Abstract This study aims to explore early reading comprehension in Chinese-speaking children with hearing loss (HL) by examining character recognition and linguistic comprehension. Twenty-five children with HL received three measures relevant to character reading: phonological awareness (PA), morphological awareness (MA), and character recognition; two linguistic-comprehension measures: receptive vocabulary knowledge and listening comprehension; and one reading comprehension measure. Three demographic variables pertinent to children with HL were also taken into account. The results showed that the degree of HL was significantly related to reading comprehension, receptive vocabulary knowledge, and listening comprehension, the latter two of which were further correlated with the type of hearing device; however, the hearing age was associated with neither of the measures. MA made a unique contribution to character reading in the presence of PA, but the reverse was not true. Linguistic comprehension significantly accounted for additional variance in reading comprehension over and above character recognition but not vice versa. A further analysis controlling for character recognition revealed that receptive vocabulary knowledge was more contributive to early reading comprehension than was listening comprehension. The results of the current investigation have potential to inform educational practice in teaching and/or intervening Chinese children with HL regarding their reading skills. Reading is a complex process encompassing many cognitive endeavors, among which word recognition and linguistic comprehension are two essential components that are required to achieve success in reading comprehension (Gough & Tunmer, 1986). While word recognition refers to the ability to read or decode written words alone without the help of context, linguistic comprehension involves the ability to understand spoken language using one’s linguistic knowledge. Deprived of auditory stimulation at birth, many children with hearing loss (HL) do not have a good command of spoken language skills, which in turn may adversely affect their reading comprehension. Children with HL have been shown to lag heavily behind their peers with typical hearing in reading achievement. This phenomenon holds true not only for the past few decades but also for a more recent era when Universal Newborn Hearing Screening becomes commonly available, and the technology of hearing aids (HA) and cochlear implants (CI) forges ahead vigorously. For example, several previous studies have revealed that students with HL aged 15–18 years only had a reading level comparable to the third to fourth grades (Allen, 1986; Holt, 1993; Traxler, 2000). A more recent study revealed no difference in reading ability between two cohorts of children with HL recruited 10 years apart, both of whom were outperformed by their peers with typical hearing (Harris, Terlektsi, & Kyle, 2017b). In other words, although identified much earlier and fitted with more state-of-the-art amplification than before, children with HL nowadays are still in a disadvantage in terms of reading achievement. Then, what factors might explain reading inferiority in children with HL? The Simple View of Reading (SVR; Gough & Tunmer, 1986) may shed important light on how children with HL might struggle with early reading. According to the SVR, word decoding or recognition constitutes the most rudimentary reading skill (Gough & Tunmer, 1986) and has been shown to be more contributive to children’s early reading comprehension, after which linguistic comprehension plays a more critical role in reading comprehension at later stages of reading development (Florit & Cain, 2011; Garcia & Cain, 2014). Previous studies on children’s reading development have proposed several precursor skills of word recognition and subcomponents of linguistic comprehension (Catts, Herrera, Nielsen, & Bridges, 2015; de Jong & van der Leij, 2002; Kendeou, van den Broek, White, & Lynch, 2009; Proctor, Carlo, August, & Snow, 2005). Among the precursor skills of word recognition, phonological awareness (PA) serves as one critical determiner of reading success. PA, the ability to reflect upon and manipulate the phonological structure of spoken words (Oakhill & Kyle, 2000), has been shown to be correlated with word recognition not only in the population with typical hearing (Hulme et al., 2002; Stahl & Murray, 1994; Torgesen, Wagner, Rashotte, Burgess, & Hecht, 1997) but in the population with HL as well (Colin, Magnan, Ecalle, & Leybaert, 2007; Dillon, de Jong, & Pisoni, 2011; Dyer, MacSweeney, Szczerbinski, Green, & Campbell, 2003; Easterbrooks, Lederberg, Miller, Bergeron, & Connor, 2008; Harris & Beech, 1998). For example, Dyer et al. (2003) found that PA (as measured by a picture rhyming task) of English-speaking adolescents with HL was significantly related to their reading delay, which was estimated by a word recoding and comprehension test. Colin et al. (2007) observed that pre-reading implicit PA of French-speaking children with HL could predict their word recognition skills after they began receiving reading instructions, even though their PA was poorer than that of their hearing peers. A longitudinal study carried out by Esterbrooks et al. (2008) tested 3- to 6-year-old English-speaking children with HL on their PA at the syllable, rime and onset levels when school started. A significant correlation between PA and word rereading skills measured at the end of school year was observed. Dillon et al. (2011), who targeted English-speaking children with HL aged 6–14 years as the study participants, noted that the correlation between PA and word reading ability survived the control for children’s age and speech perception skills. The above findings were all concerning alphabetic-language-speaking population with HL. Only a few studies have been conducted to Chinese-speaking children with HL on the relationship between PA and word/character recognition despite the well-established relation of PA to word/character reading in Chinese-speaking hearing children (Hu & Catts, 1998; Li, Shu, McBride-Chang, Liu, & Peng, 2012; Newman, Tardif, Huang, & Shu, 2011). Ching and Nunes (2015) is among the few studies that explored the concurrent contribution of PA to word recognition in Chinese-speaking children with HL. In their study, Chinese-speaking children with HL aged 7–12 years in Hong Kong were administered two PA tests, that is, onset awareness and tone awareness. The results showed that only tone awareness but not onset awareness made an independent contribution to word recognition after children’s age and nonverbal intelligence were considered. Ching and Nunes attributed this finding to the coarse-grained correspondence between characters and syllables in the Chinese language. Such a feature makes the phonological components of a syllable invisible to children; therefore, children in Hong Kong may not resort to segmental awareness to read Chinese. On the contrary, tone is the unit that is processed simultaneously with a syllable; thus, tone awareness manifests itself as a critical ability for Hong Kong children to read Chinese. Note that formal reading instructions in Hong Kong are implemented through a whole-word approach, wherein the phonological nuts and bolts of spoken words are not specified in formal reading instructions and thus are oblivious to children (Ho & Bryant, 1997). In Mainland China and Taiwan, on the other hand, children are taught phonetic decoding systems at the onset-rime level as an aide to character reading at the beginning of elementary school. Because of this, the awareness of the finer-grained segments of a syllable, such as onsets and rimes, should be more essential in character reading of children from these two Chinese societies than from Hong Kong. Indeed, many studies on hearing children have documented the importance of onset-rime awareness in recognizing characters in China and Taiwan (Hu & Catts, 1998; Li, Shu, McBride-Chang, Liu, & Peng, 2012; Newman, Tardif, Huang, & Shu, 2011). Following this line of research, PA at the onset-rime level theoretically should also play a part in character reading of Chinese-speaking children with HL in Taiwan given the similar developmental sequence of reading between children with and without HL (Chen, 2014; Geers & Hayes, 2011); however, none of the studies in Taiwan has examined the role of PA, particularly at the onset-rime level, in character recognition in children with HL. Thus, the present study would include PA as one possible correlate for character reading in Chinese-speaking children with HL in Taiwan. In addition to PA, the role of morphological awareness (MA), “the ability to reflect upon and manipulate morphemes and employ word formation rules in one’s language” (Ku & Anderson, 2003, p. 161), has also been pronounced in the literature on Chinese-speaking children’s reading development. To understand how MA could facilitate Chinese-speaking children’s character/word reading, it is first necessary to have a general idea about the orthography and morphology in the Chinese language. Chinese is generally considered to be a morphosyllabic language because a Chinese character corresponds to a syllable in tandem with a morpheme (DeFrancis, 1984). More than 80% of Chinese characters belong to phonetic-semantic compounds consisting of a phonetic radical (cuing phonological information) and a semantic radical (cuing semantic information) (for a review, see Wang, 1981). An example is 姨 /yi2/ “aunt,” which contains the phonetic radical 夷 /yi2/ and the semantic radical 女 “female-related.” Moreover, the semantic radical of a Chinese character seems to provide a more reliable cue to its recognition than does its phonetic radical (Shu, 2003; Zhu, 1987 cited from Ho, Ng, & Ng, 2003). Regarding morphology, more than 70% Chinese words are polymorphemic compound words composed of two or more single morphemes (Institute of Language Teaching and Research [of China], 1986), and the meaning of a compound word is usually derivable from its constituent morphemes. For example, the compound word 電腦 /dian4nao3/ “computer” is composed of 電 /dian4/ “electric” and 腦 /nao3/ “brain.” MA provides two major ways for Chinese-speaking children to decode or read a character. First, the morphosyllabic nature of Chinese (DeFrancis, 1984) drives Chinese-speaking children to rely substantially on MA to map morphemes onto characters. Better MA implies a better ability to draw upon one’s