Skip to main content
NCBI home page
Language, Speech, and Hearing Services in Schools logoLink to Language, Speech, and Hearing Services in Schools
. 2024 May 17;55(3):870–883. doi: 10.1044/2024_LSHSS-23-00180

Transcription Decisions of Conjoined Independent Clauses Are Equitable Across Dialects but Impact Measurement Outcomes

Janna B Oetting a,✉, Tahmineh Maleki a
PMCID: PMC11253809  PMID: 38758707

Abstract

Purpose:

Transcription of conjoined independent clauses within language samples varies across professionals. Some transcribe these clauses as two separate utterances, whereas others conjoin them within a single utterance. As an inquiry into equitable practice, we examined rates of conjoined independent clauses produced by children and the impact of separating these clauses within utterances on measures of mean length of utterance (MLU) by a child's English dialect, clinical status, and age.

Method:

The data were archival and included 246 language samples from children classified by their dialect (African American English or Southern White English) and clinical status (developmental language disorder [DLD] or typically developing [TD]), with those in the TD group further classified by their age (4 years [TD4] or 6 years [TD6]).

Results:

Rates of conjoined independent clauses and the impact of these clauses on MLU varied by clinical status (DLD < TD) and age (TD4 < TD6), but not by dialect. Correlations between the rate of conjoined clauses, MLU, and language test scores were also similar across the two dialects.

Conclusions:

Transcription decisions regarding conjoined independent clauses within language samples lead to equitable measurement outcomes across dialects of English. Nevertheless, transcribing conjoined independent clauses as two separate utterances reduces one's ability to detect syntactic differences between children with and without DLD and document syntactic growth as children age.

Supplemental Material:

https://doi.org/10.23641/asha.25822675

Language sample analysis is a recommended clinical practice and commonly utilized in research across various dialects of a language and across languages (e.g., Bernstein Ratner & MacWhinney, 2023; Bùi & Leonard, 2024; Pearce & Flannagan, 2024). Unlike norm-referenced tests that rely on structured tasks and materials that may be unfamiliar to a child, language samples can be unstructured and elicited with materials from a child's home or community. Language samples can also be tailored to a child's age—elicited through conversational play for young children, through narratives for slightly older children, and through expositions and persuasive monologues for adolescents and adults (Brimo & Hall-Mills, 2019; Coughler et al., 2024; Shirley et al., 2024). Across each of these elicitation contexts, language samples provide multiple opportunities for individuals to talk about their experiences, daily routines, special events, and a host of other topics. For these reasons, language samples are often viewed as an ecologically valid measurement context for children, regardless of children's cultural backgrounds, languages, and dialects (Ebert & Pham, 2017; Kapantzoglou et al., 2017; Liebenberg et al., 2023; Mills et al., 2017).

Once elicited, language samples must be transcribed, coded, and analyzed. Not surprisingly, multiple systems exist for guiding language sample practices (Finestack et al., 2014). Repeated attention to three language sample systems has shed light on some of the differences that exist in the field (e.g., Bernstein Ratner & MacWhinney, 2023; Garbarino et al., 2020; Guo et al., 2018). These systems include Systematic Analysis of Language Samples (SALT; Miller & Iglesias, 2024), Computerized Language Analysis (CLAN; MacWhinney, 2000), and Sampling Utterances and Grammatical Analysis Revised (SUGAR; Pavelko & Owens, 2017). Pezold et al. (2020) examined these three systems, comparing them for age of child and type of elicitation context reflected in their reference databases, treatment of abandoned and unintelligible utterances, examiner cost and time, and number and type of disagreements between examiners for transcription and coding. Although some differences were identified, the authors recommended all three systems, with the caveat that researchers and clinicians should understand the differences between the systems and select the system and reference database that best fits their needs.

Other types of language sample decisions have also been evaluated. For example, researchers have examined the impact of sample length (e.g., 25 vs. 50 vs. 100 utterances), sample time (e.g., 5 vs. 60 min.), elicitation context (e.g., conversation vs. narrative; oral vs. written), and examiner race and dialect on children's utterances (e.g., Guo & Eisenberg, 2015; Heilmann et al., 2010; Liebenberg et al., 2023; Matiasovitsová et al., 2024; Pavelko et al., 2020; Southwood & Russell, 2004; Spencer et al., 2023; Wilder & Redmond, 2022). Others have examined the clinical utility of measures that can be generated from a language sample (e.g., Francois et al., 2023; Horton-Ikard et al., 2005; Oetting et al., 2010; Overton et al., 2021; Ramos et al., 2022; Stockman et al., 2016). Overton et al. (2021) did this using language samples from 37 children (15 with developmental language disorder [DLD]; 22 classified as typically developing [TD]) who were under the age of 6 years and spoke African American English (AAE).1 Using KidEval, which is part of the CLAN suite of programs, the DLD group was found to earn lower mean scores than the TD group on four measures: total score from Developmental Sentence Score (DSS), total score from Index of Productive Syntax (IPSyn), mean length of utterance in words (MLU-W), and number of verbs per utterance. From these results, Overton et al. (2021) concluded that all four measures have clinical utility within AAE for children under the age of 6 years (but for DSS and IPSyn and children aged 6 years and older, see Oetting, 2005; Oetting et al., 2010).

Less attention has been given to transcription decisions related to utterance boundaries, including how conjoined independent clauses are transcribed (e.g., I went to the store and I bought new shoes). Clinicians who transcribe utterances using terminal units (i.e., T-units) or communication units (i.e., C-units) separate conjoined independent clauses, allowing an utterance to contain one and only one independent clause (e.g., Ebert & Pham, 2017; Eisenberg & Guo, 2013, 2016; Gardner-Neblett, 2024; Miller & Iglesias, 2024). A T-unit is an utterance with an independent clause that contains a subject and verb and all related dependent clauses (Loban, 1976). T-units often are used when clinicians evaluate a child's writing or compare oral and written samples to each other (e.g., Mills et al., 2017); with T-units, nonclausal utterances are not included in the analysis set. Examples of T-units are, Then the mom had to call the hospital and My mama crashed when the road was slippery. C-units also can include independent clauses with a subject and verb and all related dependent clauses, but because C-units were devised for the analysis of conversation, a C-unit can include clauses without these elements, especially in response to an interlocuter's utterance (e.g., Examiner: I wonder who did that. Child: The gym teacher).

Others allow up to two independent clauses within an utterance if they are syntactically conjoined within an utterance (e.g., Lee & Canter, 1971; L. Leonard, personal communication, February 15, 2024; Oetting et al., 2021; Pavelko & Owens, 2023; Retherford et al., 2019; Rice et al., 2010). Examples of these types of utterances are, It went to that level, but I wanted it to go on the top level, and I spilled it, so I couldn't do it no more. Lee's (1974) DSS is an example of a language sample measure that is based on utterances with conjoined independent clauses. DSS provides scores that range from 1 to 8 for each utterance analyzed, and one of the grammatical categories scored is conjunctions. IPSyn also includes the scoring of conjunctions that join two independent clauses together (i.e., Item 12 within the sentence score items; Altenberg et al., 2018).

In 1990s, when the first author began collecting language samples from children who spoke various dialects of rural and urban AAE and various rural dialects of Southern White English (SWE), we allowed up to two syntactically conjoined independent clauses within an utterance (e.g., Berry & Oetting, 2017; Cleveland & Oetting, 2013; Garrity & Oetting, 2010; Lee & Oetting, 2014; Oetting, 2005; Oetting & Horohov, 1997; Oetting et al., 2021; Oetting & McDonald, 2001; Pruitt & Oetting, 2009; Pruitt et al., 2011; Rivière et al., 2018). In many studies, we included 5- and 6-year-olds with DLD and age-matched TD controls; in a few studies, we included language-matched TD controls. During the conduct of these studies, we never questioned our transcription practices or those of others. Instead, our concern focused on transparency (i.e., maintaining a manual within the lab with directions and examples to guide transcription) and reliability (i.e., engaging in regular and systematic checks of transcription agreement across samples). In other words, we assumed that the same transcription method applied to all language samples would lead to equitable measurement outcomes for children, regardless of their dialect, clinical status, and age.

