Perceptions of Black Children's Narrative Language: A Mixed-Methods Study
This article has been corrected.
VIEW CORRECTIONPublication: Language, Speech, and Hearing Services in Schools
Volume 52, Number 1
Pages 84-99
Abstract
Purpose
In this mixed-methods study, we address two aims. First, we examine the impact of language variation on the ratings of children's narrative language. Second, we identify participants' ideologies related to narrative language and language variation.
Method
Forty adults listened to and rated six Black second-grade children on the quality of 12 narratives (six fictional, six personal). Adults then completed a quantitative survey and participated in a qualitative interview.
Results
Findings indicated that adults rated students with less variation from mainstream American English (MAE) more highly than students with greater variation from MAE for fictional narratives, but not for personal narratives. Personal narratives tended to be evaluated more favorably by parents than teachers. Black raters tended to assign higher ratings of narrative quality than did White raters. Thematic analysis and conversation analysis of qualitative interviews supported quantitative findings and provided pertinent information about participants' beliefs.
Conclusion
Taken together, quantitative and qualitative results point to a shared language ideology among adult raters of variation from MAE being more acceptable in informal contexts, such as telling a story of personal experience, and less acceptable in more formal contexts, such as narrating a fictional story prompted by a picture sequence.
Black students continue to lag behind their Asian and White peers on standardized tests of math, reading (Vanneman et al., 2009), and writing (Leu et al., 2014; National Center for Education Statistics, 2011). They are, simultaneously, underrepresented in gifted and talented programs (Ford, 2011) and overrepresented in special education programs (Donovan & Cross, 2002; Robinson & Norton, 2019). Although the factors that contribute to these educational disparities are multifaceted, language skills form the bedrock of academic success. In particular, narration—the language of storytelling—predicts academic outcomes: Children who are competent narrators tend to become competent readers (Reese et al., 2010; Suggate et al., 2018) and writers (Griffin et al., 2004).
Ethnographic and discourse analytic research has shown that narrative patterns and preferences vary across racial groups and that this variation has consequences for children in educational contexts. Heath's (1983) multiyear ethnographic study revealed racial differences with respect to the narrative styles into which Black and White children from low-income backgrounds were socialized by family and community. Low-income Black children were encouraged to make their narratives funny and entertaining, whereas low-income White children were discouraged from telling “tall tales” and encouraged to produce factual narratives. White children's narrative practices were more closely aligned with school norms, and this contributed to a smoother home-to-school transition than the Black children experienced. Likewise, Michael's (1981) seminal study of classroom discourse showed that, as early as Grade 1, teachers in the classroom setting preferred listening to White children who produced topic-centered narratives (which are tightly organized around a single event or object by conjunctions such as “and” or “then”) over listening to Black students who produced topic-associating narratives (which are organized around multiple anecdotes by subtle changes in tempo or intonation). Black children who produced topic-associating narratives (also referred to as “performative narratives”) during in-class sharing time were systemically marginalized.
Descriptive studies of personal narratives indicate Black children draw upon a repertoire of diverse narrative styles, including moral centered—teaching the listener a lesson in virtue—as well as topic centered and topic associating (Champion, 1998; Champion et al., 1995; Hyon & Sulzby, 1994). In a qualitative case study, Bloome et al. (2010) identified a focal child among 100 Black preschoolers and kindergartners participating in a classroom-based storytelling project. The case child highlighted a performative style of narration characteristic of young Black children—to elaborate on topics of interest and of high status to the child (a teacher, trip to park, birthday party at Chuck E. Cheese, etc.), even if these topics do not coalesce into a topic-centered plot (Champion et al., 1995; Michaels, 1981). The focal child's performative narrative style was employed to engage her audience of teachers, fellow students, and researchers; she did not simply deliver a factual and sequential retelling of a past event. However, the importance of the performative narrative style was diminished in the classroom setting, which privileged topic-centered narration.
In addition to employing narrative styles that are undervalued in the classroom, school-age Black children may produce narratives in African American English (AAE)—a dialect of American English that, in many classrooms, is not recognized as a legitimate variety of English despite having a systematic and rule-governed grammar, vocabulary, and sound system (Burns et al., 2012; Mills, 2015a; Mills et al., 2013; Mills & Fox, 2016). Although AAE shares linguistic features with mainstream American English (MAE), it also has linguistic features not present in MAE (e.g., cold pronounced as coal, zero possessive, habitual BE). Some scholars in education research use the label African American Language to refer to the phonological, morphological, syntactic, semantic, and to pragmatic patterns of AAE, which includes discursive practices (such as narrative structures) that are used by African Americans and are associated with African American culture (e.g., Bloomquist et al., 2015). In this article, we make a distinction between AAE and narrative structures in order to analyze how they each affect adult perceptions of Black children's narratives.1
In the United States, MAE is the dialect of classroom instruction (Mills & Washington, 2015), which means that children need a strong facility with MAE to succeed academically. For 90% of Black kindergartners who enter school speaking AAE as their primary dialect, formal instruction designed for MAE speakers and implemented in MAE poses distinct challenges (Craig & Washington, 2006). Across grade levels, discontinuities between home and school language can lead to misunderstandings in the classroom as well as misinterpretation and misrepresentation of students' knowledge and skills as assessed by standardized tests (Wheeler et al., 2012) and teacher grading practices (Wheeler, 2019). There is a broad consensus among researchers that insufficient accommodation and acceptance of AAE contribute to education disparities in assessment, access to services, methods of service provision, and, ultimately, educational achievement between MAE speakers and AAE speakers (e.g., Carter, 2010; Hallett, 2015; Stockman, 2010). Educational disparities cannot be fully addressed without understanding how the language of school-age Black children is evaluated informally and formally.
Perceptions of Narrative Language
Although racial differences in narration have been studied since the 1960s, efforts to identify culturally-fair measures of narrative language for AAE speakers are relatively recent. Candidate measures include false-belief mentioning (Mills & Fox, 2016) and rare vocabulary usage (Mills et al., 2017), as they were positively correlated with measures of language productivity and educational placement but not correlated with language variation. Language variation—the extent to which children's language differed from MAE—was measured by the Diagnostic Evaluation of Language Variation–Screening Test (DELV-S; Seymour et al., 2003). These measures stem from researchers and a priori classifications developed by researchers. We can further advance the development of culturally-fair measures of Black children's narrative language by examining and taking into account the perspectives of the people for whom these assessments have real-life significance: teachers and parents of Black children.
Assessments of narratives that allow adult participants to listen to and rate the quality of children's narratives provide a shift toward culturally-fair assessment because such methods elicit and take seriously the perspectives of people who are not researchers. Evaluating narrative language through listener judgment has been undertaken in the literature on healthy adults (Christensen et al., 2009) and in the literature on children with language impairment (Newman & McGregor, 2006). For example, Christensen et al. (2009) asked 12 college students to rate two narratives produced by thirteen 18- to 74-year-old adults: a narrative elicited from a wordless picture and a narrative elicited from a picture sequence. Adult rating was correlated with traditional measures of narrative language, such as number of utterances, number of ideas expressed, and number of main events. Results indicated that visual stimuli used to elicit narratives influenced how narratives were rated. That is, ratings for the wordless book were related to traditional measures of narrative language, whereas ratings for the picture sequence did not load unto traditional narrative measures.
Perceptions of children's narrative language have also been examined. For example, Newman and McGregor (2006) asked 21 teachers and 27 parents (laypersons) to rate the narratives of twenty 5- to 7-year-old children (10 with language impairment, 10 with typical language). Results indicated that ratings from both adult groups tended to be congruent with the child's diagnostic category, such that ratings for typically developing children were higher than ratings for language-impaired children with 70% nonoverlap. However, the two groups of raters differed in that laypersons tended to listen for “sparkle” or the charm of the narratives to a greater extent than did teachers. Nevertheless, both teachers and nonteachers were sensitive to quality differences in narratives that sounded typical versus impaired.
Theoretical Framework
Perceptions of language use have been studied as social and cultural phenomena by linguistic anthropologists, giving rise to concepts and insights that can productively inform efforts to develop culturally-fair assessments of Black children's language. Particularly useful is the concept of language ideologies, “conceptualizations about languages, speakers, and discursive practices” (Irvine, 2016), which range from subconsious assumptions to explicit dogma (Riley, 2011). Language ideologies differ from the concept of language attitudes in that the latter is grounded in quantitative methodologies such as questionnaires or matched-guise experiments (cf. Blake & Cutler, 2003; Salmon, 2015), whereas language ideologies research is grounded in qualitative methodologies and emphasizes how people's beliefs and feelings about language are developed, communicated, and put into action within a socioeconomic and cultural–historical context (Kroskrity, 2018).
