Narrative skills in Turkish-speaking children with hearing loss: a comparative analysis with normal-hearing peers

Abstract

Background

The assessment of narrative skills, which reflect higher-order language abilities, provides clear evidence of children’s developmental progress and may predict later academic achievement and language development.

Aims

This study aimed to compare the narrative skills of children with hearing loss (HL) and those with normal hearing (NH).

Methods

The study group included of 27 children with HL and 27 children with NH aged 4–8 year. The Turkish Early Language Development Test (TEDIL) was used to evaluate children’s early verbal language development, and the Edmonton Narrative Norms Instrument was used to assess narrative skills.

Results

According to the results, TEDIL scores (p = 0.000, p = 0.001, and p = 0.000), mean length of utterance in words (MLUw; p = 0.014), and morphemes (MLUm; p = 0.004) in the A1 story were significantly lower in children with HL than in those with NH. Significant differences were also found between the two groups in A1 (p = 0.002) and A3 (p = 0.029) story structure scores, in favour of children with NH. Additionally, significant differences were observed in NDW (p = 0.039) and TNW (p = 0.004) in the A1 story for children with HL, and in MLUw (p = 0.046), TNW (p = 0.007), NDW (p = 0.002), and story structure (p = 0.006) in the A3 story for children with NH.

Conclusion

The findings indicate that children with HL perform more poorly in language and narrative skills than children with NH. In addition, it was found that children with HL showed greater difficulty in making inferences from the story and establishing cause and effect relationships within stories. These results suggest that educational programmes should be developed to assess and support the narrative skills of children with HL, enabling them to catch up with their NH peers.

Introduction

Hearing loss (HL) is a common condition that may occur at any age and may be hereditary or acquired, and is associated with a wide range of aetiologies [1]. Because verbal communication is primarily acquired through auditory input, HL may lead to difficulties in speech perception and production [2]. Children with HL generally have lower scores in receptive vocabulary, language, and phonological skills than children with normal hearing (NH) [3]. In addition, they may produce shorter and simpler sentences, often consisting mainly of nouns and verbs, compared with their NH peers [4] .

Language development in children with HL is influenced by several factors, including age at onset, age at intervention, degree and configuration of HL, educational programme, and family and environmental factors [2]. However, early amplification and timely intervention have been shown to reduce some negative effects of HL on language development [5–7] .

Narrative skill is a key component of children’s language development and refers to the ability to convey a story that includes sequential information about events [8]. Narrative skills support receptive and expressive language development. Because narratives involve more complex language structures than everyday conversation, narrative assessment can reflect developmental progress and may predict later academic achievement and language development [9, 10]. Cognitive abilities such as memory, attention, organization, and planning, are essential for them to engaging with stories in a coherent manner [11]. In addition, children’s narrative skills play an important role in social development [12] and have been shown to predict later literacy skills [13, 14]. Therefore, oral narratives are commonly recommended as part of language assessment [15–18] .

Narrative skills are typically examined at two levels: macrostructure and microstructure. Macrostructure refers to the global organization and overall coherence of a story, including story grammar elements such as setting, initiating events, goals, attempts, outcomes, and resolution [19, 20]. It also reflects the logical sequencing and completeness of narrative episodes. References to characters’ thoughts, feelings, and intentions further enhance narrative coherence [19]. Story grammar represents the structural organisation of a narrative and the relationships among its components [21].

Microstructure, in contrast, refers to linguistic features at the word, phrase, and sentence levels. It includes productivity measures such as total number of words (TNW), lexical diversity such as number of different words (NDW), and syntactic complexity such as mean length of utterance in words (MLUw) and morphemes (MLUm) [22, 23]. Additional features include morphosyntactic accuracy, grammaticality, use of conjunctions, referential cohesion, and word-class diversity [24]. While macrostructure reflects global coherence, microstructure captures linguistic complexity and richness in narrative production [24]. Analysis of narrative data provides insight into how children apply linguistic knowledge in discourse [25].