However, for language sample data to be viable and useful for clinical practice, large repositories of language samples need clinicians and researchers to contribute their language samples to these repositories, including those associated with SALT, CLAN, and SUGAR. Preparing our language samples for these repositories led us to question the transcription of conjoined independent clauses. With others, we have also become more aware and critical of the foundational assumptions and practices of child language studies. Indeed, multiple scholars have noted that much of what is known about child speech and language has been developed by and for monolingual speakers of General American English (GAE; e.g., Garivaldo & Fabiano-Smith, 2023; Padia, 2023; Soto-Boykin et al., 2023; see also Hyter & Salas-Provance, 2023).

In a forum for Language, Speech, and Hearing Services in Schools, Brea-Spahn and Fabiano-Smith (2023) asked professionals to reflect on commonly accepted practices and to think deeply about the narrow lens by which speech-language pathology has been developed compared to the much broader and pluralistic society of the children we seek to serve. Inspired by these authors, the current study was designed to examine equity in language sample transcription practices in the field as it relates to conjoined independent clauses, asking if children's production of these types of clauses and their impact on language sample measures vary by a child's dialect, clinical status, and age. Studies of equitable transcription practice are needed now more than ever, with the growing popularity of computer automated software programs. These programs quickly generate analyses from language samples, but they are dependent on how utterances are transcribed (e.g., Fox et al., 2022). Equity in language sample transcription practices across dialects of English has also not been a focus of research. As an example, in a review of 40 years of studies involving African American children, Horton et al. (2018) identified the appropriateness of tests and alternative scoring methods as a focus topic, without noting a focus on the practice of language sample transcription.

The language sample measure selected for study was mean length of utterance (MLU). This measure was selected given its frequent use in clinical practice; the ease at which a child's MLU can be generated from a sample when using SALT, KidEval, or SUGAR; and Overton et al.'s (2021) findings showing MLU to be sensitive to childhood DLD within the dialect of AAE. Moreover, in a review of child language articles published between 2020 and 2023 in the American Journal of Speech-Language Pathology; Language, Speech, and Hearing Services in Schools; and Journal of Speech, Language, and Hearing Research, over 20 have included MLU either as a descriptive measure or as a dependent variable.

Why Might the Same Transcription Practice Lead to Inequitable Outcomes Across Dialects?

Transcription of conjoined independent clauses assumes that children across dialects are uniform in their clause conjoining preferences. If they are not—if one dialect group is more likely to conjoin clauses and grow their grammars with syntactic coordination, and another is more likely to embed clauses and grow their grammars in other ways (e.g., with subordination)—then dividing all conjoined independent clauses into separate utterances could affect one dialect group more than another.

Cross-dialectal studies of independent clause coordination are lacking, but there are studies showing dialectal differences among children related to narrative and communicative style, and these differences could influence children's conjoining preferences when talking. In a review of narrative studies, Champion and McCabe (2015) concluded that AAE-speaking children usually tell topic-centered, classically formed stories, but they can occasionally tell performative ones, which increases a child's use of repetition, parallelism, and digression. At the clause level, Oetting et al. (2021) found child speakers of AAE to produce higher proportions of utterances supporting past progressive clauses (e.g., He was helping; They were running) as compared to child speakers of SWE. Dialectal differences in story structure or clause production preferences could lead to style differences pertaining to a child's frequency of conjoined independent clauses within language samples.

Clinical Status and Age Effects as Indices of Equitable Practice

Across studies of AAE, SWE, and GAE, children with DLD are found to produce less mature grammars than same dialect-speaking, age-matched TD peers (Eisenberg & Guo, 2013; Garrity & Oetting, 2010; Guo & Schneider, 2016; Guo et al., 2018, 2020; Hendricks & Adlof, 2020; Oetting et al., 2016, 2019, 2021; Rice, 2020; Rudolph et al., 2019; Souto et al., 2014). Studies conducted in various dialects of English also have found that as children age, their grammars mature (e.g., Jackson & Roberts, 2001; Newkirk-Turner et al., 2015; Rice, 2020). Yet, few studies have been conducted to determine if clinical effects and age effects are similar in magnitude across dialects. For transcription practices of conjoined independent clauses to be equitable across dialects, children's rates of conjoined independent clauses and the impact of these clauses on language sample measures, such as MLU, should yield similar, across-dialect magnitudes of clinical effects and age effects.

Examining clinical effects and age effects differs from what has been done in previous studies, where the focus has been on whether there are differences in the surface forms children use to express grammatical structure. Consider Seymour et al.'s (1998) seminal study of AAE language samples. The authors examined children's use of but and and when used as a phrasal or clausal conjunction along with 16 other grammar structures. Results showed that the children's AAE conjunctions were composed of the same overt surface forms (i.e., but, and) as spoken in GAE, and the TD AAE group produced these overt surface forms in 97% of the utterances that required a conjunction. From these results, Seymour et al. (1998) concluded that expressions of conjunctions are the same across AAE and GAE.

In contrast, here we are asking whether there are dialectal differences in the number (i.e., rate) of conjoined independent clauses produced by children with and without DLD and/or children of different ages. This is important to do because Seymour et al.'s (1998) study did not find a difference in conjunctions when AAE DLD samples were compared to AAE TD samples. Instead, the AAE DLD and TD groups produced the conjunctions at similar frequencies (DLD raw total = 36; TD raw total = 31) and with similar proportions of overt surface forms (DLD = 95%; TD = 97%). These findings suggest that children's productions of conjunctions may not be as sensitive to clinical effects in AAE as it may be in other dialects. The current study was designed to examine this possibility using language samples from AAE and SWE child speakers as well as examine across-dialect age effects in the children's conjoining of independent clauses.

The research questions were as follows: (a) Do children's productions of conjoined independent clauses differ by dialect, clinical status, and/or age? If so, (b) do transcription decisions regarding conjoined clauses differentially affect children's MLU values by dialect, clinical status, and age?

Method

Data

The data were archival language samples from 246 (89 SWE and 157 AAE) children who lived in rural and urban areas in southeastern Louisiana and attended a child care center, preschool, Head Start, or public kindergarten. The samples were collected during three time periods between the 1990s and 2015 (for details regarding procedures, child profiles, and decision-making processes for classifying children as DLD or TD, see Supplemental Material S1), and various subsets of them have been analyzed as part of several studies (e.g., Berry & Oetting, 2017; Cleveland & Oetting, 2013; Garrity & Oetting, 2010; Oetting, 2005; Oetting et al., 1999, 2010, 2021; Oetting & Horohov, 1997; Oetting & McDonald, 2001, 2002; Oetting & Newkirk, 2008; Pruitt & Oetting, 2009; Rivière et al., 2018). However, at the time of data collection, all children passed a hearing screening conducted at school and demonstrated age-appropriate articulation as measured by a standard score higher than −1 SD on the Words-in-Sounds subtest of the Goldman-Fristoe Test of Articulation (Goldman & Fristoe, 1986, 2000) and/or passing either the phonological screener of the Test of Early Grammatical Impairment (Rice & Wexler, 2005) or a screener of specific phonemes involving /z/, /m/, /r/, /t/, or /d/ in the word-final position. Within previous studies, the children who contributed the samples have been classified by dialect (AAE = 157 vs. SWE = 89) and clinical status (DLD = 94 vs. TD = 152), with the children in the TD group further classified by age (6 years [TD6] = 101 vs. 4 years [TD4] = 51).

The participants' ages ranged from 39 to 92 months, and maternal education levels as reported by 178 caregivers ranged from sixth grade to beyond a college degree. To classify the children's dialects, trained graduate students listened to short audio excerpts from the language samples following the procedures of Oetting and McDonald (2002). In addition, the densities at which the children produced nonmainstream surface forms (e.g., zero forms, such as He Ø walking; dialect-specific overt forms, such as ain't and I'ma; and dialect-specific clausal features, such as multiple negation as in don't got no…) within their language samples were calculated by dividing the sum of these forms by the number of complete and intelligible utterances within each sample. Group averages of age, maternal education, and density of nonmainstream form use within the language samples are presented in Table 1 (for analyses, see Supplemental Material S2).

Table 1.

Characteristics of participants who contributed the language samples.