Research by scholars of raciolinguistics, a new field that explores how race shapes our ideas about language and how language shapes our ideas about race (Alim et al., 2016), has brought into focus language ideologies that “produce racialized speaking subjects who are constructed as linguistically deviant” (Flores & Rosa, 2015, p. 150). Take, for example, three deep-rooted language ideologies that have a profound effect on educational language policy and practice: the ideology of language standardization—the idea that there is a correct way of using the national language and that all people ought to use it this way; the ideology standardization—the idea that a single shared language is essential to the unity and strength of a nation and that mastery of that language is required for full citizenship; and the ideology of dualism (Farr & Song, 2011). The ideology of dualism, also referred to as “the great divide,” holds that orality and literacy are two separate entities, with literacy receiving a higher value than orality (Bloome et al., 2010; Collins & Blot, 2003). Orality is conceptualized as concrete and context bound and is associated with simplicity, whereas literacy is conceptualized as abstract and context free and is associated with modernity and progress.
Subsequent to an ideology of dualism, educational standards require that children speak in a literate fashion, rendering teachers ineffective in supporting (narrative) language that is aligned with orality. For example, Blake and Cutler (2003) administered questionnaires to high school teachers in New York to study teacher attitudes toward language variation. They found that “teachers' language attitudes appear to be influenced by the philosophies, or lack thereof, of the schools in which they teach” (p. 186). Moreover, results indicated that teacher coursework in linguistics seemed to be an important factor in teacher attitudes toward AAE, as was the ethnic composition of the student body where teachers worked. That is to say, their study called attention to the important role of context and experience in shaping of teachers' beliefs about AAE and language variation. Blake and Cutler also found that teachers are likely to be “circumspect” in their expression of attitudes (p. 188), a reminder that researchers need to use multiple elicitation methods if we are to gain access to research participants' beliefs and feelings about language variation, overt or covert, implicit or explicit. Taking language ideology and raciolinguistics as a theoretical framework helps us critically examine beliefs about what counts as narrative skill; how those beliefs are expressed and enacted in educational practice; and how those beliefs and practices position some children, their communities, and their language as deficient (Flores & Rosa, 2015).
The Current Study
The current study offers two methodological advances over prior studies of Black children's narrative abilities. The first is the use of a mixed-methods design that allows us to examine adults' evaluations of Black children's narrative language quantitatively and qualitatively. The second advance is that raters listened to two different types of narrative per child. This design is important because (a) repeated measures increase power to detect statistically significant differences and (b) prior research indicated that the visual stimuli used to elicit narratives (e.g., wordless book vs. five-picture sequence) influenced how narratives were rated (Christensen et al., 2009). Hence, raters listened to two narratives elicited under differing visual conditions (no visual = personal narrative, visual = five-picture sequence narrative). We chose to include narratives elicited from no visual and from a five-picture sequence for efficiency of listening because the two conditions yielded narratives of relatively short length compared to a wordless book (Mills, 2015b; Mills & Fox, 2016), reducing the listening time required of each adult rater.
The purpose of the study is twofold: (a) to determine the impact of language variation and narrative type on adult rating of the quality of Black children's narrative language and (b) to identify ideologies related to narration and language variation that underlie adult rating of Black children's narrative language or may not be reflected therein. Our working hypothesis was that adult rating would vary based on the following: language variation of narrator, narrative type (Christensen et al., 2009), and rater characteristics (Blake & Cutler, 2003). We also expected qualitative data to align with quantitative findings.
Method
Approach
We employed a concurrent triangulation mixed-methods design, which aims to confirm, cross-validate, or corroborate findings within a single study (Creswell et al., 2003). Concurrent triangulation offsets weaknesses that are inherent in each method. In the current study, we corroborated quantitative findings from narrative rating and a survey with qualitative findings from interviews. Quantitative and qualitative data were gathered in a single session, analyzed separately, and integrated in our discussion of the research findings.
The rationale for the mixed-methods approach employed in this study rests on its potential to unearth new insights and novel indices of narrative quality by using methods of data collection and analysis that, by nature of how the different methods are structured, allow participants to express their perceptions and perspectives in diverse ways. The numerical ratings of narrative quality and the survey present participants with a priori dimensions or categories that were identified by researchers as relevant on the basis on prior research. These methods for measuring participants' responses generated standardized, numerical forms of data, which we analyzed using statistical methods. Like the quantitative methods, the semistructured interviews were grounded in prior research, but they were designed and implemented so as to create a space for participants to express beliefs and feelings that was not provided by the highly structured (and thus constraining) rating and survey tasks. Interview data were analyzed using thematic analysis and conversation analysis (CA) in order to identify patterns in participants' talk about the children's narrative language and its variation, more broadly.
Sampling Context and Participants
Convenience and snowball sampling were used to recruit parents and teachers of second- and third-grade students. We aimed to recruit teachers and parents from schools that were near the district mean for socioeconomic status, and we recruited participants from local schools in two different school districts and in two local charter schools. After we received approval from the institutional review board at The Ohio State University2 and school district authorities, we first contacted principals via phone and e-mail to obtain their permission to contact teachers at their school. We then reached out to teachers who had been recommended by their principals or had contacted us after receiving information about the study from their principals. District staff members with whom we already had relationships helped us identify potential school sites but did not recruit participants on our behalf. With principals' permission and teachers' assent, the research team sent a recruitment flier home with second-grade students and attended several school events in order to recruit parents. After participants completed the rating, survey, and interview, we then asked them to give a recruitment flier to acquaintances who fit our inclusionary criteria, if they felt comfortable doing so. Recruitment slowed after we had collected data from more than half of the total participants. Once we received institutional review board and district approval, we began recruiting online. We posted the recruitment flier on Facebook pages: two pages targeting parents of school-age children and three pages targeting teachers, including a district official teacher association page.
A total of 40 adult raters participated in the study. Adult raters were grouped by role: teacher versus parent. Among 40 raters, 20 were teachers of second- or third-grade students. Teachers held at least 1 year of teaching experience (M = 3.8 years, range: 1–5 years). Twenty parents of second- or third-grade students participated in the study. Among all adult raters, 20 were Black and 20 were White. All data collection took place in either teachers' classrooms or in a quiet study room at a public library.
Tasks
Listener Rating
Adults listened to audio samples of six second-grade monolingual Black children from central Ohio telling two narratives: (a) a fictional narrative elicited from a five-picture sequence from the Test of Narrative Language (Gillam & Pearson, 2004) and (b) a personal narrative elicited from a model narrative (see Appendix A). Critically, child narrators were selected who represented different levels of language variations, per classifications from Part I of the DELV-S (Seymour et al., 2003). As shown in Table 1, two children had no variation from MAE (MAE speakers), two children had some variation from MAE (bidialectal speakers), and two children had strong variation from MAE (AAE speakers). Stimuli were presented electronically using the E-Prime 3.0 software (Psychology Software Tools, Inc., 2016).
| Participants | Gender | Age | DELV | TONI | TNL | PPVT-4 | WIAT-III |
|---|---|---|---|---|---|---|---|
| 1 | Male | 7;3 | 2 | 96 | 112 | 97 | 98 |
| 2 | Female | 8;3 | 1 | 95 | 91 | 103 | 84 |
| 3 | Male | 7;8 | 0 | 95 | 103 | 99 | 119 |
| 4 | Male | 8;2 | 2 | 104 | 85 | 89 | 98 |
| 5 | Male | 7;7 | 1 | 110 | 106 | 101 | 96 |
| 6 | Female | 7;8 | 0 | 106 | 109 | 120 | 112 |
| Mdn | 7;8 | 1 | 100 | 104.5 | 100 | 98 | |
| M | 7;8 | 1 | 101 | 101 | 102 | 101 | |
| SD | 4.27 | 0.89 | 6.51 | 10.68 | 10.27 | 12.46 | |
| Min | 7;3 | 0 | 95 | 85 | 89 | 84 | |
| Max | 8;3 | 2 | 110 | 112 | 120 | 119 |
Note. Mean performance on test battery including criterion scores on the Diagnostic Evaluation of Language Variation (DELV) Screening Test and standard scores on the following: Test of Narrative Language (TNL), Test of Nonverbal Intelligence (TONI), Peabody Picture Vocabulary Test–Fourth Edition (PPVT-4), and Wechsler Individual Achievement Test–Third Edition (WIAT-III). DELV criterion scores are classified as follows: 0 = no variation from mainstream American English (MAE), 1 = some variation from MAE, 2 = strong variation from MAE. Standard scores are based on a mean of 100 and an SD of 15. Age is presented in years;months.