The literature on narrative skills in children with HL remains limited. Existing studies suggest that HL negatively affects narrative performance at macrostructural and microstructural levels. Maria et al. [26] reported that children with HL performed significantly worse than their NH peers on most narrative measures. Muthu et al. [27] found that children aged 3–5 years with cochlear implants (CI) had greater difficulty in understanding and using story structure than NH peers. Walker et al. [28] also reported significant difficulties in macrostructural narrative skills among children with HL. Boons et al. [29] found that children with cochlear implants performed adequately in structural components but showed limitations in narrative richness and productivity. Overall, the findings suggest a complex relationship between HL and narrative development.

Given these findings, it is important to examine the narrative difficulties experienced by children with HL to inform educational interventions. In the Turkish literature, existing studies have mainly focused on limited microstructural measures such as MLU, NDW, TNW, and narrative errors [10, 30] .

However, studies examining macrostructural and microstructural narrative skills in Turkish-speaking children with HL remain limited. In addition, comparative research on children using hearing aids (HAs) and cochlear implants (CIs) is still scarce.

Accordingly, this study aimed to compare narrative skills and story structure in Turkish-speaking children aged 4–8 years with HL (using HA or CI) and children with NH. It was hypothesised that children with HL would perform lower than NH peers in microstructural and macrostructural measures. Based on these considerations, the present study aimed to compare the narrative skills and story structure of Turkish-speaking children aged 4–8 years with HL who use HA or CI with those of children with NH. It was hypothesized that children with HL would perform lower than children with NH in both microstructural narrative measures and macrostructural story structure scores. With this purpose, answers to the following questions were sought in the study:

1. Do children with HL differ significantly from children with NH in microstructural narrative measures, including MLU, NDW, TNW, and macrostructural story structure scores?

2. Among children with HL, is there a relationship between narrative skills and age, type of amplification, age at diagnosis of HL, and maternal education level?

Materials and methods

This study was conducted with children with HL and NH who were admitted to the Audiology Unit of the Department of Otolaryngology of Ege University Faculty of Medicine Hospital. Following the otolaryngology examination, the researcher provided general information about the study to the parents and children. Then, the demographic information form containing some information was completed by the researcher through interviews with the families.

Ethics committee permission

All procedures performed in studies involving human participants were conducted in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Medical Research Ethics Committee of Ege University (No.92659746-100-1011944 ).

Participants

The study included 54 children aged 4–8 years, of whom 27 had HL, and 27 had NH, matched for age and gender. Children with pure-tone audiometry hearing thresholds of 15 dB HL or better at 500–8,000 Hz, no middle ear problems (Type A tympanogram), and no additional disabilities or language disorders were included in the NH group. The HL group included children with severe-to-profound HL (thresholds ≥ 71 dB HL) who had been using cochlear implants (CIs) for at least 1 year and children with moderate-to-severe HL who had used hearing aids (HAs) for more than 6 months. Participants were required to have aided free-field thresholds of 40-dB HL or better, no comorbid disabilities, no inner ear malformations or auditory neuropathy, and language development within 1.5 years of their chronological age according to TEDIL scores.

Among the children with HL, 12 had bilateral CIs, and 15 had bilateral HAs. The mean age at diagnosis of HL was 14.33 ± 18.09 months. The mean duration of HA use was 60.52 ± 18.25 months, while the mean age at cochlear implantation was 20.75 ± 9.33 months, and the mean age at initiation of special education was 18.59 ± 18.26 months. The demographic characteristics of the children and their parents are presented in Table 1.

Table 1.

Demographic information of the children in the study group

Variable	Hearing Loss	Normal Hearing	p
Children’s age average (M ± SD)	74.74 ± 13.49	75.19 ± 13.12	0.903
Mothers’ age average (M ± SD)	35.74 ± 5.05	35.52 ± 4.26	0.862
Fathers’ age average (M ± SD)	38.93 ± 4.19	37.19 ± 3.62	0.108
Gender, n (%)			1.000
Female	14(51.9)	14(51.9)
Male	13(48.1)	13(48.1)
Mothers’ Education Level, n(%)			0.165
Elementary School	6 (22.2)	3 (11.1)
High School	10 (37.0)	6 (22.2)
University	11 (40.7)	18 (66.7)

M Mean, SD Standard Deviation

Procedure

The children underwent a hearing assessment during the first session. The TEDIL test was administered to children who met the inclusion criteria. In the second session, the ENNI narrative tool was applied. During ENNI administration, the children’s narratives were recorded using a voice recorder.

The Turkish SALT (Systematic Analysis of Language Transcripts) computer program [31] was used to transcribe the children’s utterances word by word in Turkish. To calculate interjudge reliability, 30% of all language samples were independently analysed by another researcher in terms of utterance segmentation, morpheme segmentation, and coding. Interjudge reliability was calculated by the agreement / (agreement + disagreement) x 100 formula [32]. As a result of the interjudge reliability calculations, division into utterances reliability was 94%, division into morphemes reliability was 95%, and coding reliability was 93%.

Assessment tools

Turkish Test of Early Language Development (TEDIL)

TEDIL, adapted into Turkish in 2013 by Topbaş and Güven from the Test of Early Language Development (TELD-3), language development test, is used to assess the early verbal language development of Turkish-speaking children aged 2;0–7;11 year. It evaluates receptive and expressive language skills and consists of two equivalent forms, A and B. Both forms include receptive and expressive language subtests assessing semantic development and syntax–morphology.

Correct responses receive one point, whereas incorrect responses receive zero points. The total raw score is converted into a standard score, with higher scores indicating better verbal language skills [33]. In this study, Form A was used. Composite scores for receptive language, expressive language, and verbal language were calculated.

The Edmonton Narrative Norms Instrument- ENNI

The ENNI narrative tool, developed by Schneider et al. [34] and adapted into Turkish by Piştav Akmeşe in 2025 [35], was used to assess narrative language skills. ENNI evaluates the expressive language skills of children 4–9 year through storytelling. It includes one training story and three test stories presented as picture sequences without written text. Children are asked to narrate the stories based on the pictures. The stories increase progressively in length, story complexity, character sex, and number, ranging from simple two-character stories to more complex four-character stories. In this the study, the training story and the A1 and A3 stories were used. To familiarize all the children with the test procedure, the training story was administered first. The training story includes two characters and consists of five pictures representing one event/episode. Following the training story, the A1 and A3 test stories were presented.

Children were asked to narrate the stories page by page while viewing the pictures. Subsequently, questions related to the A1 and A3 stories were administered. The narratives were transcribed, and the following measures were obtained:

• The Mean length of utterance: Calculated by dividing the number of morphemes by the total number of utterances.

• Number of Different Words (NDW): Calculated by counting the number of different words used in the language sample.

• Total Number of Words (TNW): The total number of intelligible words in the language sample [36].

The A1 and A3 story structures were scored using the Story Grammar Scoring Sheet for ENNI A1 and A3 Stories. The scoring system includes characters, the setting of the story, internal response, internal plan, initiative, outcome, and character reactions. Each unit is scored as 0, 1, or 2, and the scores are summed to obtain a total raw score [34] .

Data analysis

A priori sample size analysis was performed using G*Power 3.1 based on an effect size of 0.91 [10, 37], α = 0.05, and statistical power = 0.95, indicating that a minimum of 27 participants was required for each group. Statistical analyses were performed using SPSS version 24.0 [38] .

The distribution of continuous variables was assessed using the Shapiro–Wilk test. Because TEDIL and ENNI narrative variables were not normally distributed, descriptive statistics were presented as median and interquartile range (Q1–Q3), and group comparisons were performed using the Mann–Whitney U test. The Mann–Whitney U test was also used to compare TEDIL and narrative skill scores according to the type of amplification used by children with HL.

Relationships between narrative ability scores were examined using Spearman correlation analysis. A p value of < 0.05 was considered statistically significant.

Results

In this study, the narrative skills of children with HL and NH were compared. The findings included language scores obtained from TEDIL test and narrative language measures derived from the ENNI As shown in Table 2, children with NH achieved significantly higher scores in receptive, expressive, and total verbal language (p < 0.001) as well as in A1 MLUw (p = 0.014), A1 MLUm (p = 0.004), and story structure scores for both A1 (p = 0.002) and A3 (p = 0.029). No significant group differences were found for NDW and TNW in either story or for A3 MLUw and MLUm scores (p > 0.05).

Table 2.