Measure AAE
 SWE
DLD
n = 61		 TD6
n = 64		 TD4
n = 32		 DLD
n = 33		 TD6
n = 37		 TD4
n = 19
Agea 71.33
(6.84)		 69.17
(5.33)		 57.28
(4.55)		 70.52
(7.89)		 71.49
(7.03)		 48.26
(4.90)
MEDb 11.30
(2.18)		 13.94
(2.45)		 14.24
(2.15)		 12.41
(2.61)		 13.08
(2.69)		 14.00
(1.63)
Nonmainstream form densityc .36
(.10)		 .29
(10)		 .27
(.10)		 .23
(.13)		 .11
(.05)		 .12
(.05)
Open in a new tab

Note. Means reported first followed by standard deviations in parentheses. AAE = African American English; SWE = Southern White English; DLD = developmental language disorder, TD6 = typically developing 6-year-olds; TD4 = typically developing 4-year.

a

Reported in months.

b

MED = highest level of maternal education reported in years, with 12 = high school.

c

Proportion of utterances in language sample with a nonmainstream dialectal form.

All children were classified as either DLD or TD using a battery of assessments. These assessments varied across studies, but for the children's grammar abilities, we administered the grammar subtests from either the Test of Language Development–Primary: Second Edition (Newcomer & Hammill, 1988) or Third Edition (Hammill & Newcomer, 1997) or the Diagnostic Evaluation of Language Variation: Norm Referenced (Seymour et al., 2005). To measure the children's nonverbal IQ, we administered either the Primary Test of Nonverbal Intelligence (Ehrler & McGhee, 2008), the Columbia Mental Maturity Scale (Burgmeister et al., 1972), or subtests from the Leiter International Performance Scale–Revised (Roid & Miller, 1998). Finally, all children completed a version of the Peabody Picture Vocabulary Test–Revised (PPVT-R; Dunn & Dunn, 1981), Third Edition (PPVT-3; Dunn & Dunn, 1997), or Fourth Edition (Dunn & Dunn, 2007). In studies with children classified as DLD, the DLD group was age-matched to a TD6 group, and in some studies, they were language-matched to a TD4 group using either raw score on the PPVT-R or PPVT-3, or MLU. In addition, 15 of the AAE TD4 samples came from children classified as middle-income, vocabulary-matched TD controls for a group of AAE-speaking TD 6-year-olds with low vocabulary scores and maternal education levels below 12th grade (Pruitt & Oetting, 2009).2 Given the various tests administered, each child's standard scores from the tests was converted to a z score using the normative mean and standard deviation of the test, and group means of these z scores by dialect and clinical group are presented in Table 2 (for analyses, see Supplemental Material S2).

Table 2.

Participants' z scores on tests administered.

Measure AAE
 SWE
DLD TD6 TD4 DLD TD6 TD4
Nonverbal IQa n = 244 −0.29 (0.58) −0.001 (0.55) 0.36 (0.63) −0.17 (0.48) 0.08 (0.57) 0.29 (60)
Syntaxb n = 234 −1.84 (0.43) 0.01 (0.55) −0.09 (0.71) −1.69 (0.57) 0.23 (0.74) −0.24 (0.49)
Vocabularyc n = 246 −1.44 (0.69) 0.10 (0.64) −0.19 (0.69) −1.31 (0.71) 0.34 (0.60) 0.15 (0.47)
Open in a new tab

Note. Means reported first followed by standard deviations in parentheses. AAE = African American English; SWE = Southern White English; DLD = developmental language disorder; TD6 = typically developing 6-year-olds; TD4 = typically developing 4-year-olds.

a

Nonverbal IQ (intelligence quotient) reflects z scores calculated from the Primary Test of Nonverbal Intelligence (Ehrler & McGhee, 2008), the Columbia Mental Maturity Scale (Burgmeister et al., 1972), or subtests from the Leiter International Performance Scale–Revised (Roid & Miller, 1998).

b

Syntax reflects z scores calculated from either the syntax subtest of the Test of Language Development–Primary: Second Edition (Newcomer & Hammill, 1988) or Third Edition (Hammill & Newcomer, 1997) or the syntax subtest of the Diagnostic Evaluation of Language Variation: Norm Referenced (Seymour et al., 2005).

c

Vocabulary measure reflects z scores from the Peabody Picture Vocabulary Test–Revised (Dunn & Dunn, 1981), Third Edition (Dunn & Dunn, 1997), or Fourth Edition (Dunn & Dunn, 2007).

Procedure

Collection and Transcription of Language Samples

At the time of data collection, each study was approved by Louisiana State University's institutional review board, and caregiver consent and child assent were secured. The language samples were elicited through conversational-based play by an examiner at each child's school using a standardized play kit involving a baby doll set, gas station set, picnic set, and three pictures depicting familiar activities (e.g., going fishing, grocery shopping, washing a car, and visiting the doctor). Language samples were transcribed and coded using guidelines from SALT (for the most recent version, see Miller & Iglesias, 2024); however, utterances were allowed to contain up to two independent clauses if these clauses were syntactically conjoined with the conjunctions for, and, nor, but, or, yet, or so. These conjunctions often are remembered and referred to using the acronym FANBOYS.

At the time of data collection, each sample was transcribed and coded using a three-pass system, with at least two different examiners completing one of the passes. All complete and intelligible utterances from the samples were included within the analysis; utterances abandoned, interrupted, or containing unintelligible content were excluded. The complete and intelligible utterances within the samples totaled 53,066 and averaged 215.72 (SD = 66.14) per sample. A two-way analysis of variance (ANOVA) indicated that sample length differed by the children's dialect, F(1, 240) = 9.26, p = .003, ηp2 = .04, and group, F(2, 240) = 7.41, p < .001, ηp2 = .06. Follow-up analyses showed the samples to be longer in SWE than in AAE and longer in the DLD group than in the TD6 group, with the TD4 samples being the shortest (although no group mean sample length was less than 180 utterances).

Within the original studies, reliability of language sample transcription was evaluated by having a second set of examiners complete the three-pass system on a subset of the samples or on subsets of utterances from all samples collected. Agreement was at or above 96% for identifying and transcribing complete and intelligible utterances.

Extraction of Conjoined Independent Clauses Within the Language Samples

Using the list function in SALT, we searched for utterances with the seven coordinating conjunctions (i.e., for, and, nor, but, or, yet, so), and utterances were tagged if they conjoined two related independent clauses. Conjunctions conjoining noun or verb phrases within an independent clause were not tagged. For example, in the utterances, I like hotdogs, but John likes burgers, and I like hotdogs but not burgers, only the but in the first utterance was identified as a conjunction conjoining two independent clauses. Two student research assistants independently tagged the conjoined clauses, and disagreements were resolved by the first author. Then, utterances with conjoined independent clauses were divided into two utterances within the samples and saved with new file names. This process led to each child having two samples, the original with the independent clauses conjoined, and another with these clauses separated into two utterances.

Results

Number and Rate of Conjoined Independent Clauses Within the Samples

A total of 2,496 utterances with syntactically conjoined independent clauses were identified in the language samples. Table 3 lists the mean number and proportion (i.e., rate) of these utterances from the total in the children's original samples by each dialect (AAE, SWE) and group (DLD, TD6, TD4). Using rate to control for language sample size differences, the data were analyzed with a 2 × 3 ANOVA and follow-up least significant difference (LSD) t tests. Children's rate of conjoined independent clauses differed by group, F(2, 240) = 20.24, p < .001, ηp2 = .14, but not by dialect (p = .32, ηp2 = .004). Follow-up analysis of the group effect revealed a lower rate of conjoined independent clauses by the DLD and TD4 groups than by the TD6 group. These clinical group and age differences held even when we completed the analyses with the dialects separated, and the effect sizes of the differences were similar in magnitude when the DLD and TD6 groups were compared, AAE: t(1, 123) = −3.83, p < .001, d = .04; SWE: t(1, 68) = −3.81, p < .001, d = .04, and when the TD4 and TD6 groups were compared, AAE: t(1, 94) = −2.75, p < .007, d = .04; SWE: t(1, 53.98) = −5.95, p < .001, d = .03.

Table 3.

Number of utterances and number and rate of conjoined independent clauses in language samples.

Variable AAE
 SWE
DLD TD6 TD4 DLD TD6 TD4
Complete and intelligible utterances per sample 220.98 (63.62) 203.67 (69.42) 181.75 (54.72) 252.52 (77.93) 230.41 (54.95) 204.05 (38.14)
Number of conjoined independent clauses 8.48 (8.52) 13.13 (9.62) 6.56 (6.02) 7.79 (6.44) 15.62 (10.45) 4.95 (4.55)
Rate of conjoined independent clauses per utterance .04 (.04) .07 (.04) .04 (.04) .04 (.03) .06 (.04) .02 (.02)
Open in a new tab

Note. Means reported first followed by standard deviations in parentheses. AAE = African American English; SWE = Southern White English; DLD = developmental language disorder; TD6 = typically developing 6-year-olds; TD4 = typically developing 4-year-olds.

