Abstract
Purpose
Proficiency with complex syntax is important for language and reading comprehension, and production of complex syntax begins to emerge shortly after a young child begins using two-word combinations. Complex syntax production in preschool children with hearing loss who use spoken language has been explored minimally. The purpose of this study was to compare complex syntax production of 4-year-old children with hearing loss to age-matched and language-matched peers with normal hearing.
Method
Seventy-two children completed a language assessment battery, including a structured language sample. Complex syntax density and number and accuracy of productions of particular types of complex syntax were compared across three groups: 4-year-old children with hearing loss, an age-matched group of children with normal hearing, and an mean length of utterance (MLU)–matched group of children with normal hearing.
Results
Children with hearing loss had lower complex syntax density and fewer correct productions of coordinated clauses, subordinate clauses, and simple infinitives than their age-matched, but not language-matched, peers. Furthermore, children with hearing loss had lower accuracy than the age-matched group on simple infinitives and lower accuracy than both the age- and MLU-matched groups on full propositional complements and subject relative clauses.
Conclusion
Children with hearing loss exhibit delays in complex syntax acquisition as compared to their same-age peers and disruptions in development on some complex structures as compared to MLU-matched, younger children.
Supplemental Material
Complex syntax acquisition is a major linguistic accomplishment during the preschool years. We use the term complex syntax here to refer to utterances that contain more than one verb phrase (Barako Arndt & Schuele, 2013). Complex syntax production at school entry is a robust predictor of reading comprehension in elementary school (Craig et al., 2003). Additionally, decreased ability in academic language, including complex syntax knowledge, results in increased risk of writing difficulties of fourth- and eighth-grade students (Truckenmiller & Petscher, 2019). Children with hearing loss have documented deficits in morphosyntax development (e.g., Werfel & Douglas, 2017), but their complex syntax acquisition has been rarely studied, particularly during the preschool years. The purpose of this study, therefore, was to compare complex syntax production of 4-year-old children with hearing loss to that of their age- and language-matched peers with normal hearing.
Complex Syntax Acquisition in Children With Normal Hearing
Children with normal hearing and typical language development begin using complex syntax structures spontaneously in spoken language well before their third birthday, shortly after they begin to combine words (Bloom & Capatides, 1987; Bloom et al., 1984; Diessel, 2004; Limber, 1973). For children with normal hearing and language disorders, this emergence of complex syntax production in spontaneous spoken language has been reported between ages 3 and 4 years (Schuele & Dykes, 2005), with delays that continue well into the school-age years (Owen & Leonard, 2006; Owen Van Horne & Lin, 2011).
Children With Normal Hearing and Typical Language
A seminal longitudinal study provides insight into the order of acquisition of complex syntax types by children with normal hearing and typical language development (Diessel, 2004). The earliest emerging types of complex syntax production in children include reduced infinitives (e.g., I wanna play), infinitival complement clauses (e.g., I want to play), participle complement clauses (e.g., I like playing outside), and coordinate clauses (I like dogs and mom likes cats). A few months after the initial emergence of complex syntax, additional types of complex syntax begin to emerge. These types include full propositional complements (e.g., I think he's nice), wh-complement clauses (e.g., I know what it is; I know where to go), let's clauses (e.g., Let's see what's inside), and subordinate clauses (I like dogs because they're funny). Finally, relative clauses (e.g., I saw the girl who won [subject relative clause]; The dog (that) I like is brown [object relative clause]; I saw him at the park (that) is new [oblique relative clause]) emerge. In this study, complex syntax began to emerge around the children's second birthdays, and the latest emerging types of interest (e.g., relative clauses) emerged during the 3-year-old year. The results of Diessel (2004) were broadly consistent with previous reports of complex syntax acquisition for children with typical development (e.g., Tyack & Gottsleben, 1986).
Children With Normal Hearing and Specific Language Impairment
Because children with hearing loss often have secondary language impairments (Geers et al., 2016; Halliday et al., 2017), it is helpful also to consider complex syntax for children with hearing loss in light of what is known about children with specific language impairment (SLI), the most common type of developmental language disorder. Children with SLI and children with hearing loss similarly exhibit not just delays but disruptions in morphosyntax acquisition (e.g., Rice, 2004; Werfel & Douglas, 2017). Delays can be identified by comparing a group of interest with their age-matched peers with typical development, and disrupted development can be identified when a group of interest differs in production of language structures from a comparison group that is matched on language skills, such as mean length of utterance (MLU; Rice, 2004). See Rice (2004) for a deeper discussion of delayed and disrupted development.
In a preschool child with normal hearing and SLI, Schuele and Dykes (2005) reported a similar overall order of emergence for complex syntax types as for children with normal hearing and typical language development; however, this emergence was delayed by approximately 1 year from what would be expected based on the Diessel (2004) data. Additionally, differences were seen in the order of emergence of coordinate clauses and participle clauses for the child with SLI, such that these emerged relatively later than was observed for the children with typical development, suggesting that children with SLI may have disrupted, as well as delayed, complex syntax acquisition. Schuele and Dykes reported that complex syntax began to emerge for this child with SLI when MLU was approximately 2.0, but it was not until the child's MLU exceeded 3.0 that a variety of complex syntax types emerged. This finding is consistent with previous reports of children with typical language acquisition (Tyack & Gottsleben, 1986).
Marinellie (2004) reports that delays in complex syntax production for children with normal hearing and SLI continue through the school-age years. In a language sample analysis of students in Grades 3 through 5, children with SLI used fewer complex syntax features than children with typical language, even when language samples were matched for utterance length. Children with SLI produced fewer relative clauses, coordinate clauses, and subordinate clauses than children with typical language; however, the groups did not differ on errored complex syntax attempts.
Syntax in Children With Hearing Loss
Although complex syntax has been minimally studied in children with hearing loss, research indicates that morphosyntax is an area of established difficulty for the population. Prior research reveals that children with hearing loss are at risk for having shorter utterances, as measured by MLU, as compared with their peers with normal hearing (Koehlinger et al., 2013; Werfel, 2018). These studies have consistently shown that there is a significant delay in utterance length in children with hearing loss compared to children with normal hearing. The overall shorter length of utterances in the spoken language of children with hearing loss serves as an indicator of overall syntactic weakness.
In addition to shorter MLU, research has also revealed specific syntax deficits exhibited by children with hearing loss. For example, Moeller et al. (2010) identified weaknesses in syntax of children with mild-to-moderate hearing loss compared to their age-matched peers with normal hearing using the measure of the Index of Productive Syntax (Scarborough, 1990). In particular, weaknesses were observed in the use of verb tense marking and wh-clauses; the authors suggest their findings are indicative of delays in complex syntax. Additionally, Werfel (2018) reported that preschool children with hearing loss exhibited difficulty in marking plurals, as well as past tense and third person singular verbs, compared to age-matched and language-matched children with normal hearing. Koehlinger et al. (2013) similarly reported deficits in marking verb morphology for 3- and 6-year-old children with hearing loss. Thus, there is evidence that a gap between children with hearing loss and children with normal hearing occurs with regard to morphosyntax.