awareness of morphemes within a word in the oral language. Second, children could transfer their ability to analyze and synthesize constituent morphemes of a spoken word (i.e., MA) to the domain of character decoding due to the analytical characteristics enjoyed by compound words and phonetic-semantic compound characters in the Chinese language (Liu, Li, & Wong, 2017). For example, children who can decompose the spoken compound word, such as 黑板 /hei1ban3/ “blackboard,” into its morphemic constituents 黑 /hei1/ “black” and 板 /ban3/ “board” are more ready to analytically decode a phonetic-semantic compound character, such as 板 /ban3/ “board,” into its semantic radical 木 “wood” and phonetic radical 反 /fan3/. Given that the nature of the Chinese orthography and morphology is fairly different from that of alphabetic languages, such as English, many researchers of Chinese literacy development have argued that semantically-related MA may function as a more crucial precursor skill than PA for learning to read Chinese (Chen, Hao, Geva, Zhu, & Shu, 2009; Li et al., 2012; McBride-Chang, Shu, Zhou, Wat, & Wagner, 2003; Shu, 2003; Tong, McBride-Chang, Shu, & Wong, 2009; Tong et al., 2011). For example, McBride-Chang et al. (2003) explored the relative contributions of PA versus MA to character/word reading in Chinese kindergartners and second graders. It was found that MA explained all the variance in character/word reading that could be accounted for by PA and other reading-related skills, such as vocabulary, visual processing, and speeded naming. Tong et al. (2011) conducted a longitudinal study, where a group of 4-year-old kindergartners were followed by 2 years on the effect of earlier MA on later character/word recognition. It was found that Time 1 MA but not PA stayed persistent in predicting character/word recognition at both Time 2 and Time 3. Li et al. (2012) investigated the independent correlates of character recognition in a group of kindergartners and primary school children from China. The children were administered a battery of tests, including several MA measures. The results revealed that MA remained stable in predicting children’s character recognition across grades. To summarize, MA seems to be an indispensible precursor skills in developing character/word reading skills in Chinese-speaking hearing children; however, very few studies have examined a similar issue on Chinese-speaking children with HL. Since children with HL are restricted in their auditory perception ability, they may turn to other approaches to reading characters/words, such as using semantically-based MA. Ching and Nunes’s (2015) study, in addition to the effect of PA, also examined whether MA was of any consequence on word recognition in Chinese-speaking children with HL. Two MA tasks were administered to the children, namely, a morpheme identification task, which required children to distinguish homophones, and a morpheme construction task, which required children to construct words based on a designated morpheme. They found that MA, along with semantic radical awareness, could account for more individual variance in word recognition of children with HL than could PA, underscoring the importance of MA in this population. To replicate Ching and Nunes’s (2015) study, the current study would also take MA into consideration as a candidate for predicting character reading in Chinese-speaking children with HL. Apart from the two rudimentary word-decoding precursor skills mentioned above, the SVR also suggests that a more advanced cognitive ability, that is, linguistic comprehension, determines how well a child understands written texts (i.e., reading comprehension) after successfully decoding each word (Gough & Tunmer, 1986). As pointed out by Adlof, Catts, and Little (2006), linguistic comprehension is usually gauged with listening comprehension in most studies. In addition, Braze et al. (2016) and Tunmer and Chapman (2012), who employed a latent variable analysis to revisit the SVR, found that vocabulary knowledge and listening comprehension loaded on the same construct, that is, linguistic comprehension. In other words, listening comprehension as well as vocabulary knowledge constitutes two important subcomponents of linguistic comprehension. Unlike the precursor skills of word recognition that need to be dependent upon different orthographies (e.g., MA for morphosyllabic Chinese and PA for alphabetic English), the elements of linguistic comprehension should be language-universal among speakers of languages with different types of orthographies because the main channel for comprehending spoken languages is through listening. So far there have been virtually no studies on Chinese-speaking children with HL that scrutinized how reading comprehension is influenced by their listening comprehension and receptive vocabulary knowledge, still less the studies that looked at how all of the variables mentioned previously (namely, PA, MA, character recognition, vocabulary knowledge, listening comprehension) altogether shape the development of reading comprehension in Chinese-speaking children with HL. Available research on reading comprehension in Chinese-speaking children with HL has generally focused on reading interventions (Lee & Lee, 2003; Lu & Hong, 2008). For example, Lee and Lee (2003), adopting a single-subject design, reported the efficacy of reciprocal teaching in improving reading comprehension in three Chinese students with moderately severe HL. Lu and Hong (2008), who also employed a single-subject design, compared the effects of cooperative learning and self-instruction on reading comprehension in two students with severe and profound HL. They found that cooperative learning yielded better reading comprehension results than self-instruction. Although the above two studies attempted to provide some solutions to reading comprehension deficits in Chinese-speaking children with HL, the understanding of the causes for this deficit may be more important before an intervention program can be customized and implemented because these causes can guide the design of the intervention program. To untangle the complex relationships among different factors shaping the development of reading comprehension in Chinese-speaking children with HL, such as aforementioned variables, it is preferable and pertinent to first look at extant literature on Chinese-speaking hearing children’s reading comprehension. To the best of the authors’ knowledge, only a few studies on the reading comprehension development of Chinese-speaking hearing children simultaneously included word decoding and linguistic comprehension skills. For instance, Hsuan’s (2010) study recruited Chinese-speaking kindergartners as the subjects and prospectively investigated the influences of character recognition, along with its precursor skill (i.e., PA), and linguistic comprehension on later reading comprehension in elementary schools. The results indicated that preschool PA significantly predicted character recognition in the first semester of Grade 1, which predicted reading comprehension in the second semester of Grade 1. In addition, preschool linguistic comprehension as measured solely by listening comprehension also exerted significant influence on reading comprehension though the magnitude of this influence was weaker than that of character recognition. Joshi, Tao, Aaron, and Quiroz (2012) studied Chinese-speaking second and fourth graders and observed that both character recognition and listening comprehension accounted for less variance in Chinese reading comprehension at Grade 2 than at Grade 4 and that listening comprehension explained much more variance in reading comprehension at Grade 4 than at Grade 2. Although Hsuan (2010) and Joshi et al. (2012) included the two essential components of reading comprehension, namely character recognition and linguistic comprehension, in their studies, they did not take account of MA and vocabulary knowledge. The inclusion of more reading-related variables in a single study, especially the one for children of minority (e.g., children with HL), allows us to get a more comprehensive picture of the strengths and/or weaknesses in developing their reading skills. To fill this research gap and provide some potential implications for literacy instructions and/or interventions for children with HL, the purpose of the present study was to investigate the relative contributions of all the aforementioned reading-related factors to reading comprehension in Chinese-speaking children with HL. Given that some previous studies have suggested that the degree of HL, the type of hearing device and the hearing age (i.e., time since the child is fitted with amplification) may or may not to some extent influence children’s reading and/or linguistic performance (Chen & Fu, 2013; Svirsky & Meyer, 1999; Tomblin, Spencer, Flock, Tyler, & Gantz, 1999), the present study also look into what roles of children’s degree of HL, type of hearing device (i.e., HA and CI) and hearing age may play in their reading comprehension. Specifically, three research questions were addressed in the current study. (1) How are the degree of HL, the type of hearing device and the hearing age related to the abovementioned reading-related factors? (2) What are the relative contributions of PA and MA to character recognition? (3) What are the relative contributions of character recognition and linguistic comprehension to reading comprehension? Methods Participants A total of 25 Chinese-speaking second graders with HL (12 girls) were recruited from Children’s Hearing Foundation, Taiwan to participate in this study. They all suffered from sensorineural HL at birth. Their mean chronological and hearing ages at testing were 94.76 months (SD = 4.45) and 66.12 months (SD = 21.45), respectively. The degree of HL of the better unaided ear in these children ranged from mild to profound based on the international criterion for children’s hearing levels (Clark, 1981; see below). The major reason for recruiting children with mild HL is that although they seem to be less prone to developmental issues than their counterparts with more severe HL, they are nevertheless at a relative disadvantage when compared to children with typical hearing in terms of reading acquisition. For instance, Bess, Dodd-Murphy, and Parker (1998) observed that school children with mild HL were more likely to lag behind their peers with typical hearing in reading-related performance, such as listening