To further examine the children's conjoined independent clauses within each dialect, we divided the children into five subgroups based on their rate data. Subgroup 1 included the 17 children who did not show evidence of conjoining independent clauses (i.e., their rate of utterances with a conjoined independent clause = .00). For the others, the subgroups were defined by their rates as follows: Subgroup 2 = .01 to .05 (n = 149), Subgroup 3 = .06 to .10 (n = 53), Subgroup 4 = .11 to .15 (n = 20), and Subgroup 5 = .16 to .20 (n = 7), with .20 being the highest rate produced within the language samples. When these data were subjected to a chi-square analysis, differences were detected between the DLD, TD6, and TD4 groups, χ2(8, N = 246) = 37.86, p < .001. Most (80%) of the children with DLD were classified in Subgroup 1 or 2, indicating that either they did not conjoin independent clauses or they conjoined infrequently, whereas 51% of the TD6 group were classified in Subgroup 3, 4, or 5. By comparison, the TD4 group showed a more variable distribution, with 78% classified in Subgroups 2 and 3, 18% classified in Subgroup 1, and 4% classified in Subgroup 4 or 5.

Figures 1 and 2 present the same data for the dialects separately. Here we see the same distributional pattern for the DLD, TD6, and TD4 groups as was found for the dialects combined: AAE: χ2(8, N = 157) = 17.69, p = .024, SWE: χ2(8, N = 89) = 26.16, p < .001. The lower alpha level in AAE as compared to SWE is likely due to a wider (but not different) distribution of children classified in the five subgroups within AAE as compared to SWE.

Figure 1.

A bar graph for the African American English category depicts the data for the distribution in terms of the percent of children by subgroups. The data for the DLD children are as follows. 0: 7 percent. 0.01 to 0.05: 73 percent. 0.06 to 0.10: 11 percent. 0.11 to 0.15: 8 percent. 0.16 to 0.20: 1 percent. The data for the TD6 children are as follows. 0: 5 percent. 0.01 to 0.05: 48 percent. 0.06 to 0.10: 28 percent. 0.11 to 0.15: 13 percent. 0.16 to 0.20: 4 percent. The data for the TD4 children are as follows. 0: 19 percent. 0.01 to 0.05: 56 percent. 0.06 to 0.10: 19 percent. 0.11 to 0.15: 3 percent. 0.16 to 0.20: 3 percent.

Open in a new tab

Distribution of children by subgroup based on rates of conjoined clauses: African American English (AAE). DLD = developmental language disorder; TD4 = typically developing 4-year-olds; TD6 = typically developing 6-year-olds.

Figure 2.

A bar graph for the Southern White English category depicts the data for the distribution in terms of the percent of children by subgroups. The data for DLD children are as follows. 0: 2 percent. 0.01 to 0.05: 76 percent. 0.06 to 0.10: 16 percent. 0.11 to 0.15: 2 percent. 0.16 to 0.20: 2 percent. The data for the TD6 children are as follows. 0.01 to 0.05: 41 percent. 0.06 to 0.1: 43 percent. 0.11 to 0.15: 11 percent. 0.16 to 0.20: 4 percent. The data for the TD4 children are as follows. 0: 16 percent. 0.01 to 0.05: 79 percent. 0.06 to 0.10: 5 percent.

Open in a new tab

Distribution of children by subgroup based on rates of conjoined clauses: Southern White English (SWE). DLD = developmental language disorder; TD4 = typically developing 4-year-olds; TD6 = typically developing 6-year-olds.

Finally, using the 233 language samples with at least one utterance with two independent clauses conjoined, we examined the types of conjunctions the children produced by dividing the sum of each type by the total for each child. The conjunction and was produced most often, followed by but and so, and these findings were observed for both dialects (see Figures 3 and 4). The conjunctions yet and nor were never produced, and the conjunction for was produced twice, once by an AAE TD4 child, And he was holding them for he can't fall in there, and once by an AAE DLD child, No, pull him back (from the, from, he can't, for he can/'t, um) for he can't fall in the water. Although these two utterances were elicited by children who lived in different rural communities, they were spoken while the children were talking about the same picture of two boys and a man fishing. It is also telling that the utterance by the child with DLD presents clause formulation difficulties as indicated with mazing, whereas the utterance by the TD4 child does not.

Figure 3.

A pie chart depicts the proportion of conjunction types in African American English. The data are as follows. And: 76 percent. But: 10 percent. Or: 4 percent. So: 10 percent.

Open in a new tab

Proportion of each type of conjunction: African American English (AAE).

Figure 4.

A pie chart depicts the proportion of conjunction types in Southern White English. The data are as follows. And: 70 percent. But: 20 percent. Or: 1 percent. So: 9 percent.

Open in a new tab

Proportion of each type of conjunction: Southern White English (SWE).

Effects of Conjoined Independent Clauses on MLU

To examine the impact of separating conjoined independent clauses into single clause utterances on the MLU values, we calculated a difference score (MLU from sample without clauses separated minus MLU from sample with clauses separated) for each child. We did this using MLU in morphemes and MLU-W; however, the results did not differ, and correlations between the two calculations of MLU were extremely high (r = .99, p < .001). Here we report the results for MLU in morphemes (for findings with MLU-W, see Supplemental Material S2). As shown in Table 4, the MLU difference score varied by group, F(2, 240) = 23.61, p < .001, ηp2 = .16, but not by dialect, p = .36, ηp2 = .004. LSD follow-up of the group effect showed the MLU difference score to be smaller for the DLD and TD4 groups than for the TD6 group.

Table 4.

MLU in morphemes in samples without and with conjoined independent clauses separated and mean difference score.

MLU in morphemes AAE
 SWE
DLD TD6 TD4 DLD TD6 TD4
Without clauses separated 5.25 (0.99) 6.36 (1.15) 5.30 (0.94) 4.99 (0.80) 6.34 (1.08) 4.85 (0.55)
With clauses separated 5.05 (0.81) 5.95 (0.92) 5.08 (0.78) 4.81 (0.66) 5.90 (0.85) 4.74 (0.48)
Difference score 0.20 (0.24) 0.41 (0.32) 0.22 (0.25) 0.18 (0.21) 0.44 (.31) 0.11 (0.10)
Open in a new tab

Note. Means reported first followed by standard deviations in parentheses. AAE = African American English; SWE = Southern White English; DLD = developmental language disorder; TD6 = typically developing 6-year-olds; TD4 = typically developing 4-year-olds.

Correlations

The final analysis involved correlations using the 195 samples from the DLD and TD6 groups. As shown in Table 5 and for both dialects, the rates of conjoined independent clauses and MLU in morphemes—calculated with and without conjoined clauses separated—were correlated to language test z scores but not to maternal education, nonverbal IQ z score, or density of nonmainstream forms within language samples. These findings further demonstrate equity of transcription outcome across the two dialects studied.

Table 5.

Correlations between variables.

Variable Rate of conjoined clauses within utterances MLU without clauses separated MLU with clauses separated
AAE
 MEDa −.01 .11 .14
 Nonmainstream form densityb .04 −.05 −.04
 Nonverbal IQc −.02 .10 .13
 Syntaxd .36** .49** .48**
 Vocabularye .32** .46** .46**
SWE
 MED .08 −.03 −.05
 Nonmainstream form density .01 −.13 −.15
 Nonverbal IQ .13 −.10 −.13
 Syntax .34** .53** .55**
 Vocabulary .37** .52** .53**
Open in a new tab

Note. MLU = mean length of utterance; AAE = African American English; ME = maternal education; SWE = Southern White English.

a

Highest level of MED reported in years, e.g., 12 = high school degree.

b

Proportion of utterances in language sample with a nonmainstream dialectal form.

c

Nonverbal IQ reflects z scores from the Primary Test of Nonverbal Intelligence (Ehrler & McGhee, 2008), the Columbia Mental Maturity Scale (Burgmeister et al., 1972), or subtests from the Leiter International Performance Scale–Revised (Roid & Miller, 1998).

d

Syntax measure reflects z scores from either the syntax subtest of the Test of Language Development–Primary: Second Edition (Newcomer & Hammill, 1988) or Third Edition (Hammill & Newcomer, 1997) or the syntax subtest of the Diagnostic Evaluation of Language Variation: Norm Referenced (Seymour et al., 2005).

e

Vocabulary measure reflects z scores from the Peabody Picture Vocabulary Test–Revised (Dunn & Dunn, 1981), Third Edition (Dunn & Dunn, 1997), or Fourth Edition (Dunn & Dunn, 2007).

**

p < .01.

Discussion

Transcription decisions for conjoined independent clauses within language samples vary across professionals. Some transcribe these clauses as two separate utterances, whereas others conjoin them within a single utterance. As an inquiry into equitable practice, we examined the number (i.e., rate) of conjoined independent clauses produced by children and the impact of separating these clauses within utterances on measures of MLU by a child's English dialect, clinical status, and age. The children's rate of conjoined independent clauses varied by clinical status and age but not by dialect. Also, effect sizes of the clinical differences and age differences were similar across the two dialects, and children in both dialects conjoined independent clauses primarily with and, but, and so. Similarities between the two dialects were further confirmed by examining the impact of separating the conjoined independent clauses on measures of MLU, which again revealed clinical status effects and age effects but no dialect effects. Finally, within both dialects, rates of conjoined independent clauses and MLU values (with and without the conjoined clauses separated) were similarly correlated to each other and the language test scores.

The findings support the practice of transcribing conjoined independent clauses across dialects of English in the same way—by either conjoining them or separating them within utterances. Recall that all results related to the children's dialects were null. The children's MLU values from the two transcription methods also were highly correlated (r = .99, p < .001). Nevertheless, the results indicated that children's abilities to conjoin independent clauses within utterances and the impact of these clauses on their MLUs led to clinical status effects and age effects within both dialects. These effects can easily become hidden when transcribers separate conjoined clauses within utterances, as is done when using T-unit or C-unit conventions for utterance boundaries. Adding a code to utterances with conjoining conjunctions (even when the conjoined clauses are transcribed as separate units) is a possible solution that could be implemented when utterances are transcribed as T- or C-units.

For large repositories of language samples such as those associated with SALT, CHAT, and SUGAR, the findings also support either transcription method, although it is important for the developers of repositories to ask contributors to make explicit their transcription methods, especially if the repositories are to provide normative values for measures like MLU. Alternatively, developers of large repositories could move away from generating normative summaries and instead maintain language samples for public use, with the caveat that they require additional coding and analysis by others who choose to use them for normative purposes.

The clinical status effects and age effects documented in the current study also demonstrate the across-dialect appropriateness of measuring children's conjoining of independent clauses within clinical practice. Recall that in Seymour et al.'s (1998) AAE study, children's use of conjunctions did not differ between the DLD and TD language samples, but here they did. An important difference between the two studies relates to the type of conjunction measured. As far as we can tell, Seymour et al. (1998) considered all and and but conjunctions, including those that conjoined nouns and verbs within phrases, whereas we only considered conjunctions that conjoined independent clauses. Given the different findings in these two studies and the clinical status differences and age differences found in the current study, we recommend that clinical practice focus more on children's development of clausal coordination as opposed to phrasal coordination. Strategically focusing on clausal coordination may also help clinicians move their assessments and treatment goals toward children's use of complex syntax as compared to morphology, which is well known to yield dialectal differences when language samples are evaluated (e.g., Oetting et al., 2021).

Arndt and Schuele (2013) provide a review of studies focused on children's development of complex syntax, noting repeatedly that complex syntax begins emerging soon after children begin combining words (for a similar recommendation, see Owens et al., 2024). Thus, across dialects of English, treatment activities focused on complex syntax can be initiated when children with DLD are young. A systematic review of interventions focused on complex syntax also shows positive changes in children's language abilities, especially when clinicians use scaffolding techniques (e.g., recasts, expansions, modeling) as compared to imitation tasks (Wisman Weil & Schuele, 2019).

Other measures of syntax from language samples need similar across-dialect validation. Oetting and Newkirk (2008) is an example of this type of study. Using 146 of the AAE and SWE language samples studied here, clinical status differences but not dialectal differences were documented in the children's productions of relative clauses. With the dialects combined, those with DLD produced not only lower rates of utterances with relative clauses but also lower proportions of overt relative forms within these clauses compared to those classified as TD.

In the future, it will also be important to explore the diagnostic accuracy of clausal coordination in combination with other measures of syntax, such as rates of relative clauses. In the current study of conjoined independent clauses, the effect sizes associated with the clinical group differences were small (d < .05), and with the dialects combined, these effect sizes were associated with low levels of diagnostic accuracy (65%; sensitivity [Se] = .78; specificity [Sp] = .55). These small effects and low levels of diagnostic accuracy indicate that a measure of clausal conjoining cannot be used in isolation. Identifying other measures, such as children's rates of relative clauses or their use of other types of clausal subordination, may generate larger effect sizes and higher rates of diagnostic accuracy. Future studies may also want to examine the diagnostic accuracy of syntax measures along with other types of measures unrelated to syntax. Ramos et al. (2022) provide an example of this type of work. Within a systematic review, these authors examined the diagnostic accuracy of measures spanning the domains of morphology, syntax, semantics, discourse, and pragmatics, and they included in their review studies of children who spoke mainstream and nonmainstream dialects of English in the United States, Canada, and Britain.

In addition to diagnostic accuracy studies, across-dialect studies of language acquisition are needed. This type of work is being pursued in studies with bilingual children. As an example, Castilla-Earls et al. (2023) examined the growth trajectories of Spanish articles, clitics, verbs, and subjunctive mood markers as produced by Spanish–English bilingual children with and without DLD. At the start of the study, the children were 5 years of age, and they were tested 3 times over a 2-year period. Among the observed findings, those with DLD were found to lag behind but show parallel growth to those classified as TD, a finding that has also been documented for monolingual GAE-speaking children. Studies such as this one need to be conducted with children who speak dialects of English that differ from GAE.

Limitations

The study was limited to language samples from 4- to 6-year-olds, which is a narrow age range when considering the much wider age range of children served by clinicians. The dialects represented in the language samples also reflected two spoken in the Southern United States. Cross-dialectal studies conducted in other areas of the country are needed to guide practice. The language samples were elicited using a conversational-based play context. Other types of language elicitation contexts, including narration, exposition, and persuasion, may lead to dialectal differences in children's conjoining preferences and/or their use of other syntactic structures.

Conclusions

Transcription of conjoined independent clauses within language samples varies across professionals. Some transcribe these clauses as two separate utterances, whereas others conjoin them within a single utterance. Findings from the current study support either transcription method, as well as the use of the same method when transcribing dialectally diverse, English language samples. Nevertheless, transcribing conjoined independent clauses as two separate utterances, as is done with T-units and C-units, reduces one's ability to detect syntactic differences between children with and without DLD and document syntactic growth as children age. Adding a code to T-units and C-units when clausal coordination is evident offers a solution to detect these differences. At least for the dialects studied here, the findings also support adding a measure of clausal coordination to clinical practice because finding similar magnitudes of clinical effects and age effects within the children's rates of conjoined independent clauses shows this measure of syntax to be appropriate for children learning various dialects of English.

Data Availability Statement

The data files analyzed during the current study are available from the corresponding author on reasonable request.

Supplementary Material

Supplemental Material S1. Procedures, language profiles, and decision processes used to classify children as DLD or TD.
LSHSS-55-870-s001.pdf (51.2KB, pdf)
Supplemental Material S2. Results from a series of 2 (dialect) × 3 (group) ANOVAs, with group main effects followed by least significance difference t-test procedures.
LSHSS-55-870-s002.pdf (33.4KB, pdf)

Acknowledgments

The data were collected with support from National Institute on Deafness and Other Communication Disorders Grants DC03609 (awarded to Janna Oetting) and DC009811 (awarded to Janna Oetting, Michael Hegarty, and Janet McDonald). Appreciation is extended to Jessica Berry, Lesli Cleveland, Kyomi Gregory-Martin, Janice Horohov, Sonja Pruitt-Lord, April Garrity, and Andy Rivière who helped create the language sample archive and corresponding test measures through their dissertations; the many graduate and undergraduate students who collected, transcribed, and coded the language samples; and C. Melanie Schuele who provided guidance on the coding of conjunctions and manuscript editing.

