts. Findings from this preliminary study suggest that fictional narratives from a wordless book may be the best context in which to elicit a story with Evaluations, which are valued in African American homes, as well as Appendages, which are valued in U.S. schools. Fictional and personal narratives would yield similar rates of Orientations, which are highly-esteemed in the school setting. Thus, expressive elaboration analysis may bridge home and school language expectations. Further, the current study did not find a correlation between dialect density and expressive elaboration. This may indicate that expressive elaboration analysis would not penalize the use of AAE. Additional studies with children from additional dialect groups, Mainstream and non-Mainstream, are needed to further validate the use of expressive elaboration as a culture-fair narrative assessment.

African American children in this study performed similarly to an ethnically diverse group of American same-age peers in a previous study of expressive elaboration with respect to the patterns of Appendage, Orientation and Evaluation production (Ukrainetz & Gillam, 2009). That is, children across these two studies produced more Orientations than Appendages, and more Appendages than Evaluations. Ukrainetz and Gillam (2009) demonstrated that expressive elaboration skills differed as a function of age and language ability. In contrast, we did not find a relationship between expressive elaboration skills and age in our small sample of school-age children with typical language ability.

Results from this preliminary study suggest that microstructural and macrostructural narrative language may present differently across different narrative contexts. Macrostructural narrative language may be weaker in personal than in fictional contexts, but microstructural narrative language may be less mature in fictional than in personal contexts. Therefore, clinicians may design their narrative language interventions accordingly.

Conclusion

Findings from this study indicate that elicitation context may have relevant influences on the quality of narrative macrostructure and microstructure, but not on dialect use. Fictional narrative contexts may support production of artful narrative macrostructure, while personal narrative contexts may facilitate production of diverse lexical items in school-age African American children. Therefore, the best overall view of a child's narrative ability will be obtained by assessment of fictional as well as personal contexts.

Supplementary Material

Appendix A

Acknowledgments

The authors are grateful to the children who participated, to the parents who granted them permission, and to the school staff who created space for data collection. We thank the research assistants who helped code narrative data. This research note is based on the first author's dissertation, which was completed under the advisement of the second author. This research study was supported by the Diversifying Higher Education Faculty in Illinois Program, grant T32HD007489 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, and a research grant from the University of Wisconsin Systems Institute on Race and Ethnicity awarded to first and third authors.