comprehension, word usage skills and story-telling ability. In addition, their retention rate was significantly higher than that of their peers with typical hearing. In other words, mild HL may still exert a consequential influence on children’s academic and reading achievement; therefore, there is no reason for not including children with mild HL when it comes to their reading performance. All of the children were fitted with appropriate amplification, such as HA and/or CI, when being tested. The hearing levels of the aided ears in these children ranged from normal to moderate based on the international criterion for children’s hearing levels (Clark, 1981; see below). Specifically, the HA and CI groups consisted of 17 (68%) and 8 (32%) children, respectively. All of the participants were mainstreamed in school and used spoken language to interact with others in daily communications. In Taiwan, formal literacy instruction begins in elementary schools, and children receive instructions on the Mandarin Phonetic Alphabet Symbols, where they are trained to read and write the alphabetic symbols in isolation or in combination, during the first 10 weeks of Grade 1, after which the Mandarin Phonetic Alphabet Symbols are printed alongside each character on the textbooks to aide children in recognizing the characters on their own. Not until the fourth grade are the Mandarin Phonetic Alphabet Symbols removed from the characters already taught to children (Ministry of Education of Taiwan, 2011). The parental informed consents of these children were obtained prior to the implementation of this study. According to the children’s background database established by Children’s Hearing Foundation and the parental reports, none of these children had any known disorders except for HL, such as intellectual disability, autism spectrum disorders (ASD), attention deficit hyperactivity disorders (ADHD), emotional disorders, irreparable visual impairments, etc. Background Measures The Degree of HL of the Better Ear Each child’s pure-tone averages (PTA) of the hearing thresholds at 0.5, 1, 2, and 4 kHz for the better aided and unaided ear were obtained from the latest result of the hearing test administered at Children’s Hearing Foundation, Taiwan and was employed as an index for the severity of his or her HL. According to Clark’s (1981) criterion, children’s degrees of HL can be classified into seven levels (from the least to the most sever): normal (–10 to 15 dB HL), slight (16 to 25 dB HL), mild (26 to 40 dB HL), moderate (41 to 55 dB HL), moderately severe (56 to 70 dB HL), severe (71 to 90 dB HL), and profound (91 + dB HL). This criterion is a commonly used classification system all over the world. Based on the PTA of the unaided better ears, there were 4 children with mild HL, 3 children with moderate HL, 2 children with moderately severe HL, 6 children with severe HL, and 10 children with profound HL. Parental Education Level It has been suggested that parental education plays an important part in children’s reading achievements (Snow, Burns, & Griffin, 1998). Given this, parental education level for each child was included as a background variable in the current study. The education level was represented by the number of years the parents had spent in receiving formal education in Taiwan. For example, education at the elementary, junior-high, senior-high, vocational, undergraduate, and graduate levels in Taiwan is represented by 6, 9, 12, 15, 16, and 19 years, respectively. Pilot Study In the present study, two measures, Morphological Awareness Test and Listening Comprehension Test, were developed by the first author. Prior to the administration of the two tests to the children, a pilot study recruiting 30 children aged 7–9 years was conducted to select appropriate test items for these two tests. An extreme-group-comparison approach to item analysis was employed to select the items. Specifically, the participants whose score obtained from the original item pool was on the two extremes on the score scale (higher group: upper 27%; lower group: lower 27%) were compared on their performances on each item. The item that yielded significantly different scores (p < .05) between the higher and lower groups was considered to have a high power of discrimination and would be retained. Through this approach, 15 out of 30 items on the MA test and 13 out of 31 items on the listening comprehension test were shown to have a high power of discrimination and were thus selected for the present study. Regarding the reliability, the internal consistency as indexed by Cronbach’s Alpha was calculated for each of the two tests. The two tests had an acceptable Cronbach’s Alpha (i.e., above .70) based on Kline’s (1999) criterion, indicating that the selected items are all reliable and measure the same construct. Reading-related Measures The participants were administered a battery of tests measuring character recognition and linguistic comprehension, along with reading comprehension. Two important precursor skills of character recognition in Chinese, PA and MA, were assessed in the present study. In addition, receptive vocabulary knowledge and listening comprehension were gauged as two indicators of linguistic comprehension, as reasoned by Braze et al. (2016) and Tunmer and Chapman (2012). The details of each test are described below. Phonological Awareness Test Children’s PA was measured at the onset-rime level but not at the syllable level because onset-rime awareness was found in previous studies (Newman, Tardif, Huang, & Shu, 2011) to better predict elementary school children’s reading ability than syllable awareness in the Chinese societies, where the phonetic coding systems was taught to children. The onset-rime awareness test was adopted directly from Hu and Catts’s (1998) study, which required the child to identify out of three spoken syllables the one that sounded different from the other two in terms of the beginning sounds (i.e., onset) or the ending sounds (i.e., rime). There were two sets of test trials with eight in each: one set assessed onset awareness (e.g., /bi3/, /ban3/, /gou3/), and the other set assessed rime awareness (e.g., /ta1/, /po1/, /ma1/). In each set, four test trials contained a triad of spoken syllables with the same tone (e.g., /sou1/, /keng1/, /you1/), while the other four test trials contained a triad of spoken syllables with different tones (e.g., /po2/, /da4/, /ma3/). No information relevant to the occurrence of frequencies of the test spoken syllables was available in Hu and Catts’s (1998) study. The three spoken syllables were accompanied by three printed numbers, 1, 2, and 3, respectively, on a test sheet. In each trial, the experimenter said the three spoken syllables (one per second) twice while pointing to the corresponding number on the test sheet. The child had to circle the number that represented the spoken syllable that she/he thought sounded different from the other two based on the designated sound unit. Each of the two sets was preceded by three practice trials, where we made sure that each child understood the task. The feedback on the correctness of the response was given to the child in the practice trials. If the child made an error in the practice trial, she/he was given the correct answer and told why it was correct. During the test trials, no feedback was provided to the child, and she/he was asked to try his or her best to complete the task. This PA test consisted of 16 test trials and lased for about 5–8 min. One point was given for each correct response (max = 16). The test–retest reliability of this task was .78. Morphological Awareness Test Children’s MA was measured using a self-developed compound word construction task. According to Chen et al. (2009), the completion of a compound word construction task requires two major abilities: the ability to analyze a spoken word into its consistent morphemes and the ability to create novel words with the already-analyzed morphemes. Recall that MA could facilitate Chinese-speaking children in efficiently mapping morphemes onto characters and analyzing a phonetic-semantic compound character into its constituent radicals. Thus, though not directly measuring children’s character-decoding ability, MA did play an indispensible role with this regard. There were a total of 15 test trials in this test, where the experimenter orally prompted the child to say out loud a novel word based on the morphological structure of a familiar compound word. All of the familiar compound words used for the demonstration items in the prompt were chosen from the Chinese textbooks for elementary school students in Taiwan and embedded within the same carrier sentence: “… (A short description for the demonstration item) is called … (the name of the demonstration item), and what would you call … (a short description of the target item)?” For example, in one trial, the child was told “A deer that has a very long neck is called a long-neck-deer ‘giraffe,’ what would you call a deer with a very short neck?” The answer was “short-neck-deer.” Before the test trials, the child was provided with four practice trials, which required them to say a real word based on a scenario orally given by the experimenter that described another real word. If the child gave a wrong answer in response to the given scenario, the experimenter corrected her/him and explained how the answer was derived from the given scenario. For the test trials, no feedback was provided, and one point was given for each correct response (max = 15). This test lased for about 10–15 min. The internal consistency (Cronbach’s Alpha) of this test based on the pilot study was .81. Character Recognition Test Children’s ability to recognize Chinese characters was assessed using the Graded Chinese Character Recognition Test (Huang, 2004). This standardized test, designed for testing students of Grades 1–9 in Taiwan, consists of 200 Chinese characters arranged from high to low frequency. This test requires the child either to read aloud each character or to write down its pronunciation using the Mandarin Phonetic Alphabet Symbols. To avoid the articulatory problems that the children with HL may have (Eriks-Brophy, Gibson, & Tucker, 2013), the child in this study was asked to write down the Mandarin Phonetic Alphabet Symbols alongside each character. According to the test manual, the grading of this test started from the very first character and stopped right until 20 consecutive errors were made by the child. Each