Child narrators had participated in a larger study examining the correlates of narrative language in school-age Black children (Mahurin-Smith et al., 2021, this issue; Mills, 2015b; Mills & Fox, 2016). Accordingly, children were assessed and performed within normal limits in the following areas: cognition (Test of Nonverbal Intelligence–Fourth Edition; Brown et al., 2012), diagnostic risk for language impairment (DELV-S), vocabulary (Peabody Picture Vocabulary Test–Fourth Edition; Dunn & Dunn, 2007), literacy (Wechsler Individual Achievement Test–Third Edition; Wechsler, 2009), and narration (Test of Narrative Language). Table 1 shows how child narrators in the current study performed on the testing battery.
Adult raters were seen individually in either a classroom or quiet, private rooms reserved at local public libraries.Raters were trained by hearing a spoken narrative presented and rating it before continuing to rate the study narratives. See Appendix B for the instructions provided to participants.
Before beginning, the sound level was tested by playing an audio clip of a child narrating from a laptop; participants were asked to indicate if the level of volume was at an acceptable level and adjusted accordingly. Participants assigned a number between 1 (poor) and 7 (excellent) for each of the 12 prerecorded stories by moving the mouse or sliding their finger along the laptop screen. To establish intrarater agreement, E-Prime randomly selected one narrative to repeat: intraclass correlation coefficient for both absolute and relative (consistency) agreement. Intraclass correlation coefficient indicated that consistency of test–retest rating reached 66.5% and that the agreement of test–retest rating reached 66.8%. Adults tended to increase ratings after hearing narratives for the second time.
Survey
Following narrative rating, participants answered a 20-item survey regarding the aspects of narration that influenced their ratings (e.g., I listened to see if the details included were relevant to the story; I listened for the use of correct grammar) presented in E-Prime 3.0 from a laptop. Participants rated the level of influence on a Likert scale from 1 = no influence to 5 = strong influence. This survey has been used in prior studies examining listener perceptions of narrative language and is displayed in Appendix C (Christensen et al., 2009; Newman & McGregor, 2006).
Qualitative Interview
Following the survey, participants completed individual qualitative interviews that were designed to elicit information about participants' perspectives on and experiences with different varieties of English and explore their responses to the narratives they had just listened to and rated. The interviews were conducted by the fourth author, a multilingual (Korean, English, Spanish) doctoral candidate in education who had come from South Korea to the United States for graduate study. Interviews lasted from 11:44 (min:s) to 36:53 (M = 20:48, SD = 0.06). See Appendix D for interview instructions and questions.
The interviewer positioned herself as a nonnative speaker of English who was relatively new to the United States. This position enabled her to ask questions about English language variation as someone who was neither Black nor White and who was not likely to know or feel much about different varieties of English spoken in the United States. The interviewer introduced some basic terms used by linguistics to talk about different varieties of English to establish some shared vocabulary and to frame that vocabulary as technical and thus “neutral.”
As is standard practice in qualitative research methodologies, the interviewer asked a set of preplanned questions, but she also followed the lead of each participant, posing unplanned additional questions that built upon participant responses, thereby encouraging participants to discuss views and experiences that were significant to them but might not have emerged in a more tightly structured interview. The interview was organized such that different varieties of English were discussed before the children's stories because we wanted participants to have language variation in mind when the interviewer asked them about the stories they had just listened to.
We used the Systematic Analysis of Language Transcripts (SALT; Miller & Iglesias, 2012) for the initial transcription of interview data. Research assistants in communication sciences and disorders, linguistics, and psychology segmented the narratives into communication units using Loban's (1976) scoring criteria. Communication units are independent clauses plus their modifiers, including one main clause along with accompanying subordinate clauses. SALT transcripts were used for thematic analysis. For finer grained analysis of interview interactions, excerpts of the interviews were transcribed following the conventions of CA (Hepburn & Bolden, 2013).
Data Analysis
Quantitative Analysis
Quantitative data were analyzed by the first and third authors. Our analysis examined the following: (a) adult rating of children's narrative quality, (b) adult surveyed listening influences, and (c) the factors that might account for adult rating.
To address our first and third research questions, a two-way analysis of variance (ANOVA) model was built to determine whether adult rating differed on the basis of children's narrative language variation and narrative type. In addition, role and race were added to the model to identify differences in ratings. An alpha level of < .5 was set. Eta squared measured effect size of significant differences. Effect sizes of the ANOVA tests were characterized as small (.01), medium (.09), or large (.25).
To address our second question regarding adult surveyed listening preferences, we calculated that descriptive statistics of all 20 items using the R package psych (Revelle, 2014). Cronbach's alpha and McDonald's omega were estimated to confirm the reliability of the surveyed items. In recent years, omega has been suggested as a readily available alternative for item reliability and validity, as Cronbach's alpha may overestimate reliability due to its assumption of unidimensionality (Dunn et al., 2014). According to Lance et al. (2006), commonly, an alpha value of .70 is acceptable for measurement reliability. The Cronbach's alpha for the surveyed items was .84; a well-acceptable reliability of the items demonstrating adults' responses to their rating preference were consistent. All the items as a whole produced the McDonald's omega value of .89, showing a very strong internal consistency and content validity of the items in measuring adults' rating behaviors.
Using a Shapiro–Wilk normality test, we found that adult responses to the survey items were not normally distributed. Therefore, we used the nonparametric Mann–Whitney U test to examine differences in adult listening preferences on the bases of role and race. An alpha level of < .5 was set. Cohen's d measured effect size of significant differences. Effect sizes of the Mann–Whitney U tests were characterized as small (.20), medium (.50), or large (.80).
Qualitative Analysis
Interview data were analyzed qualitatively by the second and fourth authors, employing macro and micro traditions (cf. Simmons-Mackie, 2014). Thematic analysis (a macro tradition) was used to identify patterns in the interviews in terms of “what” participants said, whereas CA (a micro tradition) was used to identify patterns in terms of “how” participants said what they said. Both approaches provide insights into research participants' perspectives associated with particular phenomena or groups, and both have been used previously in communicative sciences and disorders research (Anderson & Felsenfeld, 2003; Morgan et al., 2019; Samuelsson & Plejert, 2014; Simmons-Mackie, 2014; Wilkinson, 2012).
Thematic analysis was done in five phases as follows: familiarization with the data, first-level coding, second-level coding, searching for themes, and reviewing and defining themes (cf. Nowell et al., 2017). In the first, exploratory phase, we listened to the audio recordings and followed along on the transcripts in order to become familiar with the interview data, making analytic memos but not assigning codes. In the second phase, we took an open approach, meaning we did not start with a coding framework. However, we did have some prior conceptions about patterns we might find based on published research, our research questions, and our initial review of the data. We generated and assigned initial codes that described the content of segments of participants' speech (first-level coding). For example, the utterance “Oh right so like southern dialects” was assigned the code “Refers to Southern dialect.” In the third phase, we created categories and subcategories by grouping descriptive codes that were similar in order to get a sense of patterns in the data and move to a more abstract level of coding (second-level coding or pattern coding). Thus, the codes “Refers to Southern dialect,” “Refers to Midwest accent,” and “Refers to British English” were grouped under the subcategory “Reference to regional variety/variation.” “Reference to regional variety/variation” was clustered with “Reference to socioeconomic status” and “Reference to ethnic/racial variety/variation” under “Reference to language variation.” These categories are what Braun et al. (2018) call domain summaries—summaries of “what participants said in relation to a topic or issue” (p. 4). In the fourth phase, we developed themes, reviewing categories, codes, and underlying data to identify broad ideas that illuminated large portions of the data set, what Braun et al. call shared meaning-based patterns (examples of which are discussed in the Findings section). During the fifth phase, we shifted from using SALT to using Transana Professional (Transana 3.32d, 2020), a qualitative data analysis software that facilitated the re-examination of transcripts, codes, and categories and the connections we had made among them that was necessary to refine and define themes (cf. Silver & Lewins, 2017).
The tools of CA were used to identify patterns in how participants structured their talk in the interview, the social actions they achieved by using these structures, and what the structures reveal about their stances with regard to narration, different varieties of English and speakers thereof, and the Black children's narratives they had just heard. To use CA means to focus “on the social conventions and practices which participants in an interaction draw upon in order to produce talk and other conduct in interaction which is treated by recipients as coherent and meaningful” (Wilkinson, 2012, p. 963). This micro-analytic approach to human communicative interaction is grounded in anthropological and sociological theories of language use as “a key locus of social life and an observable instantiation of culture and social organization” (Simmons-Mackie, 2014, p. 24). Conversation analysts typically focus on naturally occurring talk, but CA is also used to analyze research interviews (particularly open-ended qualitative interviews, such as those conducted for this study) and has been shown to be an effective tool for providing insights into participants' perspectives by making visible how they co-construct their accounts, descriptions, and assessments that are pertinent to the research topic (Roulston, 2006). Taking a CA approach to the interview data complemented the thematic analysis by obliging us to consider the larger course of action to which particular turns at talk or exchanges contributed, as opposed to seeing small units of talk only in isolation as instances of a specific code (Bolden, 2015). In a final, confirmatory phase of qualitative analysis, we took a concept-driven approach, looking at the thematic analysis and conversation analytic findings data after the quantitative analyses were complete in order to determine if our qualitative findings confirmed, complicated, and/or contradicted findings from the rating and survey data.