It is reasonable to hypothesize (supported by Moeller et al., 2010) that deficits in morphosyntax for this population point to widespread difficulty with syntax. Therefore, the purpose of this study was to, for the first time to our knowledge, compare the complex syntax production of 4-year-old children with hearing loss to two groups of children with normal hearing: age-matched peers and language-matched peers. We hypothesized that the children with hearing loss would exhibit delays in complex syntax production and possible differences in productivity. For each research question, group differences between children with hearing loss and age-matched, but not language-matched, children with normal hearing would indicate a delay in complex syntax use. Group differences between children with hearing loss and both the age- and language-matched children with normal hearing would indicate disrupted production. Specifically, we asked the following research questions:
Do children with hearing loss differ from age-matched and/or language-matched children with normal hearing on broad measures of complex syntax? 1
Do children with hearing loss produce fewer accurate productions on specific types of complex syntax than age-matched and/or language-matched children with normal hearing?
Do children with hearing loss have lower percent accuracy on specific types of complex syntax than age-matched and/or language-matched children with normal hearing?
Method
This study was approved by the University of South Carolina Institutional Review Board, which served as the institutional review board of record for this multisite study. All participants' parents provided informed consent prior to study participation, and participants provided assent at each assessment session. The data reported here are part of a longitudinal study on early language and literacy acquisition in children with hearing loss (e.g., Werfel, 2017). In the larger study, children complete a language and literacy assessment battery every 6 months from ages 4 to 6 years and annually after first to fifth grade. In this article, we present data from one preschool session.
Participants
Seventy-two preschool children participated in this study. There were three groups: a group of preschool children with permanent bilateral hearing loss (CHL) and two control groups of preschool children with normal hearing (a group matched for chronological age [CNH-Age] and a group matched for MLU in morphemes [CNH-MLU]). Recruitment occurred through preschools and social media groups for both groups, as well as speech-language pathology/audiology clinics and schools for the deaf for the CHL group.
CHL Group
The CHL group was composed of 26 children with bilateral, permanent hearing loss (M age = 52.46 months, SD = 3.13, range: 48–59). All CHL used amplification and spoken English. Amplification type varied among participants: 13 CHL used bilateral hearing aids, 12 CHL used bilateral cochlear implants, two CHL were bimodal, and one CHL used a bone anchored hearing aid. Chi-square analysis indicated that spoken language abilities did not differ by amplification type, as measured by the Test of Early Language Development–Third Edition or Fourth Edition 2 Spoken Language Quotient (Hresko et al., 1999, 2018), as well as MLU and complex syntax density from spontaneous language samples (p = .271, .294, and .203, respectively). All participants in the CHL group had been diagnosed with permanent hearing loss by a certified audiologist, and all were enrolled in speech-language therapy at the time of study entry. See Table 1 for additional audiologic descriptive data.
Table 1.
Audiologic information for the children with permanent bilateral hearing loss group.
| Variable | Age at identification | Age at first hearing aid | Age at first implant | Duration of device use |
|---|---|---|---|---|
| M | 8.12 | 10.93 | 18.08 | 41.95 |
| SD | 12.08 | 12.17 | 9.39 | 11.68 |
| Range | 0–36 | 1.25–36 | 9–36 | 18–56 |
| Mdn | 1.5 | 3.5 | 15 | 47.5 |
| Degree of hearing loss a | ||||
| Mild – 1 | Mild to moderate – 1 | Mild to moderately severe – 2 | Moderate –1 | Moderate to severe – 1 |
| Moderately severe – 3 | Severe – 1 | Severe to profound – 1 | Profound – 11 | Not reported – 4 |
Note. All ages listed in months.
Degree of hearing loss that includes “to” and multiple degrees indicates that the child had a sloping loss that covered multiple degrees (e.g., mild loss in low-frequency sloping to moderately severe in high frequency).
Age-Matched Control Group
The age-matched control group consisted of 26 children with normal hearing (CNH-Age). Mean age in months of the CNH-Age group was 52.23 (SD = 3.22, range: 48–58) and did not differ from the CHL group (p = .956). Mean MLU of the CNH-Age group was 5.92 (SD = 1.17, range: 4.27–8.94), which was significantly higher than the CHL group (M = 4.25, SD = 1.44, range: 2.18–7.73, p < .001).
Language-Matched Control Group
The language-matched control group consisted of 20 children with normal hearing (CNH-MLU). Mean MLU of the CNH-MLU group was 4.38 (SD = 1.22, range: 2.62–7.86), which did not differ from MLU of the CHL group (p = .980). Mean age in months of the CNH-MLU group (M = 47.85, SD = 2.03, range: 45–52) was significantly lower than the CHL group (p < .001).
All children in the CNH-Age and CNH-MLU groups passed a bilateral hearing screening prior to study participation and achieved a standard score above 85 on the Test of Early Language Development. There was no overlap of participants across the two control groups.
Participants in all groups had nonverbal intelligence standard scores above 70, as measured by the Primary Test of Nonverbal Intelligence (Ehrler & McGhee, 2008). Additionally, all participants spoke English as their primary language and had no diagnoses known to affect language or cognition (e.g., autism, Down syndrome) other than hearing loss for the CHL group, determined by parent report. Distributions of gender, ethnicity, and race did not differ across the groups (p = .523, .076, and .767, respectively). As shown in Table 2, the groups did not differ relative to maternal education, nonverbal intelligence, or speech sound production, but differences were observed on all measures of spoken language.
Table 2.
Descriptive information by group.
| Variable | CHL (n = 26) M (SD) |
CNH-Age (n = 26) M (SD) |
CNH-MLU (n = 20) M (SD) |
p |
|---|---|---|---|---|
| Age in months a | 52.46 (3.13) | 52.23 (3.22) | 47.85 (2.03) | < .001 |
| Maternal education in years | 16.02 (2.76) | 17.57 (1.71) | 16.95 (2.74) | .172 |
| Nonverbal intelligence | 109.69 (15.07) | 112.64 (10.06) | 114.70 (10.12) | .464 |
| Speech sound production | 90.31 (10.92) | 94.65 (8.87) | 91.60 (8.80) | .343 |
| Omnibus spoken language b | 93.73 (19.31) | 113.62 (11.00) | 112.65 (10.62) | < .001 |
| Expressive vocabulary b | 102.15 (13.40) | 116.24 (9.09) | 116.45 (11.56) | < .001 |
| Receptive vocabulary b | 101.12 (15.29) | 114.31 (9.42) | 112.16 (12.28) | .002 |
| Morphosyntax b | 44.68 (28.28) | 83.32 (19.81) | 65.29 (24.87) | < .001 |
| MLU c | 4.25 (1.44) | 5.92 (1.17) | 4.38 (1.22) | < .001 |
Note. Morphosyntax reported in percent correct. MLU reported in morphemes. All other measures reported in standard score. CHL = children with permanent bilateral hearing loss; CNH-Age = children with normal hearing matched for chronological age; CNH-MLU = children with normal hearing matched for MLU in morphemes; MLU = mean length of utterance.
CHL are significantly different from CNH-MLU.
CHL are significantly different from CNH-Age and CNH-MLU.
CHL are significantly different from CNH-Age.