correct response was worth one point (max = 200). This test lasted for about 18–25 min. The test manual specifies that the internal consistency is .99 and test–retest reliabilities range from .81 to .95. Vocabulary Knowledge Test Children’s receptive vocabulary has been treated in previous studies as an essential element of linguistic comprehension (Braze et al., 2016; Tunmer & Chapman, 2012). In this study, children’s receptive vocabulary knowledge was assessed employing the Chinese version of Peabody Picture Vocabulary Test-Revised (Lu & Liu, 1998). This test was standardized based on the data of 886 children in Taiwan and was composed with 125 test trials. In this test, the child was presented with a four-picture plate, from which she/he was required to choose one picture that best fit the meaning of a word spoken by the experimenter. Two practice trials were given prior to the test trials to make sure that the child had met the minimum criterion of taking this test. One point was given for each correct response to the item tested. The child was tested from a basal of eight consecutive correct responses to a ceiling of six errors in eight consecutive test words. The children’s raw score of this test was calculated by subtracting the number of errors from the number of words attempted before the ceiling was reached (max = 125). This test lasted for about 10–12 min. The test manual specifies that the internal consistency is from .90 to .97. Listening Comprehension Test Most studies on the SVR employed listening comprehension as an index for children’s linguistic comprehension (Adlof, Catts, & Little, 2006; Catts, Adlof, & Weismer, 2006; Ouellette & Beers, 2010); therefore, children with HL in the current study were administered a listening comprehension test developed by the first author. This test is composed of five longer passages (128–193 characters long) and five shorter passages (51–107 characters long), accompanied by 13 open-ended questions concerning these passages that necessitated inference making using context clues. One point was given to each correct response (max = 13). The words used to comprise the passages and questions were specially chosen from the Chinese textbooks for elementary school students in Taiwan so that very few, if any, words would hinder the child from understanding the passages. The child was told that she/he was going to hear some interesting stories and then answer the questions about these stories. One sample passage, question and answer are given below. Passage Xiao-Ming and Xiao-Hua wanted to go camping nearby somewhere that has water today. They brought their tent and went to Yangminshan National Park. The ranger told them that there were two spots for camping: one was next to the river; the other was next to the forest. After camping, they all felt very happily satisfied. Question Where did Xia-Ming and Xiao-Hua go camping in Yangmingshan National Park? Answer Nearby the river. The child’s responses were recorded for the purpose of a later inter-rater reliability analysis. In this analysis, five children (20% of the 25 subjects) were randomly selected as the targets for computing a point-to-point percentage of agreement score for this test. The first author and a trained research assistant independently evaluated the answer to each question given by the five selected children. The percentage of agreement on this test was 93.8%. The disagreement on the grading was resolved by discussion until consensus was reached. This test lased for about 15–18 min. The internal consistency (Cronbach’s Alpha) of this test based on the pilot study is .78. Reading Comprehension Test Children’s reading comprehension was evaluated adopting a standardized paper-and-pencil test, Elementary School in 2nd Grade to 6th Grade Reading Comprehension Screening Test-2nd Grade Version (Ko & Chan, 2007). This test was commonly used in elementary schools in Taiwan to screen out children who are at risk for reading disorders. There were 19 multiple-choice items with four options each in this test: 12 for sentence comprehension and 7 for passage comprehension. Sentence comprehension assessed children’s abilities to understand word meanings and sentence propositions, whereas passage comprehension assessed their abilities to understand the whole text and infer the untold information from the contexts. The Mandarin Phonetic Alphabet Symbols were printed alongside each word of the sentences and passages. One may argue that the provision of the phonetic symbols in this comprehension test renders it rather auditory in nature; however, it is hard to exclude the possibility that children may still have to decode some, if not all, of the characters due to the abundance in homophones in Chinese. For example, the Mandarin Phonetic Alphabet Symbol /yi1/ could stand for 一 “one,” 伊 “she/he,” 醫 “medicine,” 衣 “clothes,” 依 “rely” and so forth. It would be less likely for the child to derive the precise meaning of a character only from decoding the Mandarin Phonetic Alphabet Symbols, such as /yi1/, even though they can take advantage of his or her knowledge of the spoken language. In addition, the Education Ministry of Taiwan (2011) stipulates that elementary school textbooks of all the subjects provide phonetic symbols before the fourth grade as an aide for children to learn to read. In other words, decoding the phonetic symbols is part of the reading processes in second graders in Taiwan, the target participants of the present study. The child was required to read each sentence or passage carefully and then answer the accompanied questions by choosing the correct option. No test sentences, passages, and questions were uttered by the experimenter; therefore, the child must read all of them on their own. One correct response was worth one point (max = 19). The child was allowed 25 min to complete this test. The user manual of this test indicates that the internal inconsistency (Cronbach’s Alpha) is .80. Procedures Each child was tested individually in a quiet room at Children’s Hearing Foundation. For those measures which required the child to listen, the experimenter spoke to her/him at a normal speech level of about 50–60 dB HL, which was above the child’s aided hearing level (M = 25.05 dB HL; SD = 7.96). All measures were administered to the child in two sessions. The first session included character recognition test and reading comprehension test, and the second session included phonological awareness test, morphological awareness test, vocabulary knowledge test and listening comprehension test. Each child was allowed a 10–15-min break between the two sessions depending on their personal needs. Data Analysis Two major statistical approaches to data analysis were undertaken in the present study. To investigate the interrelationships among all the variables, Pearson’s correlation was adopted. To examine the relative contribution of each independent variable to the dependent variable, a series of hierarchical multiple regression analyses were carried out. Owing that the type of hearing device belongs to a binary variable, that is, HA and CI, HA was coded as 0, and CI was coded as 1 in the following statistical analyses. Results For the tests requiring the child to listen without the aide of pictures (namely, phonological awareness test, morphological awareness test, and listening comprehension test), all of the children made it to the last item, suggesting that they could manage to complete these tests. Table 1 displays means and standard deviations of all the variables. The correlations among the variables are shown in Table 2. The results of zero-order correlations showed that parental education level was significantly related to PA, MA, vocabulary, and listening comprehension (r = .46 ~ .61, all ps < .05); therefore, parental education level was treated as a control variable in the following analyses. Both character recognition and linguistic comprehension (i.e., vocabulary knowledge and listening comprehension) were significantly related to reading comprehension before and even after parental education level was controlled for (r = .49 ~ .82, all ps < .05). Consistent with previous findings on Chinese-speaking hearing children, the results of zero-order correlations revealed that both PA and MA were significantly correlated with character recognition (r = .49 ~ .65, all ps < .05); however, the partial correlations considering the parental education level became insignificant with PA (r = .37, p > .05) but remained significant with MA (r = .57, p < .01). Table 1 Descriptive statistics for the variables Mean SD Parental education level (years) 14.12 2.88 Chronological age (months) 94.76 4.45 Hearing age (months) 66.12 21.45 Degree of hearing loss of the better unaided ear (dB HL) 72.60 28.06 Type of hearing device (HA = 0; CI = 1) 0.32 0.48 Reading comprehension (max = 19) 12.40 4.27 Character recognition (max = 200) 53.36 29.71 Phonological awareness (max = 16) 8.68 3.50 Morphological awareness (max = 15) 7.52 4.54 Vocabulary knowledge (max = 125) 72.84 20.85 Listening comprehension (max = 13) 5.28 3.58 Mean SD Parental education level (years) 14.12 2.88 Chronological age (months) 94.76 4.45 Hearing age (months) 66.12 21.45 Degree of hearing loss of the better unaided ear (dB HL) 72.60 28.06 Type of hearing device (HA = 0; CI = 1) 0.32 0.48 Reading comprehension (max = 19) 12.40 4.27 Character recognition (max = 200) 53.36 29.71 Phonological awareness (max = 16) 8.68 3.50 Morphological awareness (max = 15) 7.52 4.54 Vocabulary knowledge (max = 125) 72.84 20.85 Listening comprehension (max = 13) 5.28 3.58 Table 2 Bivariate correlations among the variables 1 2 3 4 5 6 7 8 9 10 11 1. Parental education level — .22 .24 −.17 .07 .31 .37 .61** .52** .56** .46* 2. Chronological age — — .33 .27 .37 .20 .25 .19 −.00 .06 −.06 3. Hearing age — .30 — −.15 .36 .34 −.03 .14 −.01 .25 .15 4. Degree of hearing loss — .32 −.11 — .64** −.46* −.26 −.38 −.35 −.47* −.57** 5. Type of hearing device .37 .35 .67*** — −.33 −.25 −.26 −.29 −.41* −.42* 6. Reading comprehension — .14 .28 −.43* −.38 — .55** .51** .74*** .82*** .65*** 7. Character recognition — .19 −.13 −.22 −.30 .49* — .49* .65*** .61** .55** 8. Phonological awareness — .07 −.01 −.36 −.39 .42* .37 — .66*** .74*** .51** 9. Morphological awareness — −.14 −.16 −.31 −.39 .71*** .57** .51* — .85*** .77*** 10. Vocabulary knowledge — −.08 .14 −.46* −.54** .81*** .53** .61** .79*** — .72*** 11. Listening comprehension — −.19 .04 −.56** −.52** .60** .46* .32 .70*** .62** — 1 2 3 4 5 6 7 8 9 10 11 1. Parental education level — .22 .24 −.17 .07 .31 .37 .61** .52** .56** .46* 2. Chronological age — — .33 .27 .37 .20 .25 .19 −.00 .06 −.06 3. Hearing age — .30 — −.15 .36 .34 −.03 .14 −.01 .25 .15 4. Degree of hearing loss — .32 −.11 — .64** −.46* −.26 −.38 −.35 −.47* −.57** 5. Type of hearing device .37 .35 .67*** — −.33 −.25 −.26 −.29 −.41* −.42* 6. Reading comprehension — .14 .28 −.43* −.38 — .55** .51** .74*** .82*** .65*** 7. Character recognition — .19 −.13 −.22 −.30 .49* — .49* .65*** .61** .55** 8. Phonological awareness — .07 −.01 −.36 −.39 .42* .37 — .66*** .74*** .51** 9. Morphological awareness — −.14 −.16 −.31 −.39 .71*** .57** .51* — .85*** .77*** 10. Vocabulary knowledge — −.08 .14 −.46* −.54** .81*** .53** .61** .79*** — .72*** 11. Listening comprehension — −.19 .04 −.56** −.52** .60** .46* .32 .70*** .62** — Note. *p < .05; **p < .01; ***p < .001. The zero-order correlations are shown above the diagonal, and the partial correlations are presented below the diagonal. Regarding Research Question 1, the degree of HL but not the hearing age was significantly related to reading comprehension, vocabulary knowledge, and listening comprehension before and after parental education level was partialled out (r = −.57 ~ −.43, all ps < .05), suggesting that the degree of HL was more important than the hearing age in reading and linguistic development of children with HL. As far as the type of hearing device is concerned, its relationship with reading comprehension was insignificant before and after parental education level was partialled out (r = −.38 ~ −.33, all ps > .05), but it remained negatively and significantly correlated with vocabulary knowledge and listening comprehension (r = −.54 ~ −.41, all ps < .05), indicating that the CI group (coded as 1) was more likely to be adversely affected in linguistic comprehension than was the HA group (coded as 0), and that the type of hearing device exerted more influence on children’s linguistic comprehension than on their reading comprehension. Given the above correlation data, the degree of HL, the type of hearing device, as well as parental education level were treated as control variables in the subsequent regression analyses. To answer Research Question 2, we examined the relative uniqueness of PA and MA in explaining the individual differences in character recognition by adopting hierarchical multiple regression analyses with character recognition being the dependent variable. Parental education level, along with the degree of HL and the type of hearing device, was entered first as the control variable, followed by the two awareness variables at the last two steps to examine their independent contributions to character recognition. The results are shown in Table 3. As shown in Table 3, while PA accounted for no additional variance in character recognition beyond the effects of MA and the background measures, MA survived the control for PA and the background measures in significantly explaining an additional 15% of variance in character recognition, indicating the greater role of MA relative to PA in character recognition. Table 3 Regression analyses for unique variance (R2 change) of phonological and morphological awareness predicting character recognition Steps Variables Character recognition β R2 ∆R2 1 Parental education level .03 .21 .21 Degree of hearing loss .02 Type of hearing device −.08 2 Phonological awareness .10 .37 .12 3 Morphological awareness .55* .43 .15* 2 Morphological awareness .55* .42 .21** 3 Phonological awareness .10 .43 .00 Steps Variables Character recognition β R2 ∆R2 1 Parental education level .03 .21 .21 Degree of hearing loss .02 Type of hearing device −.08 2 Phonological awareness .10 .37 .12 3 Morphological awareness .55* .43 .15* 2 Morphological awareness .55* .42 .21** 3 Phonological awareness .10 .43 .00 Note. *p < .05; ** p < .01. Research Question 3 was to explore whether character recognition or linguistic comprehension is the more influential determinant of reading comprehension success at the beginning stage of reading development in children with HL. Two sets of regression analyses controlling for parental education level, the degree of HL and the type of hearing device were carried out. In one set, character recognition was entered before vocabulary knowledge and listening comprehension (i.e., two linguistic comprehension variables). In the other set, the character recognition and linguistic comprehension variables were entered in a reverse order at the last two steps. The results are displayed in Table 4. As shown in Table 4, vocabulary knowledge and listening comprehension as a whole significantly contributed to additional 31% of variance in reading comprehension in the presence of character recognition but not the other way round, suggesting that character recognition was not as crucial as linguistic comprehension in predicting reading comprehension of children with HL at the initial stage of reading development. Table 4 Regression analyses for unique variance (R2 change) of character recognition and linguistic comprehension predicting reading comprehension Steps Variables Reading comprehension β R2 ∆R2 1 Parental education level −.30 .28 .28 Degree of hearing loss −.15 Type of hearing device .21 2 Character recognition .07 .42 .14* 3a Vocabulary knowledge .86*** .73 .31*** Listening comprehension .14 2a Vocabulary knowledge .86*** .73 .45*** Listening comprehension .14 3 Character recognition .07 .73 .00 Steps Variables Reading comprehension β R2 ∆R2 1 Parental education level −.30 .28 .28 Degree of hearing loss −.15 Type of hearing device .21 2 Character recognition .07 .42 .14* 3a Vocabulary knowledge .86*** .73 .31*** Listening comprehension .14 2a Vocabulary knowledge .86*** .73 .45*** Listening comprehension .14 3 Character recognition .07 .73 .00 Note. *p < .05; *** p < .001. aVocabulary knowledge and listening comprehension were entered together as an indicator for linguistic comprehension. Since linguistic comprehension, as observed earlier, was of greater significance than character recognition in reading comprehension, it would be interesting to further investigate the relative impacts of the two element skills of linguistic comprehension (namely, vocabulary knowledge and listening comprehension) on reading comprehension after character recognition and the background variables were considered. Again, we conducted two sets of regression analyses with reading comprehension being the dependent variable and the two element skills of linguistic comprehension being the independent variables. Vocabulary knowledge was entered prior to listening comprehension in the first set, but followed by listening comprehension in the second set. Table 5 presents the results. As shown in Table 5, vocabulary knowledge made a significantly unique variance (i.e., 24%) to reading comprehension over and above listening comprehension, but not vice versa, suggesting that vocabulary knowledge is more likely than listening comprehension to be a crux for initial ability of reading comprehension in children with HL. Table 5 Regression analyses for unique variance (R2 change) of vocabulary knowledge and listening comprehension predicting reading comprehension Steps Variables Reading comprehension β R2 ∆R2 1 Parental education level −.30 .28 .28 Degree of hearing loss −.15 Type of hearing device .21 2 Character recognition .07 .42 .14* 3 Vocabulary knowledge .86*** .72 .31*** 4 Listening comprehension .14 .73 .01 3 Listening comprehension .14 .49 .07 4 Vocabulary knowledge .86*** .73 .24*** Steps Variables Reading comprehension β R2 ∆R2 1 Parental education level −.30 .28 .28 Degree of hearing loss −.15 Type of hearing device .21 2 Character recognition .07 .42 .14* 3 Vocabulary knowledge .86*** .72 .31*** 4 Listening comprehension .14 .73 .01 3 Listening comprehension .14 .49 .07 4 Vocabulary knowledge .86*** .73 .24*** Note. *p < .05; ***p < .001. Discussion The purpose of this study was to investigate early reading comprehension in Chinese-speaking children with HL, an area left unexplored in previous research which, in most cases, solely examined their character/word recognition. In addition, the present study also explored the relative contributions of character recognition and linguistic comprehension to reading comprehension in Chinese-speaking children with HL. Furthermore, the current study inspected the precursor skills (namely, PA and MA) of character recognition and the subcomponents (namely, receptive vocabulary and listening comprehension) of linguistic comprehension, and the interrelationships among the aforementioned variables. With the confounding effect of parental education level considered, this study yielded three major findings: (1) While the degree of HL was not significantly correlated with PA, MA and character recognition, it was significantly related to receptive vocabulary knowledge, listening comprehension, and reading comprehension. Similarly, the type of hearing device did not have any connection with PA, MA, character recognition, and reading comprehension, but its relationship to vocabulary knowledge and listening comprehension was substantial. The hearing age was not associated with any of the reading-related skills mentioned above; (2) MA, but not PA, made a unique contribution to character reading in children with HL in the presence of the other even after the effect of the degree of HL and the type of hearing device was accounted for; and (3) linguistic comprehension significantly explained additional variance in early reading comprehension over and above character recognition, along with the background variables, in children with HL but not vice versa. A further analysis controlling for character recognition and the background variables revealed that vocabulary knowledge was more contributive to early reading comprehension in children with HL than was listening comprehension. The Effects of the Degree of HL, the Type of Hearing Device, and the Hearing Age In the present study, the degree of HL but not the hearing age was important in determining children’s performance on the reading-related measures, specifically linguistic and reading comprehension. This study not only replicates Chen and Fu's (2013) finding that the hearing age was not a good predictor of linguistic comprehension of Chinese children with HL, but it also extends its significance to reading comprehension of Chinese children with HL. The significant effect of the degree of