Results
Quantitative Findings
Adult Rating
We conducted a two-way ANOVA to determine whether adult rating differed on the basis of children's narrative language variation (no, some, or strong variation from MAE) and narrative type (personal, fictional). We found significant main effects for both language variation, F(2, 554) = 4.358, p = .013 < .05, η2 = .013, and narrative type, F(1, 554) = 99.375, p < .001, η2 = .145. In addition, the two variables yielded a statistically significant interaction effects on adult rating, F(2, 554) = 11.900, p < .001, η2 = .035: The effect of narrative type on adult rating was relatively stronger when children had less variation from MAE, as depicted in Table 2 and Figure 1. The effect size was small for this Language Variation × Narrative Type effect. Moreover, adults offered higher ratings on narratives produced in response to picture sequence than on narratives produced from no visual stimulus.
| Language variation | Narrative type | Parent rating (n = 20) | Teacher rating (n = 20) | Rating (n = 40) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Range | M | SD | Range | M | SD | Range | M | SD | ||
| MAE | Fictional | 1.0–7.0 | 4.26 | 1.197 | 1.5–6.5 | 3.59 | 1.339 | 1.0–7.0 | 3.93 | 1.304 |
| Personal | 1.0–7.0 | 5.8 | 1.352 | 1.0–7.0 | 5.83 | 1.273 | 1.0–7.0 | 5.81 | 1.306 | |
| Bidialectal | Fictional | 2.0–7.0 | 4.51 | 1.218 | 1.0–6.5 | 3.7 | 1.400 | 1.0–7.0 | 4.09 | 1.371 |
| Personal | 1.0–7.0 | 4.74 | 1.445 | 2.0–7.0 | 4.99 | 1.358 | 1.0–7.0 | 4.87 | 1.398 | |
| AAE | Fictional | 2.5–7.0 | 4.54 | 1.202 | 1.0–6.0 | 4.21 | 1.199 | 1.0–7.0 | 4.38 | 1.206 |
| Personal | 2.5–7.0 | 4.98 | 1.233 | 1.0–7.0 | 5.12 | 1.531 | 1.0–7.0 | 5.05 | 1.381 | |
Note. Language variation status was based on classifications from Part I of the Diagnostic Evaluation of Language Variation Screening Test as follows: MAE = no variation from MAE, bidialectal = some variation from MAE, and AAE = strong variation from MAE. Fictional narratives were prompted by a picture sequence. Personal narratives were prompted by a model narrative. MAE = mainstream American English; AAE = African American English.
As aforementioned, we added role and race to the model to examine differences on adult rating. Role did not show individual effect on the ratings, yet it interacted with narrative type to yield statistically significant effects on the ratings, F(1, 532) = 11.030, p < .001, η2 = .016. The effect size was medium for this Role × Narrative Type effect on narrative rating. As depicted in Figure 2, adults across the two groups did not differ on their ratings of fictional narratives; however, parents tended to score personal narratives more favorably than did teachers. With regard to race, we found a statistically significant main effect on narrative rating, F(1, 532) = 6.043, p = .014 < .05, η2 = .005, and its effect size was small. As shown in Table 3, Black adults provided higher ratings than did White adults. Race showed no interaction effects with narrative type nor with language variation.
| Roles | Race | n | Rating |
|---|---|---|---|
| Teachers | Black | 10 | 4.71 |
| White | 10 | 4.45 | |
| Total | 20 | 4.58 | |
| Parents | Black | 10 | 4.89 |
| White | 10 | 4.73 | |
| Total | 20 | 4.81 |
Surveyed Listening Influences
We present descriptive statistics for each survey item in Tables 4 and 5. Adults reported that their listening to students' narrative was most influenced by information found in the following survey items: 1 (I listened for the use of specific [vs. general] vocabulary words), 3 (I listened to see if the details included were relevant to the story), 4 (I listened for the amount of detail that the child included), 6 (I listened to see if all the critical parts of the story were included), 7 (I listened for how well the child's thoughts flowed together), and 8 (I listened to see if the child followed the theme of the story). Conversely, adults reported that their listening to students' narrative was least influenced by information found in Items 16 (I listened for the presence of hesitations, pauses, and/or the use of words like um or uh), 17 (I listened for how funny the child was), and 18 (I listened to how cute the child sounded).
| Survey items | Teacher responses | Parent responses | p | Cohen's d | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | M | SD | Range | n | M | SD | Range | |||
| 1 | 20 | 4.15 | 0.75 | 3–5 | 20 | 3.95 | 0.83 | 2–5 | .483 | 0.11 |
| 2 | 20 | 3.25 | 1.07 | 1–5 | 20 | 3.70 | 1.30 | 1–5 | .160 | 0.22 |
| 3 | 20 | 4.65 | 0.59 | 3–5 | 18 | 4.61 | 0.61 | 3–5 | .843 | 0.03 |
| 4 | 20 | 4.90 | 0.31 | 4–5 | 19 | 4.95 | 0.23 | 4–5 | .605 | 0.09 |
| 5 | 20 | 3.25 | 0.91 | 2–5 | 20 | 3.05 | 1.19 | 1–5 | .723 | 0.06 |
| 6 | 20 | 4.70 | 0.57 | 3–5 | 19 | 4.26 | 0.99 | 2–5 | .128 | 0.24 |
| 7 | 19 | 4.32 | 0.67 | 3–5 | 20 | 4.50 | 0.69 | 3–5 | .34 | 0.15 |
| 8 | 20 | 4.30 | 1.03 | 1–5 | 20 | 4.40 | 0.88 | 2–5 | .796 | 0.04 |
| 9 | 20 | 2.70 | 1.03 | 1–4 | 20 | 3.35 | 1.14 | 1–5 | .094 | 0.27 |
| 10 | 20 | 3.10 | 1.17 | 1–5 | 20 | 3.55 | 0.94 | 1–5 | .226 | 0.19 |
| 11 | 20 | 3.40 | 0.88 | 2–5 | 20 | 3.70 | 0.98 | 2–5 | .341 | 0.15 |
| 12 | 20 | 3.20 | 1.06 | 1–5 | 20 | 2.90 | 0.97 | 1–4 | .312 | 0.16 |
| 13 | 20 | 2.80 | 1.24 | 1–5 | 20 | 3.25 | 1.25 | 1–5 | .298 | 0.17 |
| 14 | 19 | 2.63 | 1.12 | 1–4 | 20 | 3.65 | 0.93 | 2–5 | .009** | 0.42 |
| 15 | 20 | 3.80 | 1.06 | 2–5 | 20 | 3.95 | 0.83 | 2–5 | .753 | 0.05 |
| 16 | 20 | 2.70 | 1.22 | 1–5 | 20 | 3.15 | 1.09 | 1–5 | .275 | 0.17 |
| 17 | 20 | 2.20 | 1.40 | 1–5 | 20 | 3.10 | 1.41 | 1–5 | .048* | 0.31 |
| 18 | 20 | 1.55 | 0.94 | 1–4 | 20 | 2.20 | 1.24 | 1–4 | .060 | 0.30 |
| 19 | 19 | 3.21 | 1.23 | 1–5 | 19 | 3.63 | 1.26 | 1–5 | .291 | 0.17 |
| 20 | 20 | 3.85 | 0.75 | 3–5 | 20 | 4.00 | 0.97 | 1–5 | .336 | 0.15 |
*
p < .05.