The comparison of the children according to their TEDIL and narrative skill scores

Variable	Hearing Loss (n = 27) Median (Q1–Q3)	Normal Hearing (n = 27) Median (Q1–Q3)	Z	p
TEDIL Receptive Language Score	87 (83–97)	98(89–115)	-3.582	< 0.001
TEDIL Expressive Language Score	92 (87–98)	104 (92–108)	-3.246	0.001**
TEDIL Verbal Language Score	89 (81–96)	102 (94–114)	-4.330	< 0.001
A1 MLUw	3.50 (3.00-4.14)	4.43 (3.57–4.86)	-2.541	0.014*
A3 MLUw	4.09 (3.43–4.50)	4.23 (3.61–5.08)	-1.497	0.134
A1 MLUm	5.92 (4.56–7.22)	7.38 (5.86–7.83)	-3.007	0.004**
A3 MLUm	7.07 (5.98–8.13)	7.55 (6.65–8.47)	-1.505	0.132
A1 NDW	19 (16–24)	24 (18–29)	-1.681	0.093
A3 NDW	40 (30–46)	41 (37–49)	-0.875	0.382
A1 TNW	33 (24–45)	39 (29–56)	-1.471	0.141
A3 TNW	91 (72–115)	89 (71–107)	-0.311	0.755
A1 Story Structure	7 (5–11)	11 (9–12)	-3.262	0.002**
A3 Story Structure	26 (21–31)	30 (27–32)	-2.187	0.029*

TEDIL Turkish Test of Early Language Development, MLUw  Mean Length Utterance in Word, MLUm Mean Length Utterance in Morpheme, NDW Number of Different Words, TNW Total Number of Words, Q1, Q3 interquartile range

p < 0.05*, p < 0.01**

Table 3 shows no statistically significant differences in TEDIL and narrative skill scores among children with HL according to the type of amplification used (p > 0.05).

Table 3.

The Comparison of TEDIL and Narrative Skills Scores of Children With Hearing Loss According to the Type of Amplification

Variable	Type of Amplification	n	Median	Q1–Q3	Z	p
TEDIL Receptive Language Score	HA CI	15 12	87.00 87.50	83.00–99.00 83.00–94.00	-0.759	0.448
TEDIL Expressive Language Score	HA CI	15 12	95.00 89.00	87.00-100.00 84.50–95.50	-1.395	0.613
TEDIL Verbal Language Score	HA CI	15 12	89.00 89.50	83.00-101.00 80.25-95.00	-1.127	0.260
A1 MLUw	HA CI	15 12	3.67 3.14	3.00-4.22 2.69–3.77	-1.123	0.261
A3 MLUw	HA CI	15 12	4.09 4.21	3.53–4.26 3.35–4.83	-0.439	0.660
A1 MLUm	HA CI	15 12	6.17 4.93	5.00-7.29 4.39–6.87	-1.294	0.196
A3 MLUm	HA CI	15 12	7.07 7.19	6.55–7.56 5.96–8.24	-0.390	0.696
A1 NDW	HA CI	15 12	19.00 23.00	15.00–23.00 16.50-27.75	-1.003	0.316
A3 NDW	HA CI	15 12	38.00 43.50	30.00–43.00 25.00-49.25	-0.366	0.714
A1 TNW	HA CI	15 12	30.00 37.00	24.00–42.00 24.00–48.00	-0.635	0.526
A3 TNW	HA CI	15 12	90.00 102.00	72.00-115.00 54.25-132.25	-0.488	0.625
A1 Story Structure	HA CI	15 12	7.00 7.50	5.00–12.00 5.25–9.75	-0.368	0.713
A3 Story Structure	HA CI	15 12	29.00 25.50	21.00–31.00 21.00-28.50	-0.489	0.625

TEDIL Turkish Test of Early Language Development, MLUw Mean Length Utterance in Word, MLUm Mean Length Utterance in Morpheme, NDW Number of Different Words, TNW Total Number of Words, HA Hearing Aid, CI Cochlear Implant, Q1, Q3 interquartile range

p < 0.05*, p < 0.01**

The correlation analysis between demographic variables and narrative performance is presented in Table 4. In children with HL, age showed significant positive correlations with MLUw and MLUm in both the A1 story (r = 0.394, p = 0.042; r = 0.429, p = 0.026) and the A3 story (r = 0.542, p = 0.004; r = 0.560, p = 0.002). Age was also positively correlated with story structure scores for both narratives (A1: r = 0.429, p = 0.026; A3: r = 0.425, p = 0.027).

Table 4.

The correlation of the narrative skill scores of the children in the study according to age and mother education level

	Hearing Loss				Normal Hearing
	Age (month)		Mother Education Level		Age (month)		Mother Education Level
Variable	r	p	r	p	r	p	r	p
A1 MLUw	0.394	0.042*	-0.114	0.573	0.419	0.030*	0.356	0.069
A1 MLUm	0.429	0.026*	-0.079	0.695	0.458	0.016*	0.243	0.221
A1 TNW	0.129	0.522	-0.25	0.534	0.097	0.632	-0.214	0.285
A1 NDW	0.189	0.345	-0.041	0.837	0.144	0.474	-0.153	0.447
A3 MLUw	0.542	0.004*	-0.300	0.129	0.159	0.427	0.488	0.010*
A3 MLUm	0.560	0.002*	-0.306	0.121	0.207	0.301	0.364	0.062
A3 TNW	0.116	0.564	-0.016	0.937	0.001	0.997	-0.062	0.760
A3 NDW	0.086	0.670	0.085	0.674	-0.013	0.947	-0.210	0.292
A1 Story Structure	0.429	0.026*	-0.067	0.742	0.607	0.001**	0.199	0.321
A3 Story Structure	0.425	0.027*	-0.168	0.402	0.631	0.001**	0.205	0.306

TEDIL Turkish Test of Early Language Development, MLUw Mean Length Utterance in Word, MLUm Mean Length Utterance in Morpheme, NDW Number of Different Words, TNW Total Number of Words

p < 0.05*, p < 0.01**

Similarly, in the NH group, chronological age was positively correlated with MLUw (r = 0.419, p = 0.030) and MLUm (r = 0.458, p = 0.016) in the A1 story and with story structure scores for A1 (r = 0.607, p = 0.001) and A3 (r = 0.623, p < 0.001).

Maternal education level was not significantly associated with any narrative scores in the HL group (p > 0.05). However, in the NH group, a higher maternal education level was significantly associated with higher MLUw scores in the A3 story (r = 0.488, p = 0.010).

As shown in Table 5, age at diagnosis of HL was not significantly associated with any narrative skill scores in children with HL (p > 0.05).

Table 5.

The correlation of narrative skill scores of children with hearing loss according to the age of hearing loss diagnosis

Variable	Age of HL Diagnosis (month)
	r	p
A1 MLUw	-0.110	0.586
A1 MLUm	-0.035	0.863
A1 TNW	-0.042	0.834
A1 NDW	-0.149	0.457
A3 MLUw	0.054	0.787
A3 MLUm	0.034	0.866
A3 TNW	0.059	0.770
A3 NDW	-0.120	0.552
A1 Story Structure	0.170	0.396
A3 Story Structure	0.187	0.350

TEDIL Turkish Test of Early Language Development, MLUw Mean Length Utterance in Word, MLUm Mean Length Utterance in Morpheme, NDW Number of Different Words, TNW Total Number of Words

p < 0.05*, p < 0.01**

Discussion

The present study compared the narrative skills of children with HL and children with NH. Specifically, the study examined group differences in TEDIL language scores, ENNI microstructural narrative measures, and A1 and A3 story structure scores. In addition, the relationship between narrative performance and age, age at diagnosis of HL, maternal education level, and type of amplification were investigated.

Differences in between the language and narrative skills between children with HL and NH

The findings showed that children with HL had significantly lower receptive, expressive, and verbal language scores than children with NH. In terms of narrative microstructure, children with NH generally obtained higher scores across microstructural measures, including MLU, NDW, and TNW. However, statistically significant differences between groups were observed only for A1 MLUw and A1 MLUm. These findings suggest that children with HL may experience difficulties not only in general language development but also in certain aspects of narrative language production.

The present findings are consistent with previous studies reporting weaker narrative performance in children with HL. Majorano et al. [39] reported that children with CI and developmental language disorder had poorer narrative skills than children with typical development. Similarly, Zanchi et al. [40] found that the stories produced by children with CI contained fewer different words than those produced by children with NH. Zamani et al. [41] ,who compared Persian-speaking children with cochlear implants, hearing aids, and NH using “The Playful Little Elephant” story, also reported lower microstructural narrative scores in children with HL than in children with NH.

Taken together, the present findings and previous literature suggest that hearing-related differences in language development may also be reflected in narrative production, particularly in lexical diversity and syntactic or morphosyntactic complexity. However, because not all microstructural variables differed significantly between groups in the present study, these findings should be interpreted with caution.

Differences in A1 and A3 story structure scores

The present study also demonstrated that children with NH had significantly higher A1 and A3 story structure scores than children with HL. This finding suggests that children with HL may require additional support in organising the macrostructural components of narratives, such as characters, setting, initiating event, internal response, plan, attempt, outcome, and reaction. This finding is consistent with previous studies showing that children with HL may experience difficulties with story structure compared with their NH peers. Walker et al. [28] reported that children with HL showed significant deficiencies in narrative macrostructure compared with children with NH. Similarly, Muthu et al. [27] analyzed story structure components, including setting, characters, initiating event, internal plan, attempt, and outcome, and found that children with NH performed better in the use of story structure. The authors also noted that the stories produced by children with NH were richer and more complex than those produced by children with HL. Furthermore, children with HL were reported to have difficulty using the story structure effectively.

However, findings in the literature are not entirely consistent. Some studies have reported narrative performances in children with CI comparable to those of children with NH [40–44]. For example, Klein and Wie [44] reported that children with CI demonstrated similar perfomans to children with NH in a study where it was reported that children with CI showed the same performance in story macrostructure. These inconsistent findings suggest that narrative outcomes in children with HL may vary depending on factors such as auditory access, age at intervention, device use, and individual language experience.

In the present study, the HL group included children using hearing aids and cochlear implants, as well as children with different degrees of HL. In addition, variables such as age at implantation and duration of device use were not analysed in relation to narrative performance. Therefore, the heterogeneity of the HL group should be considered when interpreting the findings.

Responses to story-related questions also provided descriptive information regarding children’s ability to establish cause-and-effect relationships and make inferences. In the A1 story, which included two characters, all children in the NH group were able to establish cause-and-effect relationships, and most were able to make inferences from the story. In contrast, although most children with HL were able to establish cause-and-effect relationships, a considerably smaller proportion were able to make inferences from the story.

Similarly, in the more complex A3 story, all children with NH established cause-and-effect relationships, and most were able to make inferences. However, children with HL showed lower performance in establishing cause-and-effect relationships and making inferences. These descriptive findings suggest that children with HL may experience greater difficulty than their NH peers in inferential aspects of narrative comprehension. This interpretation is supported by Jones et al. [43], who reported that children with NH achieved higher scores on inference questions than children with HL. Similarly, Crosson and Geers [45] found that children with NH included more causal connections between events in their stories, whereas children with HL used fewer causal conjunctions. The authors suggested that this limitation may be related to delayed language development rather than a lack of understanding of cause-and-effect relationships. These findings support the clinical importance of evaluating story structure and inferential skills when assessing narrative abilities in children with HL.

Therefore, narrative assessment may provide important information beyond traditional receptive and expressive language measures. In clinical and educational settings, evaluating story structure, cause-and-effect relationships, and inference-making may contribute to a more comprehensive understanding of the language profiles of children with HL.

Relationship between demographic variables and narrative skill scores

The present study also examined the relationship between demographic variables and narrative skill scores. Age at diagnosis of HL and type of amplification were not significantly associated with narrative performance. However, age was positively associated with several narrative measures in both groups. In children with HL, age was significantly correlated with MLUw, MLUm, and story structure scores for the A1 and A3 stories. Similarly, in children with NH, age was positively associated with selected microstructural measures and story structure scores. These results indicate that narrative skills improve with age, which is consistent with the developmental nature of narrative production. Similar to the present findings, Shunhua and Tianlong [46] reported that younger children showed lower narrative performance than older children. Ralli et al. [47] also found significant age-related differences across all language skills in preschool and elementary school children, with older children outperforming their younger children. In addition, older children demonstrated better use of conjunctions and higher story structure scores than younger children. Similarly, Lee [48], who analysed the story structure of Korean children using the ENNI, reported that the use of story structure increased with age. Overall, these findings suggest that narrative skills continue to develop throughout childhood.

Maternal education level showed different patterns across groups. In the HL group, maternal education level was not significantly associated with narrative skill scores. In the NH group, however, maternal education level was positively associated with A3 MLUw. Previous studies have shown that mothers with a higher education levels tend to have greater knowledge related to child development [49]. In addition, many studies have reported a relationship between maternal education and language development [50–53]. Similarly, Murri et al. [54],who investigated the effect of of age at CI activation on narrative skills, found that the TNW and narrative complexity were weakly influenced by maternal education level, whereas MLU was more strongly associated with maternal education. The absence of a significant association between maternal education and narrative scores in the HL group may be related to the limited sample size and heterogeneity of the participants. Because the sample included children with different amplification types and HL characteristics, the specific contribution of maternal education may have been difficult to detect. Therefore, the relationship between maternal education and narrative development in children with HL should be interpreted cautiously and investigated further in larger and more homogeneous samples.

Conclusions

The findings of the present study demonstrated that children with HL performed more poorly than children with NH on measures of receptive and expressive language, as well as on selected narrative measures. In particular, children with HL obtained lower story structure scores, suggesting difficulties in organizing the macrostructural components of narratives. The descriptive findings further indicate that children with HL may experience greater difficulty in establishing cause-and-effect relationships and generating inferences from stories compared with their NH peers.

In addition, age was positively associated with several narrative measures in both groups, indicating that narrative skills improve with increasing age. Maternal education level was associated only with A3 MLUw in the NH group. These findings underscore the importance of incorporating narrative-based assessments into the evaluation of children with HL. In addition to conventional assessments of receptive and expressive language abilities, narrative assessment may provide clinically meaningful information regarding higher-level language skills. Future studies involving larger and more homogeneous samples are needed to investigate the effects of the degree of HL, type of amplification, age at implantation, and duration of device use on narrative development.

Limitations

The present study has several limitations. First, the sample size was relatively small. Although the sample size was determined based on a power analysis, studies with larger samples would provide greater statistical power and enhance the generalizability of the findings. Second, the HL group was heterogeneous with respect to amplification type and degree of HL. The sample included children who used hearing aids as well as cochlear implants, and these differences may have influenced narrative performance. Future studies should examine more homogeneous groups to better clarify the specific effects of amplification type and HL characteristics on narrative skills.

Third, variables such as age at implantation and duration of device use were not analyzed in relation to narrative performance. These factors may play an important role in explaining variability in narrative development among children with HL. Fourth, the study included only Turkish-speaking children, thereby limiting the generalizability of the findings to other linguistic populations. Future research should examine narrative skills across larger samples, diverse clinical populations, and cross-linguistic contexts. Finally, intervention studies are warranted to determine whether narrative-based educational programs can improve narrative comprehension and production in children with HL.

Abbreviations

HL

Hearing loss

NH

Normal Hearing

TEDİL

Turkish Test of Early Language Development

ENNI

The Edmonton Narrative Norms Instrument

MLU

The Mean Length Utterance

NDW

Number of Different Words

TNW

Total Number of Words

SALT

Systematic Analysis of Language Transcripts

CI

Cochlear implant

HA

Hearing Aid

Authors' contributions

Deniz Tuz: Design, literature review, data analysis, writing. Beyza Akti: Design, literature review, data analysis, writing, access to patient data, implementation of scales and tests. Pelin Piştav Akmeşe: Conception, design, literature review, data analysis, writing, access to patient data, implementation of scales and tests.

Funding

This study did not receive any external funding.

Data availability

The data that support the findings of this study are available from the first author, Akti Beyza, upon reasonable request.

Declarations

Ethics approval and consent to participate

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Medical Research Ethics Committee of Ege University (No.92659746-100-1011944). Informed written consent was obtained from the parents or legal guardians of all participants aged under 16 years.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.