HL and insignificant effect of the hearing age suggest that it might be the listening quality (as indexed by the degree of HL) rather than listening quantity (as indexed by the hearing age) that is better in determining reading achievement in children with HL. Then, how might the significant effect of the degree of HL be explained? Recall that all of the children with HL in this study differed substantially in the degree of HL (from mild to profound) in the unaided better ears, implying that the severity of their hearing impairment may vary considerably at birth. It has been suggested that various degrees of HL may have different impacts on the exposure to linguistic information in children with HL such that the more severe a child’s HL is, the less and poorer linguistic information she/he will receive (Anderson & Matkin, 2007). Poor linguistic comprehension, as a consequence, may lead to poor reading comprehension (Gough & Tunmer, 1986). Therefore, it is very likely that the negative effect of unaided hearing levels prior to the time of amplification fitting in children with HL is so strong and long-lasting that it surpasses the effect of the hearing age in deleteriously affecting their linguistic and reading comprehension. On the other hand, the type of hearing device was a negative correlate of children’s linguistic comprehension with the CI group being more inclined to have poorer vocabulary knowledge and listening comprehension than was the HA group. This finding is inconsistent with previous studies showing that children with CI as a whole performed better than their HA counterparts in language development (Svirsky & Meyer, 1999; Tomblin et al., 1999). One possible reason for this inconsistency is that previous studies recruited children with HA or CI who all suffered from severe to profound HL. It is known that once children’s HL is too severe for HA to function well, CI becomes a better alternative for aiding children’s hearing. Thus, previous studies in general found the CI advantage over HA in terms of language development. The children in the current study, however, varied considerably in their degrees of HL (from mild to profound) and were fitted with HA or CI accordingly; that is, children with milder HL tended to wear HA, and children with more severe HL tended to wear CI, as demonstrated by the significantly positive moderate correlation of .64 between the type of hearing device and the degree of HL. Given this, it might be the degree of HL rather than the type of hearing device that predominantly determines children’s performances on vocabulary knowledge and listening comprehension such that the milder HL group with HA outperformed the more severe HL group with CI in their linguistic comprehension. In other words, the observed negative correlation between the type of hearing device and vocabulary knowledge/listening comprehension may just result from children’s degree of HL. The Relative Roles of PA and MA in Character Recognition in Children with HL Similar to the findings on Chinese-speaking hearing children (Chen et al., 2009; Li et al., 2012; McBride-Chang et al., 2003; Shu, 2003; Tong et al., 2009; Tong et al., 2011) and children with HL (Ching & Nunes, 2015), MA was found to play a more important role than PA in character recognition of Chinese-speaking children with HL in the present study. There are some possible reasons for this finding. One reason could lie in the fact that children with HL may resort to orthography while bypassing phonological coding in recognizing characters due to their hearing constraints (Wang, Trezek, Luckner, & Paul, 2008). In other words, auditory limitations of children with HL might lead them to rely exclusively on the nonphonological route (i.e., MA), rather than on the phonological route (i.e., PA), in recognizing a character. However, this interpretation is not without a problem. If children with HL in the current study had only taken advantage of MA, instead of PA, in character recognition, PA should not have been significantly related to character recognition in the result of zero-order correlations; thus, the observed disadvantage of PA in character recognition may just be a relative one. Another reason, according to Tong et al. (2009), is that the retrieval of a single character is sometimes facilitated by the awareness of the morphological structure in multi-character words containing that single character. Consistent with this possibility, in the character recognition test, there were some episodic cases where children attempted to retrieve a single character by referring to the word containing that character whether their attempts succeeded or failed. For example, some children mis-decoded the character 秩 /zhi4/ as 序 /xu4/, of which the two characters form a legal Chinese bimorphemic words 秩序 /zhi4xu4/ “order.” The other possible reason, as suggested by Chen et al. (2009), is that the sensitivity to morphemes in the oral language, which is necessary to succeed in the MA test used in this study, helps children map morphemes onto characters more efficiently. Recall that the MA test required children to create a novel word (e.g., 喝者 /he1zhe3/ “drinker”) by extracting the morpheme (e.g., 者 /zhe3/ “person”) from a real spoken word (e.g., 讀者 /du2zhe3/ “reader”) and combining it with another morpheme (e.g., 喝 /he1/ “drink”). Better performance on this task implies higher sensitivity to morphemes in the oral language, which, in turn, facilitates character recognition by allowing an efficient mapping between morphemes and characters. Still other reason is relevant to the compounding mechanism shared by Chinese polymorpehmic words and phonetic-semantic characters. As has been mentioned earlier, Liu, Li, and Wong (2017) argued that the ability to tease apart a polymorpehmic word into its morphemes (e.g., breaking 讀者 /du2zhe3/ “reader” into 讀 /du2/ “read” and 者 /zhe3/ “person”) and then amalgamate one of its morphemes with another morpheme to form a new polymorphemic word (e.g., combining 喝 /he1/ “drink” with 者 /zhe3/ “person” into 喝者 /he1zhe3/ “drinker”) is transferrable to the analysis and synthesis of a phonetic-semantic compound character, thereby resulting in efficient identification of a character. Finally, the fact that homophonic characters abound in the Chinese language may render PA pretty unreliable in recognizing them, which subsequently drives Chinese-speaking children with HL to make more use of MA in dealing with the sound ambiguities among homophones. The Relative Contributions of Character Recognition and Linguistic Comprehension to Reading Comprehension Consistent with the research results on Chinese-speaking hearing children shown (Hsuan, 2010; Joshi et al., 2012), this study also showed that character recognition and linguistic comprehension both significantly contributed to reading comprehension of children with HL in the absence of the other; however, the results of their relative contributions of the two components were in contrast to those of previous studies (Florit & Cain, 2011; Garcia & Cain, 2014) in that character recognition of children with HL played a lesser unique role in early stages of reading comprehension when gauged against linguistic comprehension. Note that the Mandarin Phonetic Alphabet Symbols were provided in the reading comprehension test employed in the present study. This being the case, it is likely that some children with HL decoded the Mandarin Phonetic Alphabet Symbols of the characters. With the resultant phonological information conveyed by the phonetic codes of characters, they can further make use of their oral language or linguistic experience to derive the meanings of characters. Yet, the wealth of homophones in the Chinese language makes it highly possible that some, if not all, of the homophonic characters in the reading comprehension test still needed to be orthographically decoded especially when children’s oral language experience fails to disambiguate the meanings of those homophonic characters. Generally, children with HL in this study may achieve reading comprehension not entirely by decoding the phonetic symbols alone but partly by decoding the characters. In this context, we can only conservatively argue that the role of character recognition in children’s initial reading comprehension might be underestimated in the current study. However, Chan and Yang (2016) found that Chinese-speaking second graders with HL did not differ from their hearing peers in character recognition but still gained significantly lower reading comprehension scores. They argued that it might be children’s linguistic comprehension that was more responsible for their reading comprehension performance. Similar outcomes and arguments can be seen in Wauters, Van Bon, and Tellings (2006), who showed that Dutch-speaking children with HL, albeit equal in their word identification ability when compared with their hearing peers, were otherwise lower in their reading comprehension. They also argued that poor linguistic comprehension in children with HL may give rise to their inferiority in reading comprehension. Taken together, the above findings suggest that linguistic comprehension may be more influential than character recognition in early reading comprehension in Chinese-speaking children with HL in Taiwan. Interestingly enough, after variance due to character recognition and background variables were controlled for, vocabulary knowledge explained all variance in children’s early reading comprehension that can be accounted for by their listening comprehension. This result is not in line with those of the studies on hearing second graders conducted by Nation and Snowling (2004) and Hagtvet (2003), who found that linguistic comprehension (as assessed by listening comprehension) was of greater importance than vocabulary knowledge in contributing to reading comprehension in hearing children. These inconsistent results may reflect the fact that children with HL, like their younger hearing peers, may primarily rely more on vocabulary knowledge, the more rudimentary cognitive ability, than on listening comprehension, the more advanced cognitive ability, in early reading comprehension. This result echoes Vauras, Kinnunen, and Kuusela’s (1994) findings that young children tended to process written text using the more micro-level skills (e.g., to read in a linear element-by-element manner), while order children showed a greater propensity for using the more marco-level skills (e.g., background knowledge) to process