**
p < .01.
| Survey items | White | Black | p | Cohen's d | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | M | SD | range | n | M | SD | range | |||
| 1 | 20 | 4.05 | 0.69 | 3–5 | 20 | 4.05 | 0.89 | 2–5 | .861 | 0.03 |
| 2 | 20 | 3.35 | 1.14 | 1–5 | 20 | 3.60 | 1.27 | 1–5 | .407 | 0.13 |
| 3 | 19 | 4.53 | 0.61 | 3–5 | 19 | 4.74 | 0.56 | 3–5 | .203 | 0.20 |
| 4 | 20 | 4.90 | 0.31 | 4–5 | 19 | 4.95 | 0.23 | 4–5 | .605 | 0.09 |
| 5 | 20 | 3.10 | 1.07 | 1–5 | 20 | 3.20 | 1.06 | 1–5 | .755 | 0.05 |
| 6 | 19 | 4.37 | 0.83 | 2–5 | 20 | 4.60 | 0.82 | 2–5 | .208 | 0.20 |
| 7 | 20 | 4.45 | 0.60 | 3–5 | 19 | 4.37 | 0.76 | 3–5 | .876 | 0.03 |
| 8 | 20 | 4.10 | 1.17 | 1–5 | 20 | 4.60 | 0.60 | 3–5 | .190 | 0.21 |
| 9 | 20 | 2.95 | 0.94 | 1–5 | 20 | 3.10 | 1.29 | 1–5 | .613 | 0.08 |
| 10 | 20 | 3.35 | 1.09 | 1–5 | 20 | 3.30 | 1.08 | 1–5 | .708 | 0.06 |
| 11 | 20 | 3.30 | 0.86 | 2–5 | 20 | 3.80 | 0.95 | 2–5 | .128 | 0.24 |
| 12 | 20 | 2.95 | 1.00 | 1–4 | 20 | 3.15 | 1.04 | 1–5 | .549 | 0.10 |
| 13 | 20 | 2.85 | 1.14 | 1–5 | 20 | 3.20 | 1.36 | 1–5 | .374 | 0.14 |
| 14 | 20 | 3.00 | 1.17 | 1–5 | 19 | 3.32 | 1.11 | 1–5 | .493 | 0.11 |
| 15 | 20 | 3.60 | 0.94 | 2–5 | 20 | 4.15 | 0.88 | 2–5 | .061 | 0.30 |
| 16 | 20 | 2.75 | 1.07 | 1–4 | 20 | 3.10 | 1.25 | 1–5 | .459 | 0.12 |
| 17 | 20 | 2.80 | 1.32 | 1–5 | 20 | 2.50 | 1.61 | 1–5 | .510 | 0.11 |
| 18 | 20 | 1.65 | 0.93 | 1–4 | 20 | 2.10 | 1.29 | 1–4 | .311 | 0.16 |
| 19 | 19 | 3.26 | 1.33 | 1–5 | 19 | 3.58 | 1.17 | 1–5 | .469 | 0.12 |
| 20 | 20 | 3.90 | 0.64 | 3–5 | 20 | 3.95 | 1.05 | 1–5 | .590 | 0.09 |
Results of item response differences due to role and race are reported in Tables 4 and 5, respectively. We found a main effect for role such that parents were more influenced than were teacher by information found in Item 14 (I listened for how well the child produced their speech sounds) and Item 17 (I listened for how funny the child was). We found no race-based differences in adult listening influence.
Qualitative Findings
We focus here on the three major themes that we identified in the interview data: Reluctance to talk about language variation as linked to Blackness, Conditional acceptance of variation from MAE in young children's (narrative) language, and Emphasis on sequencing and details in narratives. In our discussion of these themes, we present interview excerpts to illustrate these shared meaning-based patterns in what participants said and how they said it. We also use interview excerpts to draw attention to some interesting differences between the perspectives expressed by Black and White participants. We present excerpts from interview data, allowing our readers to “hear” the participants' voices, to see how they are co-constructing meaning with the interviewer.
Reluctance to Talk About Language Variation as Linked to Blackness
We found that none of the participants were quick to talk about English language variation linked to race/ethnicity.3 They spoke readily about varieties of English linked to generation, place (setting, city, region, nation), or people who learned English as an additional language. All but one participant addressed varieties linked to race or ethnicity, and only when asked to by the interviewer. In several cases, White participants had to be prompted to respond to the interviewer's question, and some of the participants quickly turned the conversation back to other kinds of English language variety. Across the sequences in which the interviewer pursues information about the participant's experience with language variation that is connected to Blackness, there was a consistent pattern of participants doing much more interactional work than in other question–answer sequences. Participants used several practices that signified unease in constructing their turn at talk and/or allowed them to mitigate or delay giving their answer to the interviewer's question: pauses, disfluencies, hesitation markers (e.g., “um,” “like,” sound stretches), hedges (linguistic devices that reduce the force of an utterance, such as “I think,” “I guess”), and discourse markers that signal that the response will be nonstraightforward (e.g., “well”).
Conversation analytic transcripts are highly detailed, providing information about such features as pauses, intonation, laughter, overlapping talk, and cutoff speech. This level of detail can make it difficult for the uninitiated to understand the transcript. The transcripts presented here have been simplified to make them more accessible. Transcripts 1 and 2 are from White teachers, and Transcripts 3 and 4 are from Black teachers. Transcript 1 illustrates the patterns in what participants said and how they said it that underlie our first theme.
Transcript 1. “I don't know how to say it.” R indicates researcher/interviewer, T indicates teacher/interviewee, (0.3) indicates silence, Um:: indicates sound stretch, parentheses indicate uncertain hearing.
| R: | Do you have any experience with the children who speak a variety of the | |
| English different from yours? | ||
| T: | Um::: | |
| R: | It can be one of your students or one of your previous students, | |
| T: | I think that well I mean when I like when I moved down South | |
| um obviously (they) (there are things) there are words that they have for | ||
| the same thing that we have here but different words. | ||
| R: | Mmhmm? | |
| T: | So um that was happening all the time. And I had to get used to that. | |
| Um:: so I feel like here there's a little bit of um | ||
| (I don't) I don't know how to say it. (It like) The word would be (like) | ||
| Ebonics (I guess). | ||
| R. | Mhm. | |
| T: | And I don't know if you're familiar with that word. | |
| R: | No. | |
| T: | So um (0.3) I don't know how to explain it. | |
| R: | ((laughs)) | |
| T: | So:: it would be (like) that's sometimes a term that people use (like for) | |
| um for the group that would be (like) African American or Black per se. | ||
| Um kind of like a street slang or street term to words. | ||
| And so (um) I feel like these kids here use that a lot. | ||
| (Like) they shorten their words or (they don't) they may not (like) finish | ||
| their sentences. |
Transcript 1 begins after the teacher has talked for several turns about regional language variation in the United States, particularly her experiences in the South. It exemplifies adherence to ideologies of language standardization and monolingualism.
Conditional Acceptance of Variation From MAE in Young Children's (Narrative) Language
When asked if they paid attention to a child's accent, vocabulary, and/or sentence structure, nearly all participants reported that that they did not or tried not to allow these aspects of the stories have much or any influence on their ratings. Participants from all four groups expressed the view that the children whose stories they had listened to were too young for variation from MAE to be a consideration in their ratings. Transcript 2 is one example of how White participants' talk on this topic revealed adherence to the ideology that there is a standard, proper, or correct way to use English.
Transcript 2. “Something that is like not grammatically correct.” [indicates overlapping speech.
| T: | I definitely [listened for it. | |
| R: | [Anyways | |
| T: | (I mean) I definitely listened for it so if someone says something that is | |
| like not grammatically correct or something like that then I definitely am | ||
| like wait what? | ||
| R: | ((laughs)) | |
| T: | Um:: but (I don't think that it) especially because they were little people so | |
| I wasn't really (I mean) I didn't really listen for that. | ||
| R: | Okay, | |
| T: | I heard it but (it) I didn't and when I was scoring. | |
| R: | Take it into consideration. | |
| T: | I didn't score a lower score because they may have pronounced something | |
| the wrong way or um based on that. |
When asked if they ever corrected the speech of their students and/or children, nearly all the participants said they did sometimes, depending on the context: Variation from MAE was fine in informal contexts (e.g., conversation with friends), but correction became relevant in more formal contexts (e.g., writing, testing). Black and White participants referred to a standard variety of English and the importance of children learning it. When speaking of their own practices for responding to variation from MAE in Black children's speech, Black teachers talked not only about correcting or not correcting depending on the formality or informality of the occasion but also of not wanting to interfere with children communicating.
Transcript 3. “I'd rather them be able to talk to me.”
| T: | I try to be more mindful of it because I want to model more proper English | |
| I guess. But if they're communicating and they're talking, I just let it go. | ||
| R: | Mhm. | |
| T: | Cuz because I'd rather them be able to talk to me than me just stopping | |
| and correcting them and they don't feel comfortable um trying to say what | ||
| they wanna say | ||
| R: | Mhm. | |
| T: | And then at the end, (like) the only time (I guess) I really correct it is if | |
| We're writing or we're talking the same sentences. Then I say | ||
| ((teacher voice)) This is how you would say that sentence. | ||
| But just an (infor-) informal conversation. | ||
| R: | Mmhmm | |
| T: | I'm not very strict on it. |
Emphasis on Sequencing and Details in Narratives
With regard to their assessment of the children's narratives, all the participants said that they paid attention to sequencing and details, and most said that they preferred the picture-prompted stories because they had clearer sequencing, more detail, and better flow. It was not always clear what was meant by “flow,” but in most instances, it seemed to refer to narrative sequence and/or a smooth delivery with minimal prompting. All the teachers reported attending primarily to narrative sequence (a clear beginning, middle, and end) and the provision of details, which is consistent with Ohio's Learning Standards for second grade. Parents and teachers alike associated the personal narratives with less clear sequential structure. Several participants surmised that the pictures made it easier for the children to tell better stories by providing a visual sequence of events to describe, whereas the personal narratives seemed to be more challenging for children because they had to be recounted from memory and thus placed greater demands on the children. White participants gave more negative assessments than did Black participants of stories that did not have what they recognized as a clear narrative structure.
Overall, Black teachers and parents gave more positive assessments of the children's narratives than did White teachers and parents. Black and White participants expressed appreciation for stories that included humor, emotion, creativity, and expressive delivery. However, Black participants used more positive descriptors (e.g., “funny,” “creative”), and several spoke in detail about the stories they particularly liked. Moreover, only Black participants expressed appreciation for the variety of ways in which the children told their stories, as we see in an excerpt from a Black teacher (Transcript 4).
Transcript 4. “People aren’t going to tell stories in the same way.”
| R: | Based on your answer as far as I understood you have beentraveling a lot | |
| so that you have been exposed to the different varieties of (the) English in | ||
| the United States right? | ||
| T: | Mmhmm, yeah. | |
| R: | Do you think those experiences actually affect your perception today in | |
| any ways? | ||
| T: | Probably. | |
| R: | Mhmm. | |
| T: | I think it helps me to know that people aren't all going to tell stories in the | |
| same way. | ||
| R: | Mhmm. | |
| T: | So you have to figure out what's important to you. (like) What do you | |
| think makes a good story? Um and so that's what I was trying to think | ||
| about for myself. | ||
| R: | Mhmm. | |
| T: | What are the important parts for me? And it might come in different | |
| versions from different kids. | ||
| R: | Mhmm. Okay. |
Discussion and Limitations
This article presented a mixed-methods study examining adults' perception of Black children's narrative quality. Having analyzed and reported results from quantitative and qualitative data separately, we integrate the two in our discussion of the key findings. In this section, we discuss how findings from the qualitative and quantitative analyses provide additional insights into the main findings from the quantitative analyses.
First, adults rated students with less variation from MAE more highly than students with greater variation from MAE for fictional narratives but not for personal narratives. Why would variation from MAE matter more to raters when the narrative is prompted by a five-picture sequence? Participants may have perceived the picture-prompted narratives as more formal, more academic instances of language production than the personal narratives and consequently been less accepting of variation from MAE. Although the survey results do not show that adults listened for “correct grammar,” which is often tied to language variation, the interview data indicate most participants believed that variation from MAE was more acceptable in informal contexts and less acceptable in formal contexts. These ideologies of language standardization and dualism pervade American society and influence the formal/informal distinction reflected in pedagogical approaches to teaching MAE to children who speak AAE and other nonmainstream varieties of English (cf. Wheeler & Swords 2010).
Second, parents rated personal narratives more highly than did teachers, but they did not rate the fictional narratives more highly. Our results align with those of prior studies in which teachers preferred listening to topic-centered narratives that aligned with educational benchmarks steeped in ideologies of dualism between orality and literacy (Bloome et al., 2010; Michaels, 1981). In the current study, fictional narratives tended to be favored over personal narratives because they were presented in topic-centered, literate fashion.
Survey results indicated both groups of adults tended to listen for evidence that the child's narrative was topic centered. That is, adults tended to listen for amount of relevant details and specific vocabulary, inclusion of critical parts of the narrative, and expressed thoughts that flowed together thematically. Picture-prompted narratives supported the telling of a forward-moving, sequential narrative closely woven around the theme of arriving to school late; teachers and parents deemed these fictional narratives of higher quality than personal narratives. Yet, studies of school-age children indicate that personal narration allows for more improvisation and opportunities to express comedic verve and rare vocabulary than does fictional narration (cf. Mills et al., 2017). Parents also valued these more performative narrative qualities, listening for how funny children sounded significantly more than did teachers.
Interview data aligned with results yielded from rating and survey and provided additional nuance regarding the Role × Narrative Type interaction on adult ratings. In interviews, teachers were clear that they cared most about the children's narratives having a clear beginning, middle, and end and plenty of details (consistent with Ohio Learning Standards). Parents, on the other hand, were less specific and less insistent about narrative structure, speaking more often of “flow,” and they talked more about humor, emotion, and vocal delivery of the narrative, as has been found in previous work (Newman & McGregor, 2006). Overall, the personal narratives followed conventional narrative structure less than the fictional stories, and parents seem to have been more accepting of this than were the teachers.
Although parents appeared to be less entrenched in ideologies of dualism than were teachers, several parents noted that they used the picture sequence to help them determine if the child was telling the story accurately. This suggests that at least some parents were holding children accountable to the picture sequence, which they could not do with the personal narratives.
Finally, Black adults tended to rate narratives more highly than did White adults, and this aligns with qualitative findings of both the current study and prior studies. For example, Heath's (1983) work suggests that Black adults may encourage Black children to tell funny and entertaining narratives as a language socialization practice. Likewise, the current study illustrates that Black adults tended to prefer performative, topic-associating narratives to a greater extent than did White adults. In interviews, Black participants spoke more positively and more extensively about the children's narratives, they were more accepting of narratives with less clear sequential structure, and they explicitly valued the diverse ways that the children told their fictional narratives. The ideology of monolingualism did not seem to actively inform how Black participants rated narratives, given that they welcomed narrative repertoire and diversity. Black teachers, although they are clearly oriented to academic standards in their assessments of children's stories and spoke of the value of learning MAE, also expressed appreciation for the Black children's language and an aversion to inhibiting it.
Limitations of the study reside within shortcomings of each methodological tradition. Quantitative methods, such as numerical rating and surveys, provide control and constraint that are beneficial for identifying patterns, but not for unveiling subtle language ideologies. Qualitative methods, such as open-ended interviews, provide vivid depictions of what patterns mean. The order of study tasks were such that interviews followed survey of listening influences, potentially confounding researcher meaning with participant meaning. It is unclear, for example, whether “flow” carries the same meaning for researchers who developed the survey and participants who took the survey before conversing about narrative language with the interviewer. The concurrent triangulation method was employed to bring together the two methodological traditions. We will continue working toward an even deeper integration of the findings than we achieved in the current study.
Findings from the current study have implications for how children's narrative language is assessed by researchers, school-based professionals, and parents. To address the educational disparities faced by Black students, it is critically important to reflect on the ideologies that direct the social actions we take around evaluating their narrative language. Moreover, we need to think collectively with professionals outside of niche areas to attenuate ideologies that disenfranchise Black students.
In summary, preliminary findings from this mixed-methods study indicate language variation, narrative type, and race all matter in the formation of adults' perceptions of Black children's narrative language. School-based professionals may benefit from opportunities to explore and share their language ideologies with interdisciplinary colleagues and, critically, with parents of children from historically marginalized groups, such as Black children. Future studies will examine the role of explicit bias in perceptions of narrative language quality.
Acknowledgments
This research was supported by a Cross-College Seed Grant from The Ohio State University, which was awarded to the first two authors. We are grateful to the parents and teachers who participated in the study and to those who assisted with the recruitment snowball. We thank the research assistants who helped with transcription: Ryan Giavasis, Kallan Imhoff, Larkin Iverson, Raymond Patt, and Hannah Sherman. We also thank Stephanie Antetomaso for her assistance with stimuli development.
Footnotes
1
Research has shown that Black children's AAE production is likely to vary based on how narratives are elicited. For example, Mills (2015b) found that AAE production was higher when children told stories from a detailed picture than from a picture sequence.
2
Approval for this study was obtained from the institutional review board of the University of Houston after the first author moved to that institution.
3
This theme aligns with Blake and Cutler's (2003) finding that teachers are likely to be “circumspect” in their expression of attitudes about nonstandardized varieties of English.
References
Alim, H. S., Rickford, J. R., & Ball, A. F. (2016). Raciolinguistics: How language shapes our ideas about race. Oxford University Press.
Anderson, T. K., & Felsenfeld, S. (2003). A thematic analysis of late recovery from stuttering. American Journal of Speech-Language Pathology, 12(2), 243–253.
Blake, R., & Cutler, C. (2003). AAE and variation in teachers' attitudes: A question of school philosophy? Linguistics and Education, 14(2), 163–191.
Bloome, D., Katz, L., & Champion, T. (2010). Young childrens' narratives and ideologies of language in classrooms. Reading & Writing Quarterly, 19(3), 205–223.
Bloomquist, J., Green, L. J., & Lanehart, S. L. (2015). The Oxford handbook of African American language. Oxford University Press.
Bolden, G. B. (2015). Transcribing as research: “Manual” transcription and conversation analysis. Research on Language and Social Interaction, 48(3), 276–280.
Braun, V., Clarke, V., Hayfield, N., & Gareth, T. (2018). Thematic analysis. In P. Liamputtong (Ed.), Handbook of research methods in health social sciences (pp. 1–18). Springer Nature.
Brown, L., Sherbenou, R. J., & Johnsen, S. K. (2012). Test of Nonverbal Intelligence–Fourth Edition. Pearson.
Burns, F. A., de Villiers, P. A., Pearson, B. Z., & Champion, T. B. (2012). Dialect-neutral indices of narrative cohesion and evaluation. Language, Speech, and Hearing Services in Schools, 43(2), 132–152.
Carter, P. L. (2010). Race and cultural flexibility among students in different multiracial schools. Teachers College Record, 112(6), 1529–1574.
Champion, T. (1998). “Tell me something' good”: A description of narrative structures among African American children. Linguistics and Education, 9(3), 251–286.
Champion, T., Seymour, H., & Camarata, S. (1995). Narrative discourse of African American children. Journal of Narrative & Life History, 5(4), 333–352.
Christensen, S. C., Wright, H. H., Ross, K., Katz, R., & Capiluto, G. (2009). What makes a good story? The naïve rater's perception. Aphasiology, 23(7–8), 898–913.
Collins, J., & Blot, R. K. (2003). Literacy and literacies: Texts, power, and identity. Studies in the social and cultural foundations of language. Cambridge University Press.
Craig, H. K., & Washington, J. A. (2006). Malik goes to school: Examining the language skills of African American students from preschool–5th grade. Erlbaum.
Creswell, J. W., Plano Clark, V. L., Gutmann, M. L., & Hanson, W. E. (2003). Advanced mixed methods research designs. In A. Tashakkori & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp. 209–240). Sage.
Donovan, M. S., & Cross, C. T. (2002). Minority students in special and gifted education. National Academy Press.
Dunn, L. M., & Dunn, D. M. (2007). Peabody Picture Vocabulary Test–Fourth Edition. Pearson.
Dunn, T. J., Baguley, T., & Brunsden, V. (2014). From alpha to omega: A practical solution to the pervasive problem of internal consistency estimation. British Journal of Psychology, 105(3), 399–412.
Farr, M., & Song, J. (2011). Language ideologies and policies: Multilingualism and education. Language and Linguistics Compass, 5(9), 650–665.
Flores, N., & Rosa, J. (2015). Undoing appropriateness: Raciolinguistic ideologies and language diversity in education. Harvard Educational Review, 85(2), 149–171.
Ford, D. Y. (2011). Reversing underachievement among gifted black students: Promising practices and programs (2nd ed.). Prufrock Press, Inc.
Gillam, R. B., & Pearson, N. (2004). Test of Narrative Language. Pro-Ed.
Griffin, T. M., Hemphill, L., Camp, L., & Wolf, D. P. (2004). Oral discourse in the preschool years and later literacy skills. First Language, 24(71), 123–147.
Hallett, J. (2015). Contexts for student AAE use in the classroom. Critical Inquiry in Language Studies, 12(1), 1–26.
Heath, S. B. (1983). Ways with words: Language, life, and work in communities and classrooms. Cambridge University Press.
Hepburn, A., & Bolden, G. (2013). The conversation analytic approach to transcription. In J. Sidnell & T. Stivers (Eds.), Handbook of conversation analysis (pp. 57–76). Wiley.
Hyon, S., & Sulzby, E. (1994). African American kindergartners' spoken narratives: Topic associating and topic centered styles. Linguistics and Education, 6(2), 121–152.
Irvine, J. T. (2016). Language ideology. In Oxford bibliographies in anthropology. Oxford University Press. https://www.oxfordbibliographies.com/view/document/obo-9780199766567/obo-9780199766567-0012.xml
Kroskrity, P. (2018). Language ideologies and language attitudes, Oxford bibliographies in anthropology. Oxford University Press. https://www.oxfordbibliographies.com/view/document/obo-9780199772810/obo-9780199772810-0122.xml?rskey=Xl3bNM&result=2&q=Paul+kroskrity#firstMatch
Lance, C. E., Butts, M. M., & Michels, L. C. (2006). The sources of four commonly reported cutoff criteria: What did they really say? Organizational Research Methods, 9(2), 202–220.
Leu, D. J., Forzani, E., Rhoads, C., Maykel, C., Kennedy, C., & Timbrell, N. (2014). The new literacies of online research and comprehension: Rehtinking the reading achievement gap. Reading Research Quarterly, 50(1), 37–59.
Loban, W. (1976). Language development: Kindergarten through grade twelve. National Council of Teachers of English.
Mahurin-Smith, J., Mills, M., & Chang, R. (2021). Rare vocabulary production in school-age narrators from low-income communities. Language, Speech, and Hearing Services in Schools, 52(1), 51–63.
Michaels, S. (1981). “Sharing time”: Children's narrative styles and differential access to literacy. Language in Society, 10(3), 423–442.
Miller, J., & Iglesias, A. (2012). Systematic Analysis of Language Transcripts (SALT), Research Version 2012 [Computer Software] . SALT Software, LLC.
Mills, M. T. (2015a). Narrative performance of gifted African American school-aged children from low-income backgrounds. American Journal of Speech-Language Pathology, 24(1), 36–46.
Mills, M. T. (2015b). The effects of visual stimuli on the spoken narrative performance of school-age African American children. Language, Speech, and Hearing Services in Schools, 46(4), 337–351.
Mills, M. T., & Fox, M. (2016). Language variation and theory of mind in typical development: An exploratory study of school-age African American narrators. American Journal of Speech-Language Pathology, 25(3), 426–440.
Mills, M. T., Mahurin-Smith, J., & Steele, S. C. (2017). Does rare vocabulary use distinguish giftedness from typical development?: A study of school-age African American narrators. American Journal of Speech-Language Pathology., 26(2), 511–523.
Mills, M. T., & Washington, J. A. (2015). Managing two varieties: Code-switching in context. In S. Lanehart (Ed.), Oxford handbook of African American Language (pp. 566–581). Oxford University Press.
Mills, M. T., Watkins, R. V., & Washington, J. A. (2013). Structural and dialectal characteristics of the fictional and personal narratives of school-age African American children. Language, Speech, and Hearing Services in Schools, 44(2), 211–223.
Morgan, L., Marshall, J., Harding, S., Powell, G., Wren, Y., Coad, J., & Roulstone, S. (2019). “It depends”: Characterizing speech and language therapy for preschool children with developmental speech and language disorders. International Journal of Language & Communication Disorders, 54(6), 954–970.
National Center for Education Statistics. (2011). The nation's report card: Writing 2011 (NCES 2012-470) . Institute of Education Sciences, U.S. Department of Education. http://nces.ed.gov/nationsreportcard/pdf/main2011/2012470.pdf
Newman, R. M., & McGregor, K. K. (2006). Teachers and laypersons discern quality differences between narratives produced by children with or without SLI. Journal of Speech, Language, and Hearing Research, 49(5), 1022–1036.
Nowell, L. S., Norris, J. M., White, D. E., & Moules, N. J. (2017). Thematic analysis: Striving to meet the trustworthiness criteria. International Journal of Qualitative Methods, 16(1), 1–13.
Psychology Software Tools, Inc. (2016). E-Prime 3.0. https://www.pstnet.com
Reese, E., Suggate, S., Long, J., & Schaughency, E. (2010). Children's oral narrative and reading skills in the first 3 years of reading instruction. Reading and Writing, 23(6), 627–644.
Revelle, W. (2014). Package ‘psych.’ http://cran.r-project.org/web/packages/psych/psych.pdf
Riley, K. (2011). Language socialization and language ideologies. In A. Duranti, E. Ochs, & B. Schieffelin (Eds.), Handbook of language socialization (pp. 493–534). Wiley.
Robinson, G. C., & Norton, P. C. (2019). A decade of disproportionality: A state-level analysis of African American students enrolled in the primary disability category of speech or language impairment. Language, Speech, and Hearing Services in Schools, 50(2), 267–282.
Roulston, K. (2006). Close encounters of the “CA” kind: a review of literature analysing talk in research interviews. Qualitative Research, 6(4), 515–534.
Salmon, W. (2015). Language ideology, gender, and varieties of Belizean Krol. Journal of Black Studies, 46(6), 605–625.
Samuelsson, C., & Plejert, C. (2014). On the use of conversation analysis and retrospection in intervention for children with language impairment. Child Language Teaching and Therapy, 31(1), 19–36.
Seymour, H. N., Roeper, T. W., de Villiers, J., & de Villiers, P. A. (2003). Diagnostic Evaluation of Language Variation–Screening Test. Pearson.
Silver, C., & Lewins, A. (2017). Using software in qualitative research: A step-by-step guide. Sage.
Simmons-Mackie, N. (2014). Micro and macro traditions in qualitative research. In M. J. Ball, N. Müller, & R. L. Nelson (Eds.), Handbook of qualitative research in communication disorders (pp. 17–38). Psychology Press.
Stockman, I. J. (2010). A review of developmental and applied language research on African American children: From a deficit to difference perspective on dialect differences. Language, Speech, and Hearing Services in Schools, 41(1), 23–38.
Suggate, S., Schaughency, E., McAnally, H., & Reese, E. (2018). From infancy to adolescence: The longitudinal links between vocabulary, early literacy skills, oral narrative, and reading comprehension. Cognitive Development, 47, 82–95.
Suleman, S., & Hopper, T. (2014). Mixed methods research and its use in speech-language pathology and audiology research. Canadian Journal of Speech-Language Pathology and Audiology, 38, 386–399.
Transana 3.32d. (2020). [Computer software]. Spurgeon Woods . https://www.transana.com
Vanneman, A., Hamilton, L., Baldwin, J. A., & Rahman, T. (2009). Achievement gaps: How Black and White students in public schools perform in mathematics and reading for the National Assessment of Educational Progress (NCES 2009–455) . U.S. Department of Education.
Wechsler, D. (2009). Wechsler Individual Achievement Test–Third Edition. Pearson.
Wheeler, R. (2019). Attitude change is not enough: Disrupting deficit grading practices to disrupt dialect prejudice. Proceedings of the Linguistic Society of America, 4, 1–12.
Wheeler, R., Cartwright, K. B., & Swords, R. (2012). Factoring AAVE into reading assessment and instruction. The Reading Teacher, 65(6), 416–425.
Wheeler, R., & Swords, R. (2010). Code-switching lessons: Grammar lessons for linguistically diverse writers. Heinemann.
Wilkinson, R. (2012). Conversation analysis and communication disorders. In C. A. Chapelle (Ed.), The encyclopedia of applied linguistics (pp. 962–967). Wiley-Blackwell.
Appendix A
Narrative Elicitation Tools
Personal Narrative Model
When I was your age, my mother gave me permission to go to my friend Khedra's house and play; but I had to come home at five o'clock. When I got to there, my friends from the neighborhood were watching music videos while singing and dancing along. Mrs. Graham came in and said, “If ya'll don't turn that TV down!”(Narrator giggles). We knew the rest. So, we decided to stop singing; but we took our shoes off and kept right on dancing! Ricky had two left feet so everyone steered clear of him on our makeshift dance floor. We were having so much fun that I lost track of time and came home two hours late! I was in a lot of trouble when I got home. When I was sent to my room, I closed the door, sat on my bed, and smiled.
Late for School Narrative Picture
Appendix B
Prompt to Orient Participants
Thank you so much for participating. [NAME] I am going to read you a set of instructions before we begin. There are two parts to what we are going to have you do today.
First, you’ll listen to second graders tell some stories. You will rate the stories based on your perception of the quality of the story and its telling. The number you assign to the story should match your perception of the quality of the story. For example, a higher number should indicate a better story.
In the second part, you will answer a few questions about language.
To begin, go ahead and put the headphones on. I'll play a sound file so that we can adjust the volume. Let me know if this sounds too soft or loud. [examiner pressed function and volume up or down]. Great, I'm glad that the volume is ok. [NAME] once the program begins, you'll not be able to pause or replay stories. The program will prompt you as you go. I'll be here if you have any questions.
Appendix C
Survey Questions
| Question no. | When listening to the child… | 1 No influence | 2 | 3 | 4 | 5 Strong influence |
|---|---|---|---|---|---|---|
| 1 | I listened for the use of specific (vs. general) vocabulary words. | |||||
| 2 | I listened for the variety of vocabulary words that a child used. | |||||
| 3 | I listened to see if the details included were relevant to the story. | |||||
| 4 | I listened for the amount of detail that the child included. | |||||
| 5 | I listened for the inclusion of dialogue between the characters. | |||||
| 6 | I listened to see if all the critical parts of the story were included. | |||||
| 7 | I listened for how well the child's thoughts flowed together. | |||||
| 8 | I listened to see if the child followed the theme of the story. | |||||
| 9 | I listened for the use of correct grammar. | |||||
| 10 | I listened for how complex the child's sentences were. | |||||
| 11 | I listened for the use of complete sentences. | |||||
| 12 | I listened to the length of the child's sentences. | |||||
| 13 | I listened for how quickly and/or slowly a child spoke. | |||||
| 14 | I listened for how well the child produced their speech sounds. | |||||
| 15 | I listened for how easy it was for the child to tell the story. | |||||
| 16 | I listened for the presence of hesitations, pauses, and/or the use of words like um or uh. | |||||
| 17 | I listened for how funny the child was. | |||||
| 18 | I listened to how cute the child sounded. | |||||
| 19 | I listened to see if the child sounded like he or she was telling a story (vs. having a conversation). | |||||
| 20 | I listened for how much emotion the child put into the telling of the story. |
Appendix D
Questions for the Ethnographic Interview
Opening to assure the interviewer and interviewee have shared basic terms:
“Thanks so much for listening to those stories and completing the survey. Now, I'd like to ask you some questions about your experience with speakers of different varieties of American English. Before we get started, I want to clarify that we're talking about American English used by people who learned it as their first and only language. So, not like MY English. Ok, let's get started.
Linguists use the term dialect to refer to patterns in the way people use language, patterns that differ across regions and groups of people. Linguists also use the phrases ‘varieties of English’ and ‘language variation.’”
1.
What do you know about different varieties of English in the United States? How did you learn about them?
2.
Do you regularly use or encounter different varieties of English? In which settings? With whom? For what purposes?
3.
Do you have experience with children who speak a variety of English different from yours? Please tell me about your experiences.
a. Which varieties of English are spoken in your classroom and at your school? By whom? For which activities or purposes?
4.
Is there anything about the stories that stood out to you?
5.
Tell me about a second grader who is really good at telling stories.
Information & Authors
Information
Published In
Language, Speech, and Hearing Services in Schools
Volume 52 • Number 1 • January 2021
Pages: 84-99
PubMed: 33464984
Copyright
Copyright © 2021 The Authors. 
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.History
- Received: Feb 4, 2020
- Revised: Apr 24, 2020
- Accepted: Jun 16, 2020
- Published online: Jan 19, 2021
- Published in issue: Jan 19, 2021
Authors
Metrics & Citations
Metrics
Article Metrics
View all metricsCitations
Download article citation data for:
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
Citing Literature
- Nicole Gardner-Neblett, Angelica Ramos, Allison De Marco, When Ebony and Malik share stories in school: White teachers’ perceptions of children's use of African American English during oral storytelling, Early Childhood Research Quarterly, 10.1016/j.ecresq.2025.02.006, 72, (143-155), (2025).
- Kia N. Johnson, Monique T. Mills, Exploratory Examination of Speech Disfluencies in Spoken Narrative Samples of School-Age Bidialectal Children, American Journal of Speech-Language Pathology, 10.1044/2023_AJSLP-21-00158, 32, 3, (1182-1194), (2023).
- Elysha Clark-Whitney, Gigliana Melzi, Executive Function and Narrative Language Abilities in Emergent Bilingual Preschoolers: An Exploratory Study, Language, Speech, and Hearing Services in Schools, 10.1044/2022_LSHSS-21-00190, 54, 2, (584-599), (2023).
- Nicole Gardner-Neblett, Allison De Marco, Blake D. Ebright, Do Katie and Connor tell better stories than Aaliyah and Jamaal? Teachers’ perceptions of children's oral narratives as a function of race and narrative quality, Early Childhood Research Quarterly, 10.1016/j.ecresq.2022.07.014, 62, (115-128), (2023).
- Rachel E. Schachter, Shayne B. Piasta, Doing Assessment: A Multicase Study of Preschool Teachers’ Language and Literacy Data Practices, Reading Research Quarterly, 10.1002/rrq.419, 57, 2, (515-535), (2021).
- Paul Atkinson, Natasha Carver, Ethnopoetics and Narrative Analysis, Narrative Works, 10.7202/1076918ar, 10, (60-70), (2021).
- Monique T. Mills, Forum: Serving African American English Speakers in Schools Through Interprofessional Education & Practice, Language, Speech, and Hearing Services in Schools, 10.1044/2020_LSHSS-20-00161, 52, 1, (1-3), (2021).
View Options
View options
Sign In Options
ASHA member? If so, log in with your ASHA website credentials for full access.
Member Login