Measures
Descriptive Measures
Participants completed several measures to describe their cognitive, speech, and language abilities.
Nonverbal intelligence. The Primary Test of Nonverbal Intelligence (Ehrler & McGhee, 2008) was used to measure nonverbal intelligence. Children were shown an array of line drawings and asked to select the drawing that did not belong with the others. Standard scores were calculated according to the published test manual.
Speech sound production. The Arizona Articulation Proficiency Scale–Third Edition or Fourth Edition (Fudala, 2000; Fudala & Stegall, 2017) was used to measure speech sound production. Children were shown line drawings and asked to name the pictures. The Arizona Articulation Proficiency Scale was selected because it measures vowel production, known to be impaired for CHL, as well as consonant production in single words. Standard scores were calculated according to the published test manual.
Spoken language. The Test of Early Language Development–Third Edition or Fourth Edition (Hresko et al., 1999, 2018) was used to measure omnibus spoken language. Both versions of the Test of Early Language Development contain two subtests: Expressive Language and Receptive Language. The Spoken Language Quotient was calculated according to the published test manual.
The Expressive One-Word Picture Vocabulary Test–Fourth Edition (Brownell, 2011) was used to measure expressive vocabulary. Children were shown line drawings and asked to name the pictures. Standard scores were calculated according to the published test manual.
The Peabody Picture Vocabulary Test–Fourth Edition or Fifth Edition (D. Dunn, 2018; L. Dunn & Dunn, 2007) was used to measure receptive vocabulary. Children were shown an array of four pictures and asked to point to the picture named by the examiner. Standard scores were calculated according to the published test manual.
The Test of Early Grammatical Impairment Screener (TEGI; Rice & Wexler, 2001) was used to measure morphosyntax production. The TEGI consists of two subtests: Past Tense and Third-Person Singular. On the Past Tense subtest, children were shown two pictures. In the first picture, a child is completing an action; in the second picture, the child has completed the action. The examiner said a sentence about the first picture (e.g., “The boy is riding a horse” or “The girl is kicking a ball”) and asked the child to say a sentence about the second picture (targets: “The boy rode a horse” and “The girl kicked a ball”). The subtest includes regular and irregular past tense items. On the Third-Person Singular subtest, children were shown a picture of an adult and asked to tell what that person does (e.g., “This is a dentist. Tell me what a dentist does.” target: “He cleans teeth”) The TEGI was selected because it is a criterion-based measure of expressive finite verb morphology, a particular area of oral language deficit for CHL (e.g., Moeller et al., 2007). The Screener score is the average of subtest scores (percent correct), calculated according to the published test manual.
Language Sample Measure
A 12-min conversational language sample was collected using Hadley's (1998) protocol. Language sample procedures are described in detail below.
Procedure
Participants completed approximately 2 hr of individual assessment in a private room at the first or last author's lab, a local public library, or the participant's home. The study measures described above were administered in a predetermined randomized order, along with all measures in the larger study.
Language Sample Elicitation
The Hadley (1998) protocol was selected because longer, more varied utterances and a greater range of linguistic features are elicited when expository and story retelling contexts for language samples are used compared to play-based language samples (Evans & Craig, 1992; Masterson & Kamhi, 1991). Because the children in this study were at preschool age, we used pictures to support their participation in the conversational language samples (Heilmann et al., 2010). We have successfully used these procedures in previous studies (Werfel, 2018).
Each participant's language sample consisted of three blocks and was approximately 12 min long (mean time = 12 min 52 s, SD = 45 s). The first 4-min block focused on personal narratives about a birthday party or a recent holiday, as well as stories about the child's siblings or other family members. The second 4-min block focused on expository explanations about how to care for a pet and how to play a favorite game. The final 4-min block focused on story retells of favorite movies and/or books. Length of samples did not differ between groups in minutes or number of utterances (p = .385 and .645, respectively).
Language Sample Transcription
Language samples were transcribed by trained lab members or SALT Transcription Services, following procedures detailed by Werfel (2018). The language sample transcription procedures involved three passes. The first pass involved transcribing the dialogue. The second pass involved cleaning up the transcription, and the third pass was a final check of transcription accuracy. Utterance division followed Werfel and Douglas (2017).
All transcribers were trained prior to beginning transcription responsibilities. The training consisted of reading the lab's language sample transcription manual, transcribing an example of a language sample and comparing their transcription to the key, and following up with a senior lab member or the lab director to discuss consistent errors in their transcription. SALT Transcription Services provided a pilot transcription, which we compared to our transcribed master copy. Transcribers, both lab members and SALT Transcription Services, achieved > 95% transcription accuracy before completing transcription for the study.
Language Sample Coding Procedures
After the transcription process was finalized, each language sample was coded twice: once for the purpose of calculating MLU (Brown, 1973) and once for complex syntax (Schuele, 2009).
MLU calculation. Brown's morphemes that were bound were slashed in language transcripts. In addition to Brown's morphemes, contracted negation (e.g., not [can/n't]) and contracted verbs (e.g., have [we/'ve]) were slashed for purposes of calculating MLU. The slash indicates morpheme boundaries for SALT software, which was used in analysis. Only complete and intelligible utterances were used in calculation of MLU.
Complex syntax. Complex syntax coding was adapted from Barako Arndt and Schuele (2013), who delineate a coding scheme appropriate for the analysis of the emergence of complex syntax in young children. First, any utterance that had been broken following guidelines from Werfel and Douglas (2017) was placed back together as a single utterance for ease of identifying complex syntax features. Table 3 displays the complex syntax coding categories with examples of child utterances. To code for complex syntax, a trained lab member first identified any utterance that contained more than one verb phrase and coded it with [cs] to indicate an attempt at complex syntax. Next, the lab member determined the type of complex syntax feature(s) that the utterance contained. The specific complex syntax code(s) were placed after the general [cs] code for each utterance with an attempt at complex syntax. When an error of complex syntax occurred—operationalized as a missing verb in a complex syntax feature or a missing or incorrect complex syntax marker (e.g., omitted coordinated clause marker, what/that in relative clause)—the specific code was followed by [err].
Table 3.
Complex syntax coding scheme.
| Complex syntax type | SALT code | Example utterance |
|---|---|---|
| Coordinated clause | [cc] | We draw colors and play with squishy things [cs] [cc]. |
| Subordinate clause | [sc] | We didn't miss it when we went to the movie theater [cs] [sc]. |
| Reduced infinitive | [cat] | She's gonna be okay [cs] [cat]. |
| Simple infinitive | [si] | I like to play at the lake house [cs] [si]. |
| Unmarked infinitive | [uic] | I have mom hold the door [cs] [uic]. |
| Let's clause | [lc] | Let's show my teacher all my new clothes [cs] [lc]. |
| Wh-nonfinite clause | [wnfc] | I know how to play it [cs] [wnfc]. |
| Wh-finite clause | [wfc] | I don't know what I did at school [cs] [wfc]. |
| Full propositional complement | [fpc] | How did the movie theater know we were coming [cs] [fpc]? |
| Subject relative clause | [src] | I have some that are so cool for me [cs] [src]. |
| Object relative clause | [orc] | These are the toys I like to play with [cs] [orc]. |
| Oblique relative clause | [rc] | I play with some car I can pick [cs] [rc]. |
| Adjunct relative clause | [arc] | That was the time that we went to the lake [cs] [arc]. |
| Headless relative clause | [hrc] | That's what we call him [cs] [hrc]. |
| Participle clause | [pc] | I had fun having marshmallows at the party [cs] [pc]. |
Note. All utterances that contained one or more complex syntax features or attempts received a general complex syntax code [cs]. Errors of complex syntax were marked with [err].