Funding Statement

The data were collected with support from National Institute on Deafness and Other Communication Disorders Grants DC03609 (awarded to Janna Oetting) and DC009811 (awarded to Janna Oetting, Michael Hegarty, and Janet McDonald).

Footnotes

1

Various labels have been used to refer to childhood language impairment, including “specific language impairment” and “DLD”; in the current study and regardless of the term used in the studies reviewed or our original studies, we use the term “DLD” to promote uniformity within the field.

2

All samples from the original studies were included except 15 from AAE-speaking TD children classified as reared in poverty within Pruitt and Oetting (2009). This group was excluded here because we did not have a SWE dialect group who were recruited using the same inclusionary criteria (i.e., caregivers with less than a high school degree, enrolled in a low-performing school, and with PPVT standard scores below 90).

References

  1. Altenberg, E. P., Roberts, J. A., & Scarborough, H. S. (2018). Young children's structure production: A revision of the Index of Productive Syntax. Language, Speech, and Hearing Services in Schools, 49(4), 995–1008. 10.1044/2018_LSHSS-17-0092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arndt, K. B., & Schuele, C. M. (2013). Multiclausal utterances aren't just for big kids: A framework for analysis of complex syntax production in spoken language of preschool- and early school-age children. Topics in Language Disorders, 33(2), 125–139. 10.1097/TLD.0b013e31828f9ee8 [DOI] [Google Scholar]
  3. Bernstein Ratner, N., & MacWhinney, B. (2023). Assessment and therapy goal planning using free computerized language analysis software. Perspectives of the ASHA Special Interest Groups, 8(1), 19–31. 10.1044/2022_PERSP-22-00156 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Berry, J. R., & Oetting, J. B. (2017). Dialect variation of copula and auxiliary verb BE: African American English–speaking children with and without Gullah/Geechee heritage. Journal of Speech, Language, and Hearing Research, 60(9), 2557–2568. 10.1044/2017_JSLHR-L-16-0120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brea-Spahn, M. R., & Fabiano-Smith, L. (2023). Reflect, reframe, and re-imagine theory and practice in speech, language, and hearing sciences: Introduction to the forum. Language, Speech, and Hearing Services in Schools, 54(3), 671–674. 10.1044/2023_LSHSS-23-00048 [DOI] [PubMed] [Google Scholar]
  6. Brimo, D., & Hall-Mills, S. (2019). Adolescents' production of complex syntax in spoken and written expository and persuasive genres. Clinical Linguistics & Phonetics, 33(3), 237–255. 10.1080/02699206.2018.1504987 [DOI] [PubMed] [Google Scholar]
  7. Burgmeister, B., Blum, H., & Lorge, I. (1972). Columbia Mental Maturity Scale. The Psychological Corp.
  8. Bùi, M., & Leonard, L. B. (2024). A further look at two grammatical measures from children's language samples and their contribution to the diagnostic process. Clinical Linguistics & Phonetics. Advance online publication. 10.1080/02699206.2024.2305120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Castilla-Earls, A., Ronderos, J., & Fitton, L. (2023). Spanish bilingual morphosyntactic development in bilingual children with and without developmental language disorder: Articles, clitics, verbs, and the subjunctive mood. Journal of Speech, Language, and Hearing Research, 66(12), 4678–4698. 10.1044/2023_JSLHR-23-00091 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Champion, T. B., & McCabe, A. (2015). Narrative structures of African American children: Commonalities and differences. In Lanehart S. (Ed.). The Oxford handbook of African American language (pp. 492–511). Oxford University Press. [Google Scholar]
  11. Cleveland, L. H., & Oetting, J. B. (2013). Children's marking of verbal –s by nonmainstream English dialect and clinical status. American Journal of Speech-Language Pathology, 22(4), 604–614. 10.1044/1058-0360(2013/12-0122) [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Coughler, C., Bardell, T., Schouten, M., Smith, K., & Archibald, L. M. D. (2024). Comparing spoken versus iPad-administered versions of a narrative retell assessment tool in a practice-based research partnership. Language, Speech, and Hearing Services in Schools. Advance online publication. 10.1044/2024_LSHSS-23-00022 [DOI] [PubMed] [Google Scholar]
  13. Dunn, L., & Dunn, D. (1981). Peabody Picture Vocabulary Test–Revised. American Guidance Service. [Google Scholar]
  14. Dunn, L., & Dunn, D. (1997). Peabody Picture Vocabulary Test–Third Edition. American Guidance Service. [Google Scholar]
  15. Dunn, L., & Dunn, D. (2007). Peabody Picture Vocabulary Test–Fourth Edition. Pearson. [Google Scholar]
  16. Ebert, K. D., & Pham, G. (2017). Synthesizing information from language samples and standardized tests in school-age bilingual assessment. Language, Speech, and Hearing Services in Schools, 48(1), 42–55. 10.1044/2016_LSHSS-16-0007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ehrler, D. J., & McGhee, R. L. (2008). Primary Test of Nonverbal Intelligence. Pro-Ed. [Google Scholar]
  18. Eisenberg, S. L., & Guo, L.-Y. (2013). Differentiating children with and without language impairment based on grammaticality. Language, Speech, and Hearing Services in Schools, 44(1), 20–31. 10.1044/0161-1461(2012/11-0089) [DOI] [PubMed] [Google Scholar]
  19. Eisenberg, S. L., & Guo, L.-Y. (2016). Using language sample analysis in clinical practice: Measures of grammatical accuracy for identifying language impairment in preschool and school-aged children. Seminars in Speech and Language, 37(2), 106–116. 10.1055/s-0036-1580740 [DOI] [PubMed] [Google Scholar]
  20. Finestack, L. H., Payesteh, B., Disher, J. R., & Julien, H. M. (2014). Reporting child language sampling procedures. Journal of Speech, Language, and Hearing Research, 57(6), 2274–2279. 10.1044/2014_JSLHR-L-14-0093 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Fox, C., Jones, S., Gillam, S. L., Israelsen-Augenstein, M., Schwartz, S., & Gillam, R. B. (2022) Automated progress-monitoring for Literate Language Use in Narrative Assessment (LLUNA). Frontiers in Psychology, 13, Article 894478. 10.3389/fpsyg.2022.894478 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Francois, I., Lapka, S. M., Bernstein Ratner, N., & Mills, M. (2023). Assessing for developmental language disorder in the context of African American English. EBP Briefs, 16(2), 1–8. https://hdl.handle.net/10657/13644 [Google Scholar]
  23. Garbarino, J., Bernstein Ratner, N., & McWhinney, B. (2020). Use of computerized language analysis to assess child language. Language, Speech, and Hearing Services in Schools, 51(2), 504–506. 10.1044/2020_LSHSS-19-00118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Gardner-Neblett, N. (2024). Becoming fictional storytellers: African American children's oral narrative development in early elementary school. Child Development. Advance online publication. 10.1111/cdev.14075 [DOI] [PubMed] [Google Scholar]
  25. Garrity, A. W., & Oetting, J. B. (2010). Auxiliary BE production by African American English–speaking children with and without specific language impairment. Journal of Speech, Language, and Hearing Research, 53(5), 1307–1320. 10.1044/1092-4388(2010/09-0016) [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Garivaldo, B., & Fabiano-Smith, L. (2023). Reframing bilingual acquisition and theory: An insider perspective through a translanguaging lens. Language, Speech, and Hearing Services in Schools, 54(3), 765–780. 10.1044/2023_LSHSS-22-00136 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Goldman, R., & Fristoe, M. (1986). Goldman-Fristoe Test of Articulation. American Guidance Service. [Google Scholar]
  28. Goldman, R., & Fristoe, M. (2000). Goldman-Fristoe Test of Articulation–Second Edition. American Guidance Service. [Google Scholar]
  29. Guo, L., & Eisenberg, S. (2015). Sample length affects the reliability of language sample measures in 3-year-olds: Evidence from parent-elicited conversational samples. Language, Speech, and Hearing Services in Schools, 46(2), 141–153. 10.1044/2015_LSHSS-14-0052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Guo, L., Eisenberg, S., Bernstein Ratner, N., & MacWhinney, B. (2018). Is putting SUGAR (Sampling Utterances of Grammatical Analysis Revised) into language sample analysis a good thing? A response to Pavelko and Owens (2017). Language, Speech, and Hearing Services in Schools, 49(3), 622–627. 10.1044/2018_LSHSS-17-0084 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Guo, L., Eisenberg, S., Schneider, P., & Spencer, L. (2020). Finite verb morphology composite between age 4 and age 9 for the Edmonton Narrative Norms Instrument: Reference data and psychometric properties. Language, Speech, and Hearing Services in Schools, 51(1), 128–143. 10.1044/2019_LSHSS-19-0028 [DOI] [PubMed] [Google Scholar]
  32. Guo, L., & Schneider, P. (2016). Differentiating school-aged children with and without language impairment using tense and grammaticality measures from a narrative task. Journal of Speech, Language, and Hearing Research, 59(2), 317–329. 10.1044/2015_JSLHR-L-15-0066 [DOI] [PubMed] [Google Scholar]
  33. Hammill, D., & Newcomer, P. (1997). Test of Language Development–Primary:3. Pro-Ed. [DOI] [PubMed] [Google Scholar]
  34. Heilmann, J., Nockerts, A., & Miller, J. F. (2010). Language sampling: Does the length of the transcript matter? Language, Speech, and Hearing Services in Schools, 41(4), 393–404. 10.1044/0161-1461(2009/09-0023) [DOI] [PubMed] [Google Scholar]
  35. Hendricks, A. E., & Adlof, S. M. (2020). Production of morphosyntax within and across different dialects of American English. Journal of Speech, Language, and Hearing Research, 63(7), 2322–2333. 10.1044/2020_JSLHR-19-00244 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Horton, R., Johnson, V., & Koonce, N. (2018). From here to there and back again: Forty years of research on African American children's language. Perspectives of the ASHA Special Interest Groups, 3(14), 57–70. 10.1044/persp3.SIG14.57 [DOI] [Google Scholar]
  37. Horton-Ikard, R., Weismer, S. E., & Edwards, C. (2005). Examining the use of standard language production measures in the language samples of African-American toddlers. Journal of Multilingual Communication Disorders, 3(3), 169–182. 10.1080/14769670500170768 [DOI] [Google Scholar]
  38. Hyter, Y. D., & Salas-Provance, M. B. (2023). Culturally responsive practices in speech, language, and hearing sciences (2nd ed.). Plural. [Google Scholar]
  39. Jackson, S. C., & Roberts, J. E. (2001). Complex syntax production of African American preschoolers. Journal of Speech, Language, and Hearing Research, 44(5), 1083–1096. 10.1044/1092-4388(2001/086) [DOI] [PubMed] [Google Scholar]
  40. Kapantzoglou, M., Gerasimos, F., & Restrepo, A. M. (2017). Language sample analysis and elicitation technique effects in bilingual children with and without language impairment. Journal of Speech, Language, and Hearing Research, 60(10), 2852–2864. 10.1044/2017_JSLHR-L-16-0335 [DOI] [PubMed] [Google Scholar]
  41. Lee, L. L. (1974). Developmental sentence analysis. Northwestern University Press. [Google Scholar]
  42. Lee, L. L., & Canter, S. M. (1971). Developmental sentence scoring: A clinical procedure for estimating syntactic development in children's spontaneous speech. Journal of Speech and Hearing Disorders, 36(3), 315–340. 10.1044/jshd.3603.315 [DOI] [PubMed] [Google Scholar]
  43. Lee, R., & Oetting, J. B. (2014). Zero marking of past tense in child African American English. Perspectives on Language Learning and Education, 21(4), 173–181. 10.1044/lle21.4.173 [DOI] [Google Scholar]
  44. Liebenberg, P., van der Linde, J., Schimper, I., de Wet, F., Graham, M., & Bornman, J. (2023). Describing the spoken language skills of typically developing Afrikaans-speaking children using language sample analysis: A pilot study. Language, Speech, and Hearing Services in Schools, 54(2), 518–534. 10.1044/2022_LSHSS-22-00077 [DOI] [PubMed] [Google Scholar]
  45. Loban, W. (1976). Language development: Kindergarten through grade twelve. NCTE Committee on Research Report No. 18. National Council of Teachers of English. https://eric.ed.gov/?id=ED128818 [Google Scholar]
  46. MacWhinney, B. (2000). The CHILDES project: Tools for analyzing talk (3rd ed.). Erlbaum. [Google Scholar]
  47. Matiasovitsová, K., Ĉechová, P., Sláma, J., Homolková, K., & Smolik, F. (2024). Mean length of utterance in Czech toddlers: Validity estimates and comparison of words, morphemes, and syllables. Journal of Speech, Language, and Hearing Research, 67(3), 837–852. 10.1044/2023_JSLHR-23-00251 [DOI] [PubMed] [Google Scholar]
  48. Miller, J. F., & Iglesias, A. (2024). Systematic Analysis of Language Transcripts (Version 24). SALT Software, LLC. [Google Scholar]
  49. Mills, M., 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. 10.1044/2016_AJSLP-15-0157 [DOI] [PubMed] [Google Scholar]
  50. Newcomer, P., & Hammill, D. (1988). Test of Language Development–Primary: Second Edition. Pro-Ed. [Google Scholar]
  51. Newkirk-Turner, B. L., Horton, R., & Stockman, I. J. (2015). Language acquisition in the African American child: Prior to age four. In Lanehart S. (Ed.), The Oxford handbook of African American English (pp. 439–453). Oxford Press. [Google Scholar]
  52. Oetting, J. B. (2005). Assessing language in children who speak a nonmainstream dialect of English. In Ball M. (Ed.), Clinical sociolinguistics (pp. 180–192). Blackwell. 10.1002/9780470754856.ch14 [DOI] [Google Scholar]
  53. Oetting, J. B., Berry, J. R., Gregory, K. D., Rivière, A. M., & McDonald, J. (2019). Specific language impairment in African American English and Southern White English: Measures of tense and agreement with dialect-informed probes and strategic scoring. Journal of Speech, Language, and Hearing Research, 62(9), 3443–3461. 10.1044/2019_JSLHR-L-19-0089 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Oetting, J. B., Cantrell, J., & Horohov, J. (1999). A study of specific language impairment (SLI) in the context of non-standard dialect. Clinical Linguistics & Phonetics, 13(1), 25–44. 10.1080/026992099299220 [DOI] [Google Scholar]
  55. Oetting, J. B., & Horohov, J. (1997). Past-tense marking by children with and without specific language impairment. Journal of Speech, Language, and Hearing Research, 40(1), 62–74. 10.1044/jslhr.4001.62 [DOI] [PubMed] [Google Scholar]
  56. Oetting, J. B., & McDonald, J. L. (2001). Nonmainstream dialect use and specific language impairment. Journal of Speech, Language, and Hearing Research, 44(1), 207–223. 10.1044/1092-4388(2001/018) [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Oetting J. B., & McDonald, J. L. (2002). Methods for characterizing participants' nonmainstream dialect use in child language research. Journal of Speech, Language, and Hearing Research, 45(3), 505–518. 10.1044/1092-4388(2002/040) [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Oetting, J. B., McDonald, J. L., Seidel, C., & Hegarty, M. (2016). Sentence recall by children with SLI across two nonmainstream dialects of English. Journal of Speech, Language, and Hearing Research, 59(1), 183–194. 10.1044/2015_JSLHR-L-15-0036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Oetting, J. B., & Newkirk, B. L. (2008). Subject relatives by children with and without SLI across different dialects of English. Clinical Linguistics & Phonetics, 22(2), 111–125. 10.1080/02699200701731414 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Oetting, J. B., Newkirk, B. L., Hartfield, L. R., Wynn, C. G., Pruitt, S. L., & Garrity, A. W. (2010). Index of Productive Syntax for children who speak African American English. Language, Speech, and Hearing Services in Schools, 41(3), 328–339. 10.1044/0161-1461(2009/08-0077) [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Oetting, J. B., Rivière, A. M., Berry, J. R., Gregory, K. D., Villa, T. M., & McDonald, J. L. (2021). Marking of tense and agreement in language samples by children with and without specific language impairment in African American English and Southern White English: Evaluation of scoring approaches and cut scores across structures. Journal of Speech, Language, and Hearing Research, 64(2), 491–509. 10.1044/2020_JSLHR-20-00243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Owens, R. E., Jr., Pavelko, S. L., & Hahs-Vaughn, D. (2024). Growth of complex syntax: Coordinate and subordinate clause use in elementary school–aged children. Language, Speech, and Hearing Services in Schools. Advance online publication. 10.1044/2024_LSHSS-23-00102 [DOI] [PubMed] [Google Scholar]