Contributor Information

Monique T. Mills, The Ohio State University

Ruth V. Watkins, University of Illinois at Urbana-Champaign

Julie A. Washington, Georgia State University

References

  1. Allen MS, Kertoy MK, Sherblom JC, Pettit JM. Children's narrative productions: A comparison of personal event and fictional stories. Applied Psycholinguistics. 1994;15:149–176. [Google Scholar]
  2. Berman RA, Slobin DI. Relating events in narrative: A cross-linguistic developmental study. L. Erlbaum Associates; Hillsdale, NJ: 1994. [Google Scholar]
  3. Bliss LS, McCabe A. Personal narratives: Cultural differences and clinical implications. Topics in Language Disorders. 2008;28:162–177. [Google Scholar]
  4. Bloome D, Champion T, Katz L, Morton MB, Muldrow R. Spoken and written narrative development: African American preschoolers as storytellers and storymakers. In: Harris JL, Kamhi AG, Pollock KE, editors. Literacy in African American communities. Lawrence Erlbaum Associates; Mahwah, NJ: 2001. pp. 45–76. [Google Scholar]
  5. Bloome D, Katz L, Champion T. Young children's narratives and ideologies of language in classrooms. Reading and Writing Quarterly. 2003;19:205–223. [Google Scholar]
  6. Champion T, Seymour H, Camarata S. Narrative discourse of African American children. Journal of Narrative and Life History. 1995;5:333–352. [Google Scholar]
  7. Champion TB. “Tell me somethin’ good”: A description of narrative structures among African American children. Linguistics and Education. 1998;9:251–286. [Google Scholar]
  8. Connor CM, Craig HK. African American preschoolers’ language, emergent literacy skills, and use of African American English: A complex relation. Journal of Speech, Language, and Hearing Research. 2006;49:771–792. doi: 10.1044/1092-4388(2006/055). [DOI] [PubMed] [Google Scholar]
  9. Craig HK, Washington JA. Malik goes to school: Examining the language skills of African American students from preschool–5th grade. Lawrence Erlbaum Associates; Mahwah, NJ: 2006. [Google Scholar]
  10. Craig HK, Washington JA, Thompson-Porter C. Average C-unit lengths in the discourse of African American children from low-income, urban homes. Journal of Speech, Language & Hearing Research. 1998;41:433–444. doi: 10.1044/jslhr.4102.433. [DOI] [PubMed] [Google Scholar]
  11. Craig HK, Zhang L, Hensel SL, Quinn EJ. African American English-speaking students: An examination of the relationship between dialect shifting and reading outcomes. Journal of Speech, Language, and Hearing Research. 2009;52:839–855. doi: 10.1044/1092-4388(2009/08-0056). [DOI] [PubMed] [Google Scholar]
  12. Curenton SM. The association between narratives and theory of mind for low-income preschoolers. Early Education and Development. 2004;15:121–145. [Google Scholar]
  13. Curenton SM, Justice LM. African American and Caucasian preschoolers’ use of decontextualized language: Literate language features in oral narratives. Language, Speech, and Hearing Services in Schools. 2004;35:240–253. doi: 10.1044/0161-1461(2004/023). [DOI] [PubMed] [Google Scholar]
  14. Curenton SM. Oral storytelling: A cultural art that promotes school readiness. YC: Young Children. 2006;61:78–89. [Google Scholar]
  15. Dunn LM, Dunn DM. Peabody picture vocabulary test, 4th ed. [norm-referenced test] Pearson; 19500 Bulverde Road, San Antonio, TX: 2007. San Antonio: Pearson. [Google Scholar]
  16. Fey ME, Catts HW, Proctor-Williams K, Tomblin JB, Zhang X. Oral and written story composition skills of children with language impairment. Journal of Speech, Language, and Hearing Research. 2004;47:1301–1318. doi: 10.1044/1092-4388(2004/098). [DOI] [PubMed] [Google Scholar]
  17. Gorman BK, Fiestas CE, Peña ED, Clark MR. Creative and stylistic devices employed by children during a storybook narrative task: A cross-cultural study. Language, Speech, and Hearing Services in Schools. 2011;42:167–181. doi: 10.1044/0161-1461(2010/10-0052). [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Greenhalgh KS, Strong JS. Literate language features in spoken narratives of children with typical language and children with language impairments. Language, Speech, and Hearing Services in Schools. 2001;32:114–125. doi: 10.1044/0161-1461(2001/010). [DOI] [PubMed] [Google Scholar]
  19. Heath SB. What no bedtime story means: Narrative skills at home and school. Language in Society. 1982;11:49–76. [Google Scholar]
  20. Heath SB. Ways with words: Language, life, and work in communities and classrooms. Cambridge University Press; New York: 1983. [Google Scholar]
  21. Hester EJ. Narrative correlates of reading comprehension in African American children. Contemporary Issues in Communication Sciences and Disorders. 2010;37:73–85. [Google Scholar]
  22. Horton-Ikard R. Cohesive adequacy in the narrative samples of school-age children who use African American English. Language, Speech, and Hearing Services in Schools. 2009;40:393–402. doi: 10.1044/0161-1461(2009/07-0070). [DOI] [PubMed] [Google Scholar]
  23. Hughes DL, McGillivray L, Schmidek M. Guide to narrative language: Procedures for assessment. Thinking Publications; Eau Claire, WI: 1997. [Google Scholar]
  24. Hudson J, Shapiro L. From knowing to telling: The development of children's scripts, stories, and personal narratives. In: McCabe A, Peterson C, editors. Developing narrative structure. Lawrence Erlbaum Associates; Hillsdale, NJ: 1991. pp. 89–136. [Google Scholar]
  25. Ivy LJ, Masterson J. A comparison of oral and written English styles in African American students at different stages of writing development. Language, Speech, and Hearing Services in Schools. 2011;42:31–40. doi: 10.1044/0161-1461(2010/09-0069). [DOI] [PubMed] [Google Scholar]
  26. Loban W. Language development: Kindergarten through grade twelve. National Council of Teachers of English; Urbana, IL: 1976. [Google Scholar]
  27. Mainess KJ, Champion TB, McCabe A. Telling the unknown story complex and explicit narration by African American preadolescents: Preliminary examination of gender and socioeconomic issues. Linguistics and Education. 2002;13:151–173. [Google Scholar]
  28. Mayer M. Frog, where are you? Dial Books for Young Readers; New York: 1969. [Google Scholar]
  29. McCabe A, Bliss L, Barra G, Bennett M. Comparison of personal versus fictional narratives of children with language impairment. American Journal of Speech-Language Pathology. 2008;17:194–206. doi: 10.1044/1058-0360(2008/019). [DOI] [PubMed] [Google Scholar]
  30. Michaels S. “Sharing time”: Children's narrative styles and differential access to literacy. Language in Society. 1981;10:423–442. [Google Scholar]
  31. Miller J, Iglesias A. Systematic analysis of language transcripts (SALT) (Research Version ed.) [software] SALT Software, LLC.; Madison, WI: 2010. [Google Scholar]
  32. National Governors Association Center for Best Practices, Council of Chief State School Officers . Common core state standards. National Governors Association Center for Best Practices, Council of Chief State School Officers; Washington, DC: 2010. [Google Scholar]
  33. Peterson C, McCabe A. Developmental psycholinguistics: Three ways of looking at a child's narrative. Plenum Press; New York: 1983. [Google Scholar]
  34. Price JR, Roberts JE, Jackson SC. Structural development of the fictional narratives of African American preschoolers. Language, Speech, and Hearing Services in Schools. 2006;37:178–190. doi: 10.1044/0161-1461(2006/020). [DOI] [PubMed] [Google Scholar]
  35. Ross SH, Oetting JB, Stapleton B. Preterite had plus V- ed: A developmental narrative structure of African American English. American Speech. 2004;79:167–193. [Google Scholar]
  36. Seymour HN, Roeper TW, de Villiers J, de Villiers PA. Diagnostic evaluation of language variation-screening test (DELV-S) [norm-referenced test] Pearson; 19500 Bulverde Road, San Antonio, TX: 2003. San Antonio. [Google Scholar]
  37. Shiro M. Genre and evaluation in narrative development. Journal of Child Language. 2003;30:165–195. doi: 10.1017/s0305000902005500. [DOI] [PubMed] [Google Scholar]
  38. Smitherman G. Talkin that talk: Language, culture, and education in African America. Routledge; New York: 2000. [Google Scholar]
  39. Sperry LL, Sperry DE. Early development of narrative skills. Cognitive Development. 1996;11:443–465. [Google Scholar]
  40. Steele SC, Willoughby LM, Mills MT. Learning word meanings during reading: Effects of phonological and semantic cues on children with language impairment. International Journal of Speech-Language Pathology. 2012 doi: 10.3109/17549507.2012.700322. http://www.speechpathologyaustralia.org.au/publications/international-journal-of-speech-language-pathology Advance online publication. [DOI] [PubMed]
  41. Thompson CA, Craig HK, Washington JA. Variable production of African American English across oracy and literacy contexts. Language Speech and Hearing Services in Schools. 2004;35:269–282. doi: 10.1044/0161-1461(2004/025). [DOI] [PubMed] [Google Scholar]
  42. Uccelli P, Páez MM. Narrative and vocabulary development of bilingual children from kindergarten to first grade: Developmental changes and associations among English and Spanish skills. Language, Speech, and Hearing Services in Schools. 2007;38:225–236. doi: 10.1044/0161-1461(2007/024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ukrainetz TA, Justice LM, Kaderavek JN, Eisenberg SL, Gillam RB. The development of expressive elaboration in fictional narratives. Journal of Speech, Language, and Hearing Research. 2005;48:1363–1377. doi: 10.1044/1092-4388(2005/095). [DOI] [PubMed] [Google Scholar]
  44. Ukrainetz TA, Gillam RB. The expressive elaboration of imaginative narratives by children with specific language impairment. Journal of Speech, Language, and Hearing Research. 2009;52:883–898. doi: 10.1044/1092-4388(2009/07-0133). [DOI] [PubMed] [Google Scholar]
  45. Van Keulen JE, DeBose CE, Weddington GT. Speech, language, learning, and the African American child. Allyn and Bacon; Boston: 1998. [Google Scholar]
  46. Weddington G. It's not the language: Alternative explanations of the education gap for African American children. Topics in Language Disorders. 2010;30:48–56. [Google Scholar]
  47. Williams KT. Expressive vocabulary test, 4th ed [norm-referenced test] Pearson; 19500 Bulverde Road, San Antonio, TX: 2007. San Antonio. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix A
NIHMS567492-supplement-Appendix_A.doc (39KB, doc)

RESOURCES