A three-step process was utilized for MLU and complex syntax coding. First, a trained lab member made a first pass of MLU or complex syntax coding. Second, a second trained lab member checked the coding. Third, any disagreements were resolved by consensus. Disagreements were relatively rare; generally, fewer than two codes per transcript required discussion, and many transcripts had no disagreements. Thus, there was 100% agreement of the final coded transcripts.
Language Sample Analysis Procedures
After the coding process was finalized, each transcript was analyzed using SALT 18 Research Software (Miller & Iglesias, 2018). For MLU calculation, the analysis set for each child consisted only of complete and intelligible utterances; any utterance that was abandoned, interrupted, or contained at least one unintelligible word was excluded from analysis. The Standard Measures function was used to calculate MLU, number of different words, and length of sample in time.
For complex syntax analysis, the analysis set for each child consisted of the total utterances, including utterances that were incomplete or contained unintelligible words. We chose to use total utterances to look at the emergence of complex syntax production, rather than only mastery in complete and intelligible utterances, because attempts during the process of acquiring new linguistic skills may result in abandoned, incomplete, or unintelligible utterances. First the Standard Measures function was used to calculate the length of sample in utterances. Then the Explore function was used to search for each coded instance of complex syntax attempt [cs], errors [err], and each individual code listed in Table 3. Finally, the SALT Explore output was used to calculate scores for each variable of interest in this study for each transcript:
Broad Measures of Complex Syntax
Proportion of utterances containing complex syntax attempts was calculated by dividing the number of [cs] codes by the total number of utterances.
Proportion of utterances containing correct complex syntax features was calculated by dividing the number of utterances with correct instances of at least one correct use of complex syntax by the total number of utterances.
Proportion of utterances containing complex syntax errors was calculated by dividing the number of utterances with at least one error of complex syntax by the total number of utterances.
Number of complex syntax features was the total number of correct complex syntax features.
Complex syntax density was calculated by dividing the total number of correct complex syntax features by the total number of utterances.
Specific Types of Complex Syntax
For each type of complex syntax, (a) number correct and (b) percent accurate.
Statistical Analyses
Because the variables from language sample analysis did not meet assumptions of normality for one-way analysis of variance and based on the sample size, Kruskal–Wallis H tests were used to compare performance across the three groups on all measures. Mann–Whitney U tests were used for follow-up comparisons.
Results
This study aimed to evaluate complex syntax production in 4-year-old children with hearing loss compared to groups matched for chronological age and MLU. Results of the Standard Measures calculated from language sample analysis are displayed in Table 4. The group means for the CHL group were within the average range for each norm-referenced language measure (TEGI excluded because it is a criterion-referenced measure); however, 42% of CHL scored below the average range on omnibus spoken language, 31% on speech sound production, 12% on expressive vocabulary, and 19% on receptive vocabulary, and 58% scored below benchmark on morphosyntax.
Table 4.
Results from Standard Measures analyses of language sample variables.
| Construct | CHL M (SD) min–max |
CNH-Age M (SD) min–max |
CNH-MLU M (SD) min–max |
p |
|---|---|---|---|---|
| NDW a | 140.50 (46.99) 51–238 |
187.69 (34.89) 122–287 |
143.15 (49.48) 50–258 |
< .001 |
| Length – minutes | 12.89 (0.71) 12.07–14.47 |
12.71 (0.57) 11.75–14.05 |
13.03 (1.01) 11.82–15.87 |
.743 |
| Length – utterances | 135.88 (28.75) 72–204 |
137.15 (30.55) 84–224 |
130.80 (31.17) 57–224 |
.901 |
Note. CHL = children with permanent bilateral hearing loss; CNH-Age = children with normal hearing matched for chronological age; CNH-MLU = children with normal hearing matched for MLU in morphemes; MLU = mean length of utterance.
CHL are significantly different from CNH-Age.
Broad Measures of Complex Syntax Production
Kruskal–Wallis H tests indicated that there were group differences in several broad measures of complex syntax production, including the proportion of utterances with complex syntax attempts, H(2) = 12.923, p = .002; the proportion of utterances with correctly produced complex syntax features, H(2) = 14.760, p = .001; the total number of complex syntax features, H(2) = 17.016, p < .001; and complex syntax density, H(2) = 15.605, p < .001 (see Figure 1). In all cases, follow-up Mann–Whitney U tests indicated that the CHL (d = 0.89–1.08) and CNH-MLU (d = 0.96–1.02) groups performed lower than the CNH-Age group. The CHL and CNH-MLU groups did not differ on any of these measures. Finally, no group difference was found for the proportion of utterances with an errored production of complex syntax, H(2) = 4.021, p = .134. Descriptive statistics for these analyses are presented in Table 5.
Figure 1.
Complex syntax density by group. CHL = children with permanent bilateral hearing loss; CNH-Age = children with normal hearing matched for chronological age; CNH-MLU = children with normal hearing matched for mean length of utterance in morphemes.
Table 5.
Results from overall complex syntax analyses of language sample variables.
| Construct | CHL M (SD) min–max |
CNH-Age M (SD) min–max |
CNH-MLU M (SD) min–max |
p |
|---|---|---|---|---|
| Proportion utterances with complex syntax attempts a | 0.12 (0.09) 0.01–0.31 |
0.20 (0.09) 0.06–0.47 |
0.12 (0.07) 0.02–0.36 |
.002 |
| Proportion utterances with correct complex syntax a | 0.11 (0.09) 0.01–0.30 |
0.19 (0.09) 0.07–0.45 |
0.11 (0.07) 0.02–0.34 |
.001 |
| Proportion utterances with a complex syntax error | 0.02 (0.02) 0.00–0.11 |
0.01 (0.02) 0.00–0.07 |
0.01 (0.01) 0.00–0.05 |
.134 |
| Number complex syntax features a | 18.46 (16.71) 1–70 |
37.20 (18.09) 12–89 |
20.50 (16.52) 2–67 |
<.001 |
| Complex syntax density a | 0.14 (0.12) 0.00–0.45 |
0.28 (0.14) 0.12–0.68 |
0.15 (0.11) 0.02–0.50 |
<.001 |
Note. CHL = children with permanent bilateral hearing loss; CNH-Age = children with normal hearing matched for chronological age; CNH-MLU = children with normal hearing matched for MLU in morphemes; MLU = mean length of utterance.
CHL are significantly different from CNH-Age.
Production of Specific Types of Complex Syntax Features by Group
Number of Correct Productions
Group differences were observed for number correct of three specific types of complex syntax productions using Kruskal–Wallis H tests: coordinated clauses, H(2) = 15.217, p < .001; subordinate clauses, H(2) = 6.589, p = .037; and simple infinitives, H(2) = 15.133, p = .001. For coordinated clauses and simple infinitives, follow-up Mann–Whitney U tests indicated that the CHL and CNH-MLU groups produced fewer correct features than the CNH-Age group (Cohen's d for coordinated clauses = 0.95 and 0.68, simple infinitives = 0.83 and 0.91, respectively, for CNH-Age in comparison to CHL and CNH-MLU). For subordinate clauses, the CHL group produced few correct instances than the CNH-Age group (d = 0.58). No other group comparison was significant.
Percent Accuracy
Kruskal–Wallis H tests additionally identified three specific types of complex syntax features for which percent correct productions differed across groups. Group differences were observed for simple infinitives, H(2) = 8.721, p = .013; full propositional complements, H(2) = 11.043, p = .004; and subject relative clauses, H(2) = 6.933, p = .031. A follow-up Mann–Whitney U test indicated that, for simple infinities, the CHL group performed lower than the CNH-Age group (d = 0.87), but no other group differences were observed. For full propositional complements and subject relative clauses, the CHL group performed lower than the CNH-Age (d = 0.91 and 1.03, respectively) and CNH-MLU (d = 0.85 and 0.94, respectively) groups. Descriptive statistics for all types of complex syntax analyzed in this study are presented in Supplemental Materials S1 and S2, and Figure 2 displays box plots from each of the six comparisons with group differences.
Figure 2.
Box plots of number correct and percent accurate types of complex syntax productions that differed by group. Column A shows number correct, and Column B shows percent accurate. CHL = children with permanent bilateral hearing loss; CNH-Age = children with normal hearing matched for chronological age; CNH-MLU = children with normal hearing matched for mean length of utterance in morphemes.
An outlier analysis was completed by rerunning statistics and leaving removing those cases where a participant's score was more than 1.5 times the interquartile range away from the group mean. No more than three outliers were discovered in any one group for analyses, and when outliers were removed, none of the patterns of results changed.
Discussion
The purpose of this study was to compare complex syntax production of 4-year-old children with hearing loss to that of their age- and language-matched peers. Recall that comparing children with hearing loss with their age-matched peers with typical development can identify areas of delay, and comparing children with hearing loss to their language-matched peers, in this case, MLU, can identify potential areas of disrupted development (Rice, 2004).
Children with hearing loss exhibited lower scores on broad measures of complex syntax use in spontaneous language relative to their age-matched, but not language-matched, peers. This pattern of results suggests a delay in broad measures of complex syntax for children with hearing loss. Additionally, three specific types of complex syntax emerged with fewer accurate productions for children with hearing loss than age-matched peers: coordinated clauses, subordinate clauses, and simple infinitives, again indicative of delays. Finally, three types of complex syntax types emerged with lower accuracy of productions than age-matched peers: simple infinitives, full propositional complements, and subject relative clauses. For full propositional complements and subject relative clauses, children with hearing loss also had lower accuracy scores than language-matched peers. This pattern of findings suggests a delay in acquisition of simple infinitives but potential disrupted development in full propositional complements and subject relative clauses. Therefore, we conclude that 4-year-old children with hearing loss exhibit delays in complex syntax acquisition relative to same-age peers and also disrupted development in some types (i.e., full propositional complements and subject relative clauses) that cannot be explained by lower MLU alone. The findings of the current study align with previous studies that have identified morphosyntactic weaknesses in children with hearing loss (e.g., Briscoe et al., 2001; Moeller et al., 2010; Werfel, 2018). Additionally, our findings align with previous reports of delays in complex syntax acquisition in children with SLI based on MLU, as well as disrupted development of some complex syntax structures in children with SLI (e.g., Schuele & Dykes, 2005).
Strengths in Complex Syntax Production of Children With Hearing Loss
We note that children with hearing loss presented with multiple areas of strength within complex syntax acquisition. For example, children with hearing loss did not produce utterances with errors in complex syntax production at a higher rate than either age- or language-matched peers. This finding indicates that, when children with hearing loss attempt complex syntax production, they generally are similarly successful as children with normal hearing. Additionally, in the period of emergence of complex syntax production, we identified only a few specific complex syntax structures on which children with hearing loss were less productive or less accurate than children with normal hearing. In both the count analysis and the percent accuracy analysis, only three of 15 types of complex syntax showed group differences. We discuss these differences below.
Group Differences on Broad Measures of Complex Syntax
Our findings of complex syntax productions in children with hearing loss parallels the results of previous research that clinical groups have fewer instances of complex syntax production than age-matched peers (Owen & Leonard, 2006; Owen Van Horne & Lin, 2011). The proportion of utterances with complex syntax attempts in children with hearing loss was comparable to their MLU-matched peers with normal hearing. In addition, the proportion of correct complex syntax productions also did not differ significantly between children with hearing loss and the MLU-matched group. Thus, although the proportion of complex syntax attempts and accurate productions are significantly lower than their age-matched peers, children with hearing loss are still producing correct complex syntax structures similar to their normal-hearing peers with lower MLU who are younger. Our finding parallels reported findings for children with SLI (Barako Arndt & Schuele, 2013), suggesting that children with hearing loss are likely to have delayed emergence of complex syntax use in spontaneous language. It is important to note that our MLU-matched group was approximately 5 months younger than the CHL group.
Previous work has shown that children with hearing loss produce shorter utterances with less lexical diversity compared to their peers with normal hearing (e.g., Koehlinger et al., 2013; Werfel & Douglas, 2017). In the current study, children with hearing loss differed from their age-matched peers, who produced longer sentences across a wide variety of complex syntax measures. In contrast, when children with hearing loss were compared to younger children with similar MLUs, the only differences that emerged were percent accuracy on two specific types of embedded complex syntax. These results suggest that shorter-than-expected MLU for a child's age is a potential red flag that could alert clinicians to the need for intervention goals targeting complex syntax during the preschool years for children with hearing loss, even if the difference is not very large.
In addition to lower proportion of complex syntax attempts, the CHL group demonstrated lower complex syntax density compared to their age-matched peers with normal hearing but no observable difference in complex syntax density when the CHL group was compared to language-matched peers with normal hearing. This indicates that, for children with hearing loss, complex syntax density or overall rate of use of complex syntax features is delayed rather than disrupted. If the CHL group had exhibited complex syntax density lower than both groups with normal hearing, this would have suggested their complex syntax development may follow a different trajectory than their language-matched peers. Instead, the data revealed a delayed rather than disrupted trajectory of overall syntax development. This suggests that the lower MLU of the children with hearing loss may reflect delays in use of complex syntax in their spontaneous language.
Group Differences in Number of Accurate Productions of Complex Syntax
Beyond broad measures of complex syntax production, we were additionally interested in identifying if children with hearing loss differed from either control group on the number of correct productions of any particular types of complex syntax. That is, it is important to consider how often children with hearing loss are attempting and succeeding in producing complex syntax structures as compared to children with normal hearing. Our findings identified three complex syntax features on which children with hearing loss had a lower number of accurate productions than their age-matched peers. Across all groups, the complex syntax types that were produced most frequently were coordinated clauses, subordinate clauses, and simple infinitives. For each of these three types, we found that the CHL group produced fewer instances in their spontaneous language than the CNH-Age group, but no difference from the CNH-MLU group. These findings add support to the conclusion that complex syntax acquisition is delayed in children with hearing loss relative to their peers with normal hearing and provide information on possible intervention targets for preschoolers with hearing loss.
Group Differences in Percent Accuracy on Specific Types of Complex Syntax
Finally, we were interested in identifying if children with hearing loss differed from either comparison group on percent accuracy of any particular types of complex syntax production. In other words, examination of this variable captures how often children with hearing loss are making errors when they attempt complex syntactic structures as compared to children with normal hearing, thus allowing exploration of errored productions in types of complex syntax that are produced less often. Our findings identified group differences in percent accuracy on three types of complex syntax: simple infinitives, full propositional complements, and subject relative clauses. For simple infinitives, we found that children with hearing loss had lower percent accuracy than their age-matched, but not language-matched, peers. One likely cause is the omission of the obligatory “to” marker in the production of simple infinitives; this pattern would parallel errored patterns of simple infinitive productions in preschool children with language impairment, as well as younger children with typical language (e.g., Diessel, 2004; Schuele & Dykes, 2005). It is possible, because “to” is a form rather than a content word, that it is neither semantically nor perceptually salient to children with hearing loss and therefore likely to be left out of sentences.
For full propositional complements and subject relative clauses, however, a different pattern of findings emerged. For both of these complex syntax types, children with hearing loss had lower accuracy than both their age- and language-matched peers. Previous work in children with language impairment has identified errors in both of these types of complex syntax: omission of the “if” complementizer in full propositional complements and omission of the obligatory relative marker in subject relative clauses (Schuele & Dykes, 2005). Omission of the obligatory relative marker has also been reported in younger children with typical language development (Diessel & Tomasello, 2000). Interestingly, for these two types of complex syntax, the CHL group had lower percent accuracy than the CNH-MLU group. These findings suggest that the acquisition of complex syntax production may not merely be delayed for children with hearing loss, but for some complex syntax features, particularly embedded clauses, may additionally be disrupted.
Full propositional complements and subject relative clauses are types of complex syntax that are embedded rather than sequentially placed clauses (e.g., Diessel, 2004; Schuele & Dykes, 2005). For example, a relative clause places one idea and related verb phrase inside of another clause, as compared to a subordinate clause, which lists one idea and then another. The added complexity of embedding, rather than listing clauses, may contribute to increased error rates for children with hearing loss. Extant literature indicates that, in addition to having less language learning experience as a result of experiencing a period of auditory deprivation, children with hearing loss also experience poorer speech perception than children with normal hearing (Kral et al., 2016). Increasing numbers of studies indicate that children with hearing loss experience increased listening effort and fatigue as compared to children with normal hearing, likely as a result of using cognitive effort to attend to the speech signal (Hicks & Tharpe, 2002; Hornsby, 2013; Hornsby et al., 2014; Werfel & Hendricks, 2016). Because embedded clauses are more complex and difficult to process than other types of clauses, it is possible that their acquisition by children with hearing loss is disrupted rather than simply delayed. In other words, children with hearing loss who experience difficulties with speech perception and related fatigue may be less likely to learn particularly complex syntactic structures without intervention. This possibility merits exploration in future work.
Clinical Implications
Given the results of this analysis, we conclude that it is vital for preschool children with hearing loss to receive explicit instruction for complex syntax. Our findings provide guidance for selecting specific complex syntax structures that clinicians should consider targeting. Clinicians should consider placing particular focus on embedded clauses, such as full propositional complements and subject relative clauses, because these types of complex syntax appear to be particularly difficult for children with hearing loss.
In many cases, the MLU of children with hearing loss appears to be associated with delays in acquisition of complex syntax production in spontaneous language. Shorter-than-expected MLUs could serve as a red flag for the need for complex syntax goals for children with hearing loss. It is important to note that the mean scores on each norm-referenced language measure was within the average range, but significantly lower than both peer groups of children with normal hearing, for the children with hearing loss. Therefore, it is vital that service provision decisions for preschool children with hearing loss include a holistic approach of measuring language production and consider differences in real-world language production that may not result in lower-than-average scores on standardized tests.
Although broad use of complex syntax presents as a delay (as compared to disrupted use of complex syntax) for children with hearing loss, this delay compared to peers with normal hearing cannot be ignored. Even on complex syntax measures that appear to be only delayed, it is vital to provide early intervention to increase these skills for children with hearing loss, because they are educated overwhelmingly in mainstreamed classrooms (U.S. Department of Education, 2020). Delays in complex syntax production may put children with hearing loss at risk for exacerbated linguistic delays (the delays they come to school with plus delays that are the result of not understanding classroom instruction). An early focus on increasing MLU by specifically targeting complex syntax for toddlers and early preschoolers with hearing loss might help alleviate some of the delays that we observed. Clinicians should focus additionally on grammatical morpheme interventions for young children with hearing loss. Werfel (2018) reported particularly low accuracy on tense-marking morphemes for children with hearing loss, which would be an appropriate target for most of the population early in preschool.
In other cases of complex syntax production, children with hearing loss had differences relative to the CNH-MLU group, suggesting that increasing MLU alone is unlikely to result in typical complex syntax use. Instead, we suggest that clinicians should also specifically target at least some types of complex syntax as soon as MLU reaches 3.0. From this study, it appears that subject relative clauses and full propositional complements, both embedded clauses, would both be particularly appropriate targets during language intervention.
Limitations
One limitation of this study is the relatively small sample size; however, small sample sizes are not uncommon when studying hearing loss given its low incidence. Furthermore, the participants in this study used spoken English and had limited amounts of sign language use. These findings should not be extended to populations beyond that of this study. Lastly, the types of amplification and degrees of hearing loss of the participants in this study were mixed; future research should investigate complex syntax in groups of children with hearing loss using cochlear implants and hearing aids separately.
Future Directions
We propose that future research should further explore two areas relative to complex syntax production in children with hearing loss. First, longitudinal investigation of the emergence of complex syntax production in children with hearing loss is needed. Such work could provide insight into whether a delayed or difference model of acquisition is appropriate for children with hearing loss. Second, investigations that develop and/or evaluate effectiveness of complex syntax interventions for preschool children with hearing loss are needed. One of the most common ways to target complex syntax is with sentence combining, which is an effective way to increase the production of complex syntax and its accuracy in school-age children with normal hearing (e.g., Balthazar & Scott, 2018; Graham & Perin, 2007). For preschoolers with normal hearing, narrative interventions have been shown to increase complex syntax (e.g., Petersen & Spencer, 2016; Vasilyeva et al., 2006). Intervention research could explore modifications necessary to adapt such procedures for children with hearing loss.
Conclusions
The purpose of this study was to compare complex syntax production of 4-year-old children with hearing loss to that of their age- and language-matched peers. Findings indicated that children with hearing loss exhibit broad delays in complex syntax use in spontaneous language. Percentage of utterances that contained a complex syntax attempt and complex syntax density of children with hearing loss were delayed relative to their age-matched, but not language-matched, peers. Additionally, for coordinated clauses, subordinate clauses, and simple infinitives, children with hearing loss had fewer accurate productions than their age-matched peers. For simple infinitives, full propositional complements, and subject relative clauses, children with hearing loss had lower percent accuracy of productions than age-matched peers. Finally, for full propositional complements and subject relative clauses, children with hearing loss also had lower accuracy than language-matched peers. Therefore, we conclude that children with hearing loss exhibit broad delays in complex syntax acquisition and also disrupted productions in some types that cannot be explained by low MLU alone.
Author Contributions
Krystal L. Werfel: Conceptualization (Lead), Data curation (Lead), Formal analysis (Lead), Funding acquisition (Lead), Investigation (Lead), Methodology (Lead), Project administration (Lead), Writing – original draft (Lead), Writing – review & editing (Lead). Gabriella Reynolds: Formal analysis (Equal), Writing – original draft (Supporting), Writing – review & editing (Supporting). Sarah Hudgins: Data curation (Supporting), Writing – original draft (Supporting), Writing – review & editing (Supporting). Marissa Castaldo: Data curation (Supporting), Writing – original draft (Supporting). Emily A. Lund: Funding acquisition (Equal), Methodology (Equal), Writing – original draft (Supporting), Writing – review & editing (Equal).
Supplementary Material
Acknowledgments
The work reported herein was funded by grants from the National Institute on Deafness and Other Communication Disorders (R03DC014535 to Krystal L. Werfel; R01DC017173 to Krystal L. Werfel and Emily A. Lund). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Funding Statement
The work reported herein was funded by grants from the National Institute on Deafness and Other Communication Disorders (R03DC014535 to Krystal L. Werfel; R01DC017173 to Krystal L. Werfel and Emily A. Lund). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
We use the term broad measures of complex syntax to refer to those measures that assess complex syntax production as a whole, such as complex syntax density, and the term specific types of complex syntax to refer to particular syntactic features, such as coordinated clauses or simple infinitives.
Participants were administered current versions of each assessment based on their testing date.
References
- Balthazar, C. H. , & Scott, C. M. (2018). Targeting complex sentences in older school children with specific language impairment: Results from an early-phase treatment study. Journal of Speech, Language, and Hearing Research, 61(3), 713–728. https://doi.org/10.1044/2017_JSLHR-L-17-0105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Barako Arndt, K. , & 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. https://doi.org/10.1097/TLD.0b013e31828f9ee8 [Google Scholar]
- Bloom, L. , & Capatides, J. B. (1987). Expression of affect and the emergence of language. Child Development, 58(6), 1513–1522. https://doi.org/10.2307/1130691 [Google Scholar]
- Bloom, L. , Tackeff, J. , & Lahey, M. (1984). Learning to in complement constructions. Journal of Child Language, 11(2), 391–406. https://doi.org/10.1017/S0305000900005833 [DOI] [PubMed] [Google Scholar]
- Briscoe, J. , Bishop, D. V. , & Norbury, C. F. (2001). Phonological processing, language, and literacy: A comparison of children with mild-to-moderate sensorineural hearing loss and those with specific language impairment. Journal of Child Psychology and Psychiatry and Allied Disciplines, 42(3), 329–340. https://doi.org/10.1111/1469-7610.00726 [PubMed] [Google Scholar]
- Brown, R. (1973). A first language: The early stages. Harvard University Press. [Google Scholar]
- Brownell, R. (2011). Expressive One-Word Picture Vocabulary Test–Fourth Edition. Academic Therapy Publications. [Google Scholar]
- Craig, H. K. , Connor, C. M. , & Washington, J. A. (2003). Early positive predictors of later reading comprehension for African American students: A preliminary investigation. Language, Speech, and Hearing Services in Schools, 34(1), 31–43. https://doi.org/10.1044/0161-1461(2003/004) [DOI] [PubMed] [Google Scholar]
- Diessel, H. (2004). The acquisition of complex sentences. Cambridge University Press. https://doi.org/10.1017/CBO9780511486531 [Google Scholar]
- Diessel, H. , & Tomasello, M. (2000). The development of relative clauses in spontaneous child speech. Cognitive Linguistics, 11(1–2), 131–151. https://doi.org/10.1515/cogl.2001.006 [Google Scholar]
- Dunn, D. (2018). Peabody Picture Vocabulary Test–Fifth Edition. Pearson. [Google Scholar]
- Dunn, L. , & Dunn, D. (2007). Peabody Picture Vocabulary Test–Fourth Edition (PPVT-4). Pearson. https://doi.org/10.1037/t15144-000 [Google Scholar]
- Ehrler, D. , & McGhee, R. (2008). Primary Test of Nonverbal Intelligence. Pro-Ed. [Google Scholar]
- Evans, J. , & Craig, H. (1992). Language sample collection and analysis: Interview compared to free play assessment contexts. Journal of Speech and Hearing Research, 35(2), 343–353. https://doi.org/10.1044/jshr.3502.343 [DOI] [PubMed] [Google Scholar]
- Fudala, J. B. (2000). Arizona Articulation Proficiency Scale–Third Edition. Western Psychological Services. [Google Scholar]
- Fudala, J. B. , & Stegall, S. (2017). Arizona Articulation and Phonology Scale–Fourth Edition (Arizona-4). Western Psychological Services. [Google Scholar]
- Geers, A. E. , Nicholas, J. , Tobey, E. , & Davidson, L. (2016). Persistent language delay versus late language emergence in children with early cochlear implantation. Journal of Speech, Language, and Hearing Research, 59(1), 155–170. https://doi.org/10.1044/2015_JSLHR-H-14-0173 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Graham, S. , & Perin, D. (2007). A meta-analysis of writing instruction for adolescent students. Journal of Educational Psychology, 99(3), 445–476. https://doi.org/10.1037/0022-0663.99.3.445 [Google Scholar]
- Hadley, P. (1998). Language sampling protocols for eliciting text-level discourse. Language, Speech, and Hearing Services in Schools, 29(3), 132–147. https://doi.org/10.1044/0161-1461.2903.132 [DOI] [PubMed] [Google Scholar]
- Halliday, L. F. , Tuomainen, O. , & Rosen, S. (2017). Language development and impairment in children with mild to moderate sensorineural hearing loss. Journal of Speech, Language, and Hearing Research, 60(6), 1551–1567. https://doi.org/10.1044/2016_JSLHR-L-16-0297 [DOI] [PubMed] [Google Scholar]
- 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. https://doi.org/10.1044/0161-1461(2009/09-0023) [DOI] [PubMed] [Google Scholar]
- Hicks, C. , & Tharpe, A. M. (2002). Listening effort and fatigue in school-age children with and without hearing loss. Journal of Speech, Language, and Hearing Research, 45(3), 573–584. https://doi.org/10.1044/1092-4388(2002/046) [DOI] [PubMed] [Google Scholar]
- Hornsby, B. (2013). The effects of hearing aid use on listening effort and mental fatigue associated with sustained speech processing demands. Ear and Hearing, 34(5), 523–534. https://doi.org/10.1097/AUD.0b013e31828003d8 [DOI] [PubMed] [Google Scholar]
- Hornsby, B. , Werfel, K. L. , Camarata, S. , & Bess, F. (2014). Subjective fatigue in children with hearing loss: Some preliminary findings. American Journal of Audiology, 23(1), 129–134. https://doi.org/10.1044/1059-0889(2013/13-0017) [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hresko, W. , Reid, D. , & Hammill, D. (1999). Test of Early Language Development–Third Edition. Pro-Ed. [Google Scholar]
- Hresko, W. , Reid, D. , & Hammill, D. (2018). Test of Early Language Development–Fourth Edition. Pro-Ed. [Google Scholar]
- Koehlinger, K. M. , Van Horne, A. J. , & Moeller, M. P. (2013). Grammatical outcomes of 3- and 6-year-old children who are hard of hearing. Journal of Speech, Language, and Hearing Research, 56(5), 1701–1714. https://doi.org/10.1044/1092-4388(2013/12-0188) [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kral, A. , Kronenberger, W. G. , Pisoni, D. B. , & O'Donoghue, G. M. (2016). Neurocognitive factors in sensory restoration of early deafness: A connectome model. The Lancet, Neurology, 15(6), 610–621. https://doi.org/10.1016/S1474-4422(16)00034-X [DOI] [PMC free article] [PubMed] [Google Scholar]
- Limber, J. (1973). The genesis of complex sentences. In Moore T. E. (Ed.), Cognitive development and the acquisition of language. Academic Press. [Google Scholar]
- Marinellie, S. (2004). Complex syntax used by school-age children with specific language impairment (SLI) in child–adult conversation. Journal of Communication Disorders, 37(6), 517–533. https://doi.org/10.1016/j.jcomdis.2004.03.005 [DOI] [PubMed] [Google Scholar]
- Masterson, J. , & Kamhi, A. (1991). The effects of sampling conditions on sentence production in normal, reading-disabled, and language-learning disabled children. Journal of Speech and Hearing Research, 34(3), 549–558. https://doi.org/10.1044/jshr.3403.549 [DOI] [PubMed] [Google Scholar]
- Miller, J. , & Iglesias, A. (2018). Systematic Analysis of Language Transcripts (SALT), Research Version 2018. SALT Software. [Google Scholar]
- Moeller, M. P. , McCleary, E. , Putman, C. , Tyler-Krings, A. , Hoover, B. , & Stelmachowicz, P. (2010). Longitudinal development of phonology and morphology in children with late-identified mild–moderate sensorineural hearing loss. Ear and Hearing, 31(5), 625–635. https://doi.org/10.1097/AUD.0b013e3181df5cc2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Moeller, M. P. , Tomblin, J. B. , Yoshinaga-Itano, C. , Connor, C. M. , & Jerger, S. (2007). Current state of knowledge: Language and literacy of children with hearing impairment. Ear and Hearing, 28(6), 740–753. https://doi.org/10.1097/AUD.0b013e318157f07f [DOI] [PubMed] [Google Scholar]
- Owen, A. J. , & Leonard, L. B. (2006). The production of finite and nonfinite complement clauses by children with specific language impairment and their typically developing peers. Journal of Speech, Language, and Hearing Research, 49(3), 548–571. https://doi.org/10.1044/10902-4388(2006/040) [DOI] [PubMed] [Google Scholar]
- Owen Van Horne, A. J. , & Lin, S. (2011). Cognitive state verbs and complement clauses in children with SLI and their typically developing peers. Clinical Linguistics & Phonetics, 25(10), 881–898. https://doi.org/10.3109/02699206.2011.582226 [DOI] [PubMed] [Google Scholar]
- Petersen, D. , & Spencer, T. (2016). Using narrative intervention to accelerate canonical story grammar and complex language growth in culturally diverse pre-schoolers. Topics in Language Disorders, 36(1), 6–19. https://doi.org/10.1097/TLD.0000000000000078 [Google Scholar]
- Rice, M. (2004). Growth models of developmental language disorders. In Rice M. L. & Warren S. F. (Eds.), Developmental language disorders: From phenotypes to etiologies (pp. 207–240). Erlbaum. https://doi.org/10.4324/9781410610881 [Google Scholar]
- Rice, M. , & Wexler, K. (2001). Rice/Wexler Test of Early Grammatical Impairment. The Psychological Corporation. [Google Scholar]
- Scarborough, H. S. (1990). Index of Productive Syntax. Applied Psycholinguistics, 11, 1–22. https://doi.org/10.1017/S0142716400008262 [Google Scholar]
- Schuele, C. M. (2009). Complex syntax coding manual. Vanderbilt University. [Google Scholar]
- Schuele, C. M. , & Dykes, J. C. (2005). Complex syntax acquisition: A longitudinal case study of a child with specific language impairment. Clinical Linguistics & Phonetics, 19(4), 295–318. https://doi.org/10.1080/02699200410001703709 [DOI] [PubMed] [Google Scholar]
- Truckenmiller, A. J. , & Petscher, Y. (2019). The role of academic language in written composition in elementary and middle school. Reading and Writing: An Interdisciplinary Journal, 33, 45–66. https://doi.org/10.1007/s11145-019-09938-7 [Google Scholar]
- Tyack, D. L. , & Gottsleben, R. H. (1986). Acquisition of complex sentences. Language, Speech, and Hearing Services in Schools, 17(3), 160–174. https://doi.org/10.1044/0161-1461.1703.160 [Google Scholar]
- U.S. Department of Education. (2020). Office of Special Education Programs, Individuals with Disabilities Education Act (IDEA) database, Table 204.60. https://nces.ed.gov/programs/digest/d19/tables/dt19_204.60.asp?current=yes
- Vasilyeva, M. , Huttenlocher, J. , & Waterfall, H. (2006). Effects of language intervention on syntactic skill levels in preschoolers. Developmental Psychology, 42(1), 164–174. https://doi.org/10.1037/0012-1649.42.1.164 [DOI] [PubMed] [Google Scholar]
- Werfel, K. L. (2017). Emergent literacy skills in preschool children with hearing loss who use spoken language: Initial findings from the early language and literacy acquisition (ELLA) study. Language, Speech, and Hearing Services in Schools, 48(4), 249–259. https://doi.org/10.1044/2017_LSHSS-17-0023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Werfel, K. L. (2018). Morphosyntax production of preschool children with hearing loss: An evaluation of the extended optional infinitive and surface accounts. Journal of Speech, Language, and Hearing Research, 61(9), 2313–2324. https://doi.org/10.1044/2018_JSLHR-L-17-0406 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Werfel, K. L. , & Douglas, M. (2017). Are we slipping them through the cracks? The insufficiency of norm-referenced assessments for identifying language weaknesses in children with hearing loss. Perspectives of the ASHA Special Interest Groups, 2(9), 43–53. https://doi.org/10.1044/persp2.SIG9.43 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Werfel, K. L. , & Hendricks, A. E. (2016). The relation between child versus parent report of chronic fatigue and language/literacy performance in school-age children with cochlear implants. Ear and Hearing, 37(2), 216–224. https://doi.org/10.1097/AUD.0000000000000242 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.