  63. Overton, C., Baron, T., Pearson, B. Z., & Bernstein Ratner, N. (2021). Using free computer-assisted language sample analysis to evaluate and set treatment goals for children who speak African American English. Language, Speech, and Hearing Services in Schools, 52(1), 31–50. 10.1044/2020_LSHSS-19-00107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Padia, L. (2023). Toward a model of reciprocal carryover: Learning from communication systems of families of nonspeaking bilingual children. Language, Speech, and Hearing Services in Schools, 54(3), 716–728. 10.1044/2023_LSHSS-22-00145 [DOI] [PubMed] [Google Scholar]
  65. Pavelko, S. L., & Owens, R. E., Jr. (2017). Sampling Utterances and Grammatical Analysis Revised (SUGAR): New normative values for language sample analysis measures. Language, Speech, and Hearing Services in Schools, 48(3), 197–215. 10.1044/2017_LSHSS-17-0022 [DOI] [PubMed] [Google Scholar]
  66. Pavelko, S. L., & Owens, R. E., Jr. (2023). A sweet tutorial to the SUGAR method of language sampling. Perspectives of the ASHA Special Interest Groups, 8(1), 32–49. 10.1044/2022_PERSP-22-00134 [DOI] [Google Scholar]
  67. Pavelko, S. L., Price, L. R., & Owens, R. E., Jr. (2020). Revisiting reliability: Using Sampling Utterances and Grammatical Analysis Revised (SUGAR) to compare 25- and 50-utterance language samples. Language, Speech, and Hearing Services in Schools, 51(3), 778–794. 10.1044/2020_LSHSS-19-00026 [DOI] [PubMed] [Google Scholar]
  68. Pearce, W. M., & Flannagan, K. (2024). Associations among oral narrative language measures for Indigenous and non-Indigenous Australian children in their first year of school. In Han W. & Brebner C. (Eds.), Typical and atypical language development in cultural and linguistic diversity (pp. 152–171). Routledge. [Google Scholar]
  69. Pezold, M., Imgrund, C. M., & Storkel, H. L. (2020). Using computer programs for language sample analysis. Language, Speech, and Hearing Services in Schools, 51(1), 103–114. 10.1044/2019_LSHSS-18-0148 [DOI] [PubMed] [Google Scholar]
  70. Pruitt, S., & Oetting, J. B. (2009). Past tense marking by African American English–speaking children reared in poverty. Journal of Speech, Language, and Hearing Research, 52(1), 2–15. 10.1044/1092-4388(2008/07-0176) [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Pruitt, S. L., Oetting, J. B., & Hegarty, M. (2011). Passive participle marking by African American English–speaking children reared in poverty. Journal of Speech, Language, and Hearing Research, 54(2), 598–607. 10.1044/1092-4388(2010/09-0134) [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Ramos, M. N., Collins, P., & Peña, E. D. (2022). Sharpening our tools: A systematic review to identify diagnostically accurate language sample measures. Journal of Speech, Language, and Hearing Research, 65(10), 3890–3907. 10.1044/2022_JSLHR-22-00121 [DOI] [PubMed] [Google Scholar]
  73. Retherford, K., Schreiber, L., & Jarzynski, R. (2019). Guide to analysis of language transcripts (4th ed.). Pro-Ed. [Google Scholar]
  74. Rice, M. L. (2020). Clinical lessons from studies of children with specific language impairment. Perspectives of the ASHA Special Interest Groups, 5(1), 12–29. 10.1044/2019_PERSP-19-00011 [DOI] [Google Scholar]
  75. Rice, M. L., Smolik, F., Perpich, D., Thompson, T., Rytting, N., & Blossom, M. (2010). Mean length of utterance levels in 6-month intervals for children 3 to 9 years with and without language impairments. Journal of Speech, Language, and Hearing Research, 53(2), 333–349. 10.1044/1092-4388(2009/08-0183) [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Rice, M. L., & Wexler, K. (2005). Test of Early Grammatical Impairment. The Psychological Corporation. [Google Scholar]
  77. Rivière, A. M., Oetting, J. B., & Roy, J. (2018). Effects of specific language impairment on a contrastive dialect structure: The case of infinitival TO across various nonmainstream dialects of English. Journal of Speech, Language, and Hearing Research, 61(8), 1989–2001. 10.1044/2018_JSLHR-L-17-0209 [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Roid, G., & Miller, L. (1998). Leiter International Performance Scale–Revised. Stoelting. [Google Scholar]
  79. Rudolph, J. M., Dollaghan, C. A., & Crotteau, S. (2019). The finite verb morphology composite: Values from a community sample. Journal of Speech, Language, and Hearing Research, 62(6), 1813–1822. 10.1044/2019_JSLHR-L-18-0437 [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Seymour, H. N., Bland-Stewart, L., & Green, L. J. (1998). Difference versus deficit in child African American English. Language, Speech, and Hearing Services in the Schools, 29(2), 96–108. 10.1044/0161-1461.2902.96 [DOI] [PubMed] [Google Scholar]
  81. Seymour, H. N., Roeper, T., & de Villiers, J. (2005). Diagnostic Evaluation of Language Variation–Norm Referenced. Ventris Learning. [Google Scholar]
  82. Shirley, R., Silverman, C., Tran, C., & Nippold, M. A. (2024). Narrative discourse in youth offenders: Examining individual differences. American Journal of Speech-Language Pathology, 33(3), 1193–1208. 10.1044/2023_AJSLP-23-00240 [DOI] [PubMed] [Google Scholar]
  83. Soto-Boykin, X., Brea-Spahn, M. R., Perez, S., & McKenna, M. A. (2023). A critical analysis of state-level policies impacting racialized emergent bilinguals suspected or labeled as dis/abled. Language, Speech, and Hearing Services in Schools, 54(3), 729–745. 10.1044/2023_LSHSS-22-00137 [DOI] [PubMed] [Google Scholar]
  84. Southwood, F., & Russell, A. F. (2004). Comparison of conversation, freeplay, and story generation as methods of language sample elicitation. Journal of Speech, Language, and Hearing Research, 47(2), 366–376. 10.1044/1092-4388(2004/030) [DOI] [PubMed] [Google Scholar]
  85. Souto, S. M., Leonard, L. B., & Deevy, P. (2014). Identifying risk for specific language impairment with narrow and global measures of grammar. Clinical Linguistics & Phonetics, 28(10), 741–756. 10.3109/02699206.2014.893372 [DOI] [PMC free article] [PubMed] [Google Scholar]
  86. Spencer, T. D., Tolentino, T. J., & Foster, M. E. (2023). Impact of discourse type and elicitation task on language sampling outcomes. American Journal of Speech-Language Pathology, 32(6), 2827–2845. 10.1044/2023_AJSLP-22-00365 [DOI] [PubMed] [Google Scholar]
  87. Stockman, I. J., Newkirk-Turner, B., Swartzlander, E., & Morris, L. (2016). Comparison of African American children's performances on a minimal competence core for morphosyntax and the Index of Productive Syntax. American Journal of Speech-Language Pathology, 25(1), 80–96. 10.1044/2015_AJSLP-14-0207 [DOI] [PubMed] [Google Scholar]
  88. Wilder, A., & Redmond, S. M. (2022). The reliability of short conversational language sample measures in children with and without developmental language disorder. Journal of Speech, Language, and Hearing Research, 65(5), 1939–1955. 10.1044/2022_JSLHR-21-00628 [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Wisman Weil, L., & Schuele, C. M. (2019). Complex syntax interventions for young children with language impairments. EBP Briefs, 13(5), 1–9. [Google Scholar]

Associated Data

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

Supplementary Materials

Supplemental Material S1. Procedures, language profiles, and decision processes used to classify children as DLD or TD.
LSHSS-55-870-s001.pdf (51.2KB, pdf)
Supplemental Material S2. Results from a series of 2 (dialect) × 3 (group) ANOVAs, with group main effects followed by least significance difference t-test procedures.
LSHSS-55-870-s002.pdf (33.4KB, pdf)

Data Availability Statement

The data files analyzed during the current study are available from the corresponding author on reasonable request.

Articles from Language, Speech, and Hearing Services in Schools are provided here courtesy of American Speech-Language-Hearing Association

RESOURCES