written text. On the other hand, the present finding is in line with Harris, Terlektsi, and Kyle’s (2017a) study showing that vocabulary knowledge functioned as a key media through which deaf children made inferences in comprehending a written passage. Recall that some of the questions from the reading comprehension test adopted in this study required inference-making skills, which may account for the importance of vocabulary knowledge in explaining individual differences in early reading comprehension of children with HL in the current study. Limitations and Future Directions The present study was among the first studies that investigated reading comprehension, along with its component skills, in Chinese-speaking children with HL; however, there exist some limitations regarding the sample and the measures adopted. First, the small sample size makes impossible the generalization of the results to other Chinese-speaking children with HL. Children with HL are known to be a heterogeneous population (Spencer & Marschark, 2010); therefore, the results of the present study may only be at best applied to children with HL who have similar demographic features as shown in this study. Second, the reading comprehension test used in this study may not be the perfect one to measure children’s understanding of the written text since the Mandarin Phonetic Alphabet Symbols were printed alongside the characters. The relatively minor contribution of character recognition (in contrast to linguistic comprehension) to early reading comprehension observed in the current study may just be the artifact of children’s phonetic decoding of the characters. For children with HL in Hong Kong, it could be expected that character recognition may be of greater importance than linguistic comprehension in developing early reading comprehension because they are not initially taught using phonetic character symbols to learn to read; instead, they receive a whole-word reading instruction. This issue warrants further research. Future studies should employ a reading comprehension test that does not provide any phonological cue for the characters in order to examine the real uniqueness of character recognition and linguistic comprehension in predicting reading comprehension in Chinese-speaking children with HL in Taiwan. Third, no causal relationships among the variables could be established due to the cross-sectional and correlational nature of the current study. For the causal relationship to be established, future research should carry out a longitudinal intervention study that includes both a control and experimental group. Conclusion In conclusion, the results of the current study add to our understanding of the developmental course of early reading comprehension in Chinese children with HL. The degree of HL was more critical than the hearing age and the type of hearing device in determining linguistic and/or reading comprehension in Chinese-speaking children with HL. The effect of PA on character recognition was canceled out by the effect of MA, the awareness that has been proved more contributive to Chinese reading. In addition, linguistic comprehension seemed to dominate over character recognition in early reading comprehension in Chinese-speaking children with HL; however, this conclusion warrants further confirmation in future research using the “purer” measure of reading comprehension. These results may further inform educational practice in teaching prerequisite reading skills to children with HL who have residual hearing, are fitted with appropriate amplification, are mainstreamed and use spoken language in daily communications, as the cases in the present study. For example, since the orally-based MA task used in the present study was found to make a greater contribution than PA to character recognition, the contemporary reading education for Chinese-speaking children with HL in Taiwan may incorporate the oral training of MA in addition to the auditory training therapy (e.g., PA training) to help them more efficiently recognize characters. Moreover, given the essentiality of linguistic comprehension in initial reading comprehension in children with HL, it is recommended that the reading training and/or intervention program designed for children with HL emphasize the instructions on vocabulary knowledge and listening comprehension. Conflict of Interest No conflicts of interest were reported. References Adlof, S. M., Catts, H. W., & Little, T. D. ( 2006). Should the simple view of reading include a fluency component? Reading and Writing , 19, 933– 958. doi:10.1007/s11145-006-9024-z. Google Scholar CrossRef Search ADS Allen, T. E. ( 1986). Patterns of academic achievement among hearing impaired students: 1974 and 1983. In Schildroth A. N., & Karchmer M. A. (Eds.), Deaf children in America (pp. 161– 206). San Diego, CA: College Hill Press. Anderson, K., & Matkin, N. ( 2007). Relationship of hearing loss to listening and learning needs. Retrieved from http://www.janemadell.com/publications/Anderson.pdf. Bess, F. H., Dodd-Murphy, J., & Parker, R. A. ( 1998). Children with minimal sensorineural hearing loss: prevalence, educational performance, and functional status. Ear and Hearing , 19, 339– 354. doi:10.1097/00003446-199810000-00001. Google Scholar CrossRef Search ADS PubMed Braze, D., Katz, L., Magnuson, J. S., Mencl, W. E., Tabor, W., Van Dyke, J. A.… Shankweiler, D. P. ( 2016). Vocabulary does not complicate the simple view of reading. Reading and Writing , 29, 435– 451. doi:10.1007/s11145-015-9608-6. Google Scholar CrossRef Search ADS PubMed Catts, H. W., Adlof, S., & Weismer, S. E. ( 2006). Language deficits in poor comprehenders: A case for the simple view of reading. Journal of Speech, Language, and Hearing Research , 49, 278– 293. doi:10.1044/1092-4388(2006/023). Google Scholar CrossRef Search ADS Catts, H. W., Herrera, S., Nielsen, D. C., & Bridges, M. S. ( 2015). Early prediction of reading comprehension within the simple view framework. Reading and Writing , 28, 1407– 1425. doi:10.1007/s11145-015-9576-x. Google Scholar CrossRef Search ADS Chan, Y.-C., & Yang, Y.-J. (March 2016). An analysis of the difference in reading development between low-graders with and without hearing loss. Poster presented at the 2016 Annual Conference of Taiwan Academy for Learning Disabilities, Nantou County, Taiwan. Chen, P.-H., & Fu, J.-T. (October 2013). Prediction of outcomes of language assessment by duration of early intervention and hearing age for children with hearing loss. Paper presented at the 2013 Conference of Taiwan Speech-Language-Hearing Association, Taipei, Taiwan. Chen, X., Hao, M., Geva, E., Zhu, J., & Shu, H. ( 2009). The role of compound awareness in Chinese children’s vocabulary acquisition and character reading. Reading and Writing: An Interdisciplinary Journal , 22, 615– 631. doi:10.1007/s11145-008-9127-9. Google Scholar CrossRef Search ADS Chen, Y.-H. ( 2014). Effects of phonological awareness training on early Chinese reading of children who are deaf and hard of hearing. The Volta Review , 114, 85– 100. Ching, B. H. H., & Nunes, T. ( 2015). Concurrent correlates of Chinese word recognition in deaf and hard-of-hearing children. Journal of Deaf Studies and Deaf Education , 20, 172– 190. doi:10.1093/deafed/env003. Google Scholar CrossRef Search ADS PubMed Clark, J. G. ( 1981). Uses and abuses of hearing loss classification. Asha , 23, 493– 500. Google Scholar PubMed Colin, S., Magnan, A., Ecalle, J., & Leybaert, J. ( 2007). Relation between deaf children’s phonological skills in kindergarten and word recognition performance in first grade. Journal of Child Psychology and Psychiatry , 48, 139– 146. doi:10.1111/j.1469-7610.2006.01700.x. Google Scholar CrossRef Search ADS PubMed de Jong, P. F., & van der Leij, A. ( 2002). Effects of phonological abilities and linguistic comprehension on the development of reading. Scientific Studies of Reading , 6, 51– 77. doi:10.1207/S1532799XSSR0601_03. Google Scholar CrossRef Search ADS DeFrancis, J. ( 1984). The Chinese language: Fact and fantasy . Honolulu, HI: University of Hawaii Press. Dillon, C. M., de Jong, K., & Pisoni, D. B. ( 2011). Phonological awareness, reading skills, and vocabulary knowledge in children who use cochlear implants. Journal of Deaf Studies and Deaf Education , 17, 205– 226. doi:10.1093/deafed/enr043. Google Scholar CrossRef Search ADS PubMed Dyer, A., MacSweeney, M., Szczerbinski, M., Green, L., & Campbell, R. ( 2003). Predictors of reading delay in deaf adolescents: The relative contributions of rapid automatized naming speed and phonological awareness and decoding. Journal of Deaf Studies and Deaf Education , 8, 215– 229. doi:10.1093/deafed/eng012. Google Scholar CrossRef Search ADS PubMed Easterbrooks, S. R., Lederberg, A. R., Miller, E. M., Bergeron, J. P., & Connor, C. M. ( 2008). Emergent literacy skills during early childhood in children with hearing loss: Strengths and weaknesses. The Volta Review , 108, 91. Eriks-Brophy, A., Gibson, S., & Tucker, S. K. ( 2013). Articulatory error patterns and phonological process use of preschool children with and without hearing loss. The Volta Review , 113, 87– 125. Florit, E., & Cain, K. ( 2011). The simple view of reading: Is it valid for different types of alphabets? Educational Psychology Review , 23, 553– 576. doi:10.1007/s10648-011-9175-6. Google Scholar CrossRef Search ADS Garcia, J. R., & Cain, K. ( 2014). Decoding and reading comprehension: A meta-analysis to identify which reader and assessment characteristics influence the strength of the relationship in English. Review of Educational Research , 84, 74– 111. doi:10.3102/0034654313499616. Google Scholar CrossRef Search ADS Geers, A. E., & Hayes, H. ( 2011). Reading, writing, and phonological processing skills of adolescents with 10 or more years of cochlear implant experience. Ear and Hearing , 32, 49S– 59S. doi:10.1097/AUD.0b013e3181fa41fa. Google Scholar CrossRef Search ADS PubMed Gough, P. B., & Tunmer, W. E. ( 1986). Decoding, reading, and reading disability. Remedial and Special Education , 7, 6– 10. doi:10.1177/074193258600700104. Google Scholar CrossRef Search ADS Hagtvet, B. E. ( 2003). Listening comprehension and reading comprehension in poor decoders: Evidence for the importance of syntactic and semantic skill as well as phonological skills. Reading and Writing: An Interdisciplinary Journal , 16, 505– 539. doi:10.1023/A:1025521722900. Google Scholar CrossRef Search ADS Harris, M., & Beech, J. R. ( 1998). Implicit phonological awareness and early reading development in prelingually deaf children. The Journal of Deaf Studies and Deaf Education , 3, 205– 216. doi:10.1093/oxfordjournals.deafed.a014351. Google Scholar CrossRef Search ADS PubMed Harris, M., Terlektsi, E., & Kyle, F. E. ( 2017a). Concurrent and longitudinal predictors of reading for deaf and hearing children in primary school. The Journal of Deaf Studies and Deaf Education , 22, 233– 242. doi:10.1093/deafed/enw101. Google Scholar CrossRef Search ADS PubMed Harris, M., Terlektsi, E., & Kyle, F. E. ( 2017b). Literacy outcomes for deaf and hard of hearing primary school children: A cohort comparison study. Journal of Speech, Language, and Hearing Research , 60, 701– 711. doi:10.1044/2016_JSLHR-H-15-0403. Google Scholar CrossRef Search ADS Ho, C. S.-H., & Bryant, P. ( 1997). Phonological skills are important in learning to read Chinese. Developmental Psychology , 33, 946– 951. doi:10.1037/0012-16220.127.116.116. Google Scholar CrossRef Search ADS PubMed Ho, C. S.-H., Ng, T. T., & Ng, W. K. ( 2003). A “radical” approach to reading development in Chinese: The role of semantic radicals and phonetic radicals. Journal of Literacy Research , 35, 849– 878. doi:10.1207/s15548430jlr3503_3. Google Scholar CrossRef Search ADS Holt, J. A. ( 1993). Stanford Achievement Test—8th edition: Reading comprehension subgroup results. American Annals of the Deaf , 138, 172– 175. doi:10.1353/aad.2012.0684. Google Scholar CrossRef Search ADS Hsuan, C.-H. ( 2010). Prediction of Chinese reading difficult children in kindergarten and first grade. National Science Council Report: NSC 97-2410-H-468-024-MY2 Hu, C. F., & Catts, H. W. ( 1998). The role of phonological processing in early reading ability: What we can learn from Chinese. Scientific Studies of Reading , 2, 55– 79. doi:10.1207/s1532799xssr0201_3. Google Scholar CrossRef Search ADS Huang, H.-S. ( 2004). Graded Chinese character recognition test . Taipei, Taiwan: Psychological Publishing. Hulme, C., Hatcher, P. J., Nation, K., Brown, A., Adams, J., & Stuart, G. ( 2002). Phoneme awareness is a better predictor of early reading skill than onset-rime awareness. Journal of Experimental Child Psychology , 82, 2– 28. doi:10.1006/jecp.2002.2670. Google Scholar CrossRef Search ADS PubMed Institute of Language Teaching and Research [of China] ( 1986). A frequency dictionary of modern Chinese . Beijing, China: Beijing Language Institute Press. Joshi, R. M., Tao, S., Aaron, P. G., & Quiroz, B. ( 2012). Cognitive component of componential model of reading applied to different orthographies. Journal of Learning Disabilities , 45, 480– 486. doi:10.1177/0022219411432690. Google Scholar CrossRef Search ADS PubMed Kendeou, P., van den Broek, P., White, M., & Lynch, J. S. ( 2009). Predicting reading comprehension in early elementary school: The independent contributions of oral language and decoding skills. Journal of Educational Psychology , 101, 765– 778. doi:10.1037/a0015956. Google Scholar CrossRef Search ADS Kline, P. ( 1999). The handbook of psychological testing ( 2nd ed.). London, UK: Routledge. Ko, H.-W., & Chan, Y.-L. ( 2007). Elementary school in 2nd grade to 6th grade reading comprehension screening test . Taipei, Taiwan: Ministry of Education. Ku, Y.-M., & Anderson, R. C. ( 2003). Development of morphological awareness in Chinese and English. Reading and Writing , 16, 399– 422. doi:10.1023/A:1024227231216. Google Scholar CrossRef Search ADS Lee, Z.-T., & Lee, P.-C. ( 2003). Jiāohù jiāoxué fǎ duì zēngjìn tīngjué zhàngài xuéshēng yuèdú lǐjiě nénglì zhī yánjiù (The research of reciprocal teaching on improving the reading comprehension abilities of the students with hearing impairments). Tèshū Jiāoyù yǔ Fùjiàn Xuébào , 11, 127– 152. Li, H., Shu, H., McBride-Chang, C., Liu, H., & Peng, H. ( 2012). Chinese children’s character recognition: Visuo-orthographic, phonological processing and morphological skills. Journal of Research in Reading , 35, 287– 307. doi:10.1111/j.1467-9817.2010.01460.x. Google Scholar CrossRef Search ADS Liu, D., Li, H., & Wong, K. S. R. ( 2017). The anatomy of the role of morphological awareness in Chinese character learning: The mediation of vocabulary and semantic radical knowledge and the moderation of morpheme family size. Scientific Studies of Reading , 21, 210– 224. doi:10.1080/10888438.2017.1278764. Google Scholar CrossRef Search ADS Lu, L., & Liu, M.-X. ( 1998). Peabody vocabulary test—Revised . Taipei, Taiwan: Psychological Publishing. Lu, Y.-H., & Hong, C.-I. ( 2008). Hézuò xuéxí yǔ zìwǒ jiāodǎo duì tīngjué zhàngài értóng yuèdú lǐjiě chéngxiào zhī yánjiù (A study on the effects of cooperative learning and self-instruction on the reading comprehension of hearing-impaired children). Dōng Táiwān Tèshū Jiāoyù Xuébào , 10, 93– 119. McBride-Chang, C., Shu, H., Zhou, A. B., Wat, C. P., & Wagner, R. K. ( 2003). Morphological awareness uniquely predicts young children’s Chinese character recognition. Journal of Educational Psychology , 95, 743– 751. doi:10.1037/0022-0618.104.22.1683. Google Scholar CrossRef Search ADS Ministry of Education of Taiwan ( 2011). Ministry of Education’s standards of printing textbooks for elementary schools and junior high schools. The Executive Yuan Gazette , 24, 1. Nation, K., & Snowling, M. J. ( 2004). Beyond phonological skills: Broader language skills contribute to the development of reading. Journal of Research in Reading , 27, 342– 356. doi:10.1111/j.1467-9817.2004.00238.x. Google Scholar CrossRef Search ADS Newman, E. H., Tardif, T., Huang, J., & Shu, H. ( 2011). Phonemes matter: The role of phoneme-level awareness in emergent Chinese readers. Journal of Experimental Child Psychology , 108, 242– 259. doi:10.1016/j.jecp.2010.09.001. Google Scholar CrossRef Search ADS PubMed Oakhill, J., & Kyle, F. ( 2000). The relation between phonological awareness and working memory. Journal of Experimental Child Psychology , 75, 152– 164. doi:10.1006/jecp.1999.2529. Google Scholar CrossRef Search ADS PubMed Ouellette, G., & Beers, A. ( 2010). A not-so-simple view of reading: How oral vocabulary and visual-word recognition complicate the story. Reading and Writing , 23, 189– 208. doi:10.1007/s11145-008-9159-1. Google Scholar CrossRef Search ADS Proctor, C. P., Carlo, M., August, D., & Snow, C. ( 2005). Native Spanish-speaking children reading in English: Toward a model of comprehension. Journal of Educational Psychology , 97, 246– 256. doi:10.1037/0022-0622.214.171.124. Google Scholar CrossRef Search ADS Shu, H. ( 2003). Chinese writing system and learning to read. International Journal of Psychology , 38, 274– 285. doi:10.1080/00207590344000060. Google Scholar CrossRef Search ADS Snow, C. E., Burns, M. S., & Griffin, P. (Eds.) ( 1998). Preventing reading difficulties in young children . Washington, DC: National Academy Press. Spencer, P. E., & Marschark, M. ( 2010). Evidence-based practice in educating deaf and hard-of-hearing students . New York, NY: Oxford University Press. Stahl, S. A., & Murray, B. A. ( 1994). Defining phonological awareness and its relationship to early reading. Journal of Educational Psychology , 86, 221– 234. doi:10.1037/0022-06126.96.36.199. Google Scholar CrossRef Search ADS Svirsky, M. A., & Meyer, T. A. ( 1999). Comparison of speech perception in pediatric Clarion® cochlear implant and hearing aid users. Annals of Otology, Rhinology & Laryngology , 108, 104– 109. doi:10.1177/00034894991080S421. Google Scholar CrossRef Search ADS Tomblin, J. B., Spencer, L., Flock, S., Tyler, R., & Gantz, B. ( 1999). A comparison of language achievement in children with cochlear implants and children using hearing aids. Journal of Speech, Language, and Hearing Research , 42, 497– 511. doi:10.1044/jslhr.4202.497. Google Scholar CrossRef Search ADS Tong, X., McBride-Chang, C., Shu, H., & Wong, A. M.-Y. ( 2009). Morphological awareness, orthographic knowledge, and spelling errors: Keys to understanding early Chinese reading acquisition. Scientific Studies of Reading , 13, 426– 452. doi:10.1080/10888430903162910. Google Scholar CrossRef Search ADS Tong, X., McBride-Chang, C., Wong, A. M.-Y., Shu, H., Retisma, P., & Rispens, J. ( 2011). Longitudinal predictors of very early Chinese literacy acquisition. Journal of Research in Reading , 34, 315– 332. doi:10.1111/j.1467-9817.2009.01426.x. Google Scholar CrossRef Search ADS Torgesen, J. K., Wagner, R. K., Rashotte, C. A., Burgess, S., & Hecht, S. ( 1997). Contributions of phonological awareness and rapid automatic naming ability to the growth of word-reading skills in second-to fifth-grade children. Scientific Studies of Reading , 1, 161– 185. doi:10.1207/s1532799xssr0102_4. Google Scholar CrossRef Search ADS Traxler, C. B. ( 2000). The Stanford Achievement Test, 9th edition: National norming and performance standards for deaf and hard-of-hearing students. Journal of Deaf Studies and Deaf Education , 5, 337– 348. doi:10.1093/deafed/5.4.337. Google Scholar CrossRef Search ADS PubMed Tunmer, W. E., & Chapman, J. W. ( 2012). The simple view of reading redux: Vocabulary knowledge and the independent components hypothesis. Journal of Learning Disabilities , 45, 453– 466. doi:10.1177/0022219411432685. Google Scholar CrossRef Search ADS PubMed Vauras, M., Kinnunen, R., & Kuusela, L. ( 1994). Development of text-processing skills in high-, average-, and low-achieving primary school children. Journal of Literacy Research , 26, 361– 389. doi:10.1080/10862969409547859. Wang, W. S.-Y. ( 1981). Language structure and optimal orthography. In Tzeng O. J. L., & Singer H. (Eds.), Perception of print: Reading research in experimental psychology (pp. 223– 236). Hillsdale, NJ: Lawrence Erlbaum. Wang, Y., Trezek, B. J., Luckner, J. L., & Paul, P. V. ( 2008). The role of phonology and phonologically related skills in reading instruction for students who are deaf or hard of hearing. American Annals of the Deaf , 153, 396– 407. doi:10.1353/aad.0.0061. Google Scholar CrossRef Search ADS PubMed Wauters, L. N., Van Bon, W. H., & Tellings, A. E. ( 2006). Reading comprehension of Dutch deaf children. Reading and Writing , 19, 49– 76. doi:10.1007/s11145-004-5894-0. Google Scholar CrossRef Search ADS Zhu, Y.-P. ( 1987). Analysis of cuing functions of the phonetic in modern China. Unpublished manuscript, East China Normal University. © The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: firstname.lastname@example.org.
The Journal of Deaf Studies and Deaf Education – Oxford University Press
Published: Jan 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera