Unlike Matilda, children with Down syndrome are likely to experience reading and spelling difficulties. The two studies reported in this thesis examined which linguistic and cognitive skills predicted the reading comprehension and spelling performance of school-aged children with Down syndrome living in a multilingual environment. Each of these investigations are the first of their kind. They offer insight into the nature of reading and spelling difficulties in children with Down syndrome while also provoking future research. This final chapter identifies what was learned from both investigations and presents theoretical and clinical implications and areas of future research.
Study one: Predictors of reading comprehension in Down syndrome
There are few investigations of reading comprehension abilities at passage level in children with Down syndrome and none, to the author's knowledge, that situate their research within a multilingual environment. Of the available studies, it is difficult to directly compare reading comprehension results because stimuli have been read aloud by the examiner (e.g., Levorato, Roch & Beltram, 2009), and testing formats have varied from picture pointing responses (e.g., Levorato, Roch & Beltram, 2007) to close sentence tasks (e.g., Woodcock, McGrew & Mather, 2001). Study one presented in this thesis used the York Assessment of Reading for Comprehension (YARC) (Snowling, et al, 2009), a relatively new test that was standardised on a representative sample of English speaking children from various ethnic backgrounds living in the United Kingdom, including pupils with English as a second language (comprising 14.3% of the sample). Children were asked to read and then respond to comprehension questions posed by the examiner. Passages were also culturally neutral or had a mixed content reflective of a multi-cultural population. The YARC was therefore determined as the most appropriate reading comprehension tool for children with Down syndrome living in Singapore (a country rich in language diversity and use).
The Singapore Bilingual Vocabulary Test (Rickard Liow & Sze, 2008) which assessed receptive vocabulary was also chosen because it too best suited the cultural and linguistic context of Singapore. With these considerations in mind, six linguistic and cognitive predictors of receptive vocabulary, phonological awareness, phonological short-term working memory, listening comprehension, non-verbal intelligence and age were selected for correlation and regression analysis. Thirty eight children with Down syndrome participated in this study, a modest sample size for statistical purposes but an impressive sample size with respect to recruitment success within the developmental disabilities literature. The results of the study identified receptive vocabulary, phonological awareness (i.e., phoneme isolation and phoneme deletion) and chronological age to be critical contributors towards predicting reading comprehension.
The goal of reading is to extract meaning, and vocabulary knowledge appears to be an integral domain contributing towards passage level comprehension. A rich vocabulary allows for words to be meaningfully understood, which ultimately enables mastery of the passage (Levorato et al., 2009). Levorato, Roch and Beltrame (2009) postulated that individuals with Down syndrome between ages 8 to 16 years were able to use vocabulary to infer (i.e., use surrounding textual information) so as to understand the story passage. This is important as the meaning of a word is largely integrated within a given context. Hence, vocabulary knowledge may be vital in helping children with Down syndrome acquire the meaning of words by utilizing contextual knowledge, and aid in passage comprehension (Levorato, Roch & Nesi, 2007).
Phonological awareness (i.e., phoneme isolation and phoneme deletion) also significantly predicted passage level comprehension. Phonological awareness has been strongly linked with increasing orthographic knowledge through single word reading accuracy which enhances reading fluency (i.e., speed in which words are recognised) and automatic word recognition (Catts & Kamhi, 1999). Reading accuracy requires reading decoding skills (at least at the beginning stages of competency) which in turn utilises phonological awareness skills. Recent research has shown that children with Down syndrome use phonological skills to assist in word decoding and identification (Cupples & Iacono, 2000; Fidler et al., 2005). Word decoding skills may thus enable reading efficiency (Lyon, 1999) and indirectly aid reading comprehension in children with Down syndrome. This line of reasoning has yet to be confirmed because in Study one, the children performed close or at ceiling for the Early Word Recognition subtest of the YARC (i.e., its word decoding measure), rendering it unsuitable for statistical analyses. It is also possible that the ability to manipulate phonemes within words may reflect broader aspects of phonological processing and cognition which in turn contributes to the reading comprehension skills of children with Down syndrome.
Age also predicted reading comprehension skills in children with Down syndrome. Older children with Down syndrome were found to perform better on linguistic and cognitive tasks (e.g., phonological short-term working memory, receptive vocabulary and listening comprehension). Perhaps, with age, individuals with Down syndrome may have gleaned more literary experiences and opportunities to engage fully in learning, alongside others. With increasing age, individuals with Down syndrome may develop their own learning styles (Wishart, 1996). In addition, their life experiences may increase their understanding and exposure to real world knowledge. All of these may lend a hand in facilitating reading comprehension skills in children with Down syndrome. Nevertheless, more research is required to verify this proposition.
The findings of Study one identify sub-processes that are somewhat consistent with the Simple View of Reading (Gough & Tunmer, 1986). This framework identifies two interacting but separate components required for reading comprehension: single word decoding and listening comprehension. The results of Study one identified that although listening comprehension significantly correlated with reading comprehension, it did not contribute as a predictor of reading comprehension in the final regression model. Phonological awareness, on the other hand, did significantly predict reading comprehension, perhaps because of its involvement in word decoding. Receptive vocabulary, also found to be a significant predictor of reading comprehension, could be argued to be a sub-process strongly involved in listening comprehension ability. Indeed, Study one found a significant correlation between receptive vocabulary and listening comprehension. The results of Study one do, therefore, support the Simple View of Reading model to a certain extent but also call into question exactly what underlying skills might be involved in single word decoding and listening comprehension. In short, the results highlight that the terms used in the Simple View of Reading, are constructs which are difficult to operationalise precisely.
Study two: Predictors of spelling acquisition in Down syndrome
The second study presented in this thesis examined predictors of spelling for both spelling accuracy and absolute spellings. This is a unique contribution to the literature because much previous research has assessed spelling knowledge using correct spellings only. In a recent review of five standardised spelling tests, Calhoon, Greenberg and Hunter (2010) investigated if different spelling tests measured similar orthographic skills (e.g., syllables, consonants and vowels) and if spelling tests measured the same knowledge equally. Tests were compared according to the frequency of: (a) the number of syllables in each word, (b) the syllable types in each word, (c) consonant graphemes, and (d) the vowel types represented in each word. Results showed that all types of syllables and vowels were similarly represented across all tests, while long vowels and vowel teams (e.g., 'ue', 'ui' were least correlated). Interestingly, tests varied in their measurement of orthographic knowledge. For example, certain orthographic skills were tested more frequently, including one and two syllable words, closed and open syllables, /k/ = 'c' before a, o, u, or any consonant, /s/= 'c' before e, i, or y and short vowels (e.g., /e/). Calhoon and colleagues concluded that existing standardised spelling tests do not cover the range of orthographic skills that children are required to master.
In fact, other types of information can be used to measure spelling skills, e.g., analogy to known words, orthographic and morphological knowledge and memory for specific words (Ehri, 2000), but are not necessarily reflected in standardised spelling tests. To investigate spelling acquisition in children with Down syndrome, Study two examined both spelling accuracy and absolute spellings, and compared the performance of children with Down syndrome to typically developing peers matched for receptive vocabulary. Looking at absolute spellings (i.e., spelling approximations), in addition to spelling accuracy (i.e., correct spellings) allow for greater diagnostic and prescriptive strengths (Varnhagen et al., 1997). For example, a child with Down syndrome who spells police as 'polis' is likely to be mapping letters to sounds using a phonological approach. Incorporating absolute spellings (i.e., spelling approximations) in the analysis in Study two provide a greater insight into how children with Down syndrome may be using a combination of strategies while also reflecting different degrees of understanding about the phonological system used to spell words.
A comparison group was used in Study two because the pathway of learning to spell among typically developing children is not as well-researched as is the case for reading (Treiman, 1998). Spelling is also considered a more difficult task for English compared to reading as spelling depends on the permanent storage of information relating to component letters and their sequence (Seymour & Porpodas, 1980). There are also generally more possible spellings for a particular word than possible readings, for example, the phoneme /i/ has the following possible spellings: Y as in Entry, EY as in Key, EE as in Deep, EA as in Leaf, and IE as in Chief (Bosman & Van Orden, 1997). Thus, in view of the above complexities, a comparison group was utilised.
Study two showed an interesting pattern of results for both the children with Down syndrome and those with typical development. Correct spellings in children with Down syndrome were most consistently predicted by phonological short- term working memory and reading comprehension while absolute spellings (i.e., spelling approximations) were most significantly predicted by phonemic awareness, phonological short-term working memory and reading comprehension. For typically developing children, phonemic awareness and reading comprehension best predicted both spelling accuracy and absolute spellings. Phonemic awareness as a significant predictor of spelling for the typically developing group is a finding consistent with research (Bruck & Treiman, 1990; Holligan & Johnston, 1991). For instance, in a longitudinal study, Stuart and Masterson (1992) examined the spelling skills in 20 typically developing children (with a mean age of 10) who were first assessed as 4-year-old pre-readers. Results showed a significantly high correlation of 0.93 between phonological awareness and spelling (regular and irregular) words. The well-documented relationship between phonological awareness abilities and later spelling skills may arise due to the development of efficient phonological recoding (e.g., phoneme identification and phoneme manipulation) required for spelling.
Predictors of spelling for children with Down syndrome differed from the typically developing group. Overall, phonological short-term working memory was found to play a more important role than phonemic awareness for spelling accuracy in children with Down syndrome. In addition, although phonemic awareness emerged as a significant predictor of absolute spellings (i.e., spelling approximations), it was not as strong a predictor compared to phonological short-term working memory. This suggests that children with Down syndrome display a qualitatively different pattern in their spelling acquisition compared to typically developing children. This poses interesting questions as to whether different cognitive skills required for spelling are involved in children with Down syndrome relative to their typically developing peers. For instance, children with Down syndrome may be depending on a serial recall strategy used in attempting to spell, where orthographic patterns are matched and whole-word retrieval strategies are used rather than sounding out (i.e., decoding) strategies (Abbeduto, Warren & Conners, 2007). Digit span was chosen as the measure of phonological short-term working memory. It was postulated that children with Down syndrome would be able to perform the digit span task with greater accuracy than non-word repetition, by virtue of reducing the expressive phonological demands of the task. However, digit span performance could perhaps involve a serial recall strategy rather than reflect a pure phonological short-term working memory task. This highlights the complexity of measurement in capturing phonological short-term working memory in children with developmental disabilities. While this issue cannot be resolved in this thesis, an alternative to a serial recall strategy might be that children with Down syndrome have acquired a level of phonological awareness sufficient for spelling word approximations (e.g., spelling bus as 'bs') but inadequate for correct spellings. It may also be that children with Down syndrome are not utilising phonological awareness skills as reliably or as successfully as typically developing children.
Study two also found that reading comprehension significantly predicted spelling accuracy and absolute spellings (i.e., spelling approximations) in both groups of children. It is probable that increased exposure to print related material may facilitate spelling acquisition in children with Down syndrome and in typically developing children. Correct spelling requires precise knowledge of the phonological, orthographic and morphological information in words (Ehri, 2000; Treiman, 1998). Hence, increased exposure to reading print information may augment or boost orthographic representations stored in the internal lexicon (Mommers, 1987), which in turn assists in spelling development.
Relationships between reading and spelling development and the underlying skills that contribute
This thesis allows one to compare and reflect upon the reading and spelling performance of children with Down syndrome. After considering the results of Study one and two, perhaps the most striking observation relates to stage models of reading and spelling development. A popular conception of children's reading and spelling development is that the phonological, orthographic and morphological information and strategies that children acquire follow a sequence of stages (Ehri, 1992). Frith (1985) suggested that reading and spelling processes mutually influence each other and outlined four distinct stage models used predominantly for spelling, but also referred to in reading.
However, reading and spelling may not necessarily involve the same stages. For instance, if spelling development is characterised by similar reading stages, it should be possible to observe qualitatively different spellings at different points in development, with consistent spellings within a point of development. To this end, Varnhagen, Mccallum and Burstow (1997) investigated spelling samples from stories written by 272 typically developing children (from first to sixth grade) and found little evidence for predicted qualitative differences in stage classifications of errors or in stage consistency across grades. The authors found that children's spelling of silent -e long vowels and spelling for different types of -ed past tense words did not follow a strong developmental progression of qualitatively distinct stages from partial alphabetic to alphabetic to orthographic reading stages to correct spellings across the elementary school period. Furthermore, children may not use a letter name strategy (e.g., spelling are as 'r' versus spelling tea as 'te') to spell words all the time. Rather, they may be using a combination of strategies and be operating within different degrees of understanding about the phonological system, not explained by reading or spelling stages. For example, a young child might start off spelling 'r' for car. However, as the child improves in phonological segmentation and analysis, he or she starts applying letter name knowledge to the rime of the word which might lead a child to spell car as 'cr'. The partial alphabetic stage of reading or spelling fails to account for this type of spelling attempt (Henderson, 1990). Another way of thinking about this is to consider how a child correctly spells the word action. The child may know the spelling for act as well as the suffix -ion which therefore reflects integration of phonological, orthographical and morphological information rather than identifying and distinguishing between alphabetic and orthographic reading or spelling stages (Varnhagen et al., 1997).
Study two contributes to this issue. Spelling errors made by children with Down syndrome in the spelling study were noted to be variable and broad, such as spelling grab as 'gram', 'gurb', 'gab', gead', 'gimp' and 'goup', or club as 'clube', 'clam', 'clid', 'clap', clouds' and 'clock'. One child with Down syndrome, aged 10 spelt club as 'cub', signifying that she might be using rime units (cl-ub, c-ub) as a spelling strategy instead of a sequential progression from an alphabetic to an orthographic stage. Interestingly, some children with Down syndrome spelt the word police as 'pilling', showing that some morphological word-endings (i.e., 'ing') were utilised as a spelling strategy. Another child with Down syndrome, aged 12 spelt police as 'picoman', indicating that he might have linked the word police with a mental representation of a man and did not explicitly follow processes defined by Frith's (1985) reading stages. This shows that spelling skills may encompass multiple strategies compared to reading and may not necessarily involve nor reflect similar reading stage models.
The spelling attempts highlighted her also refelect contributions of underlying sub-processes identified in the results of Study two. Consider the spelling, of grab as 'gram'. Does this spelling attempt reflect the need for phonological short-term working memory? What about the word club and the spelling attempt 'clid'? Has the child used phonological awareness skills here? And finally, the word police, and the child's spelling as 'picoman'. Could reading comprehension processes have influenced that spelling attempt? While this thesis cannot definitively answer these questions, they serve to explore and marvel at the complex interplay between reading and spelling as well as the sub-processes that operate dynamically even within the one individual.
Notably, the linguistic and cognitive sub-processes influencing reading comprehension (Study one) differed from the sub-processes influencing spelling acquisition in children with Down syndrome (Study two). Primarily, sub-processes affecting reading comprehension included receptive vocabulary and phonological awareness skills, while phonological short-term working memory and reading comprehension were the most consistent predictors for spelling acquisition in children with Down syndrome. There are a few reasons why different sub-processes were identified in these two studies.
First, reading comprehension is concerned with recognising printed words and understanding the message that print conveys (Hoover & Gough, 1990). Hence, a wide vocabulary store of word meanings may allow an individual to construct an integrated and coherent mental representation of the text, rich in referential relationships and semantic knowledge (Ouellette, 2006). Spelling skills however, may rely primarily on using visual word recognition (i.e., lexical route) or phonological recoding (i.e., sublexical route) as described by the dual route model (Ellis, 1982), rather than tapping into a rich store of word meanings in order to spell.
Phonological short-term working memory was found to significantly predict spelling acquisition for correct spellings and approximate spellings, but not for reading comprehension. Spelling strategies may involve sounding out a word letter by letter, relying on phonological short-term working memory capacity to hold the early sounds in the word long enough to decode the whole word (Laws et al., 1995). Phonological short-term working memory deficits in children with Down syndrome may thus affect spelling attempts, in that children are less able to store as much phonological information (e.g., a spoken word) in short-term working memory at any given moment (Laws, 2004), thereby hindering their ability to spell accurately. In comparison, reading comprehension may not utilise phonological short-term working memory skills as much. A printed text provides a visual representation of language, giving a permanent reference rather than a transitional signal. Hence, this allows more processing time for the individual, and may not exclusively rely on phonological short-term working memory resources.
Implications arising from Study one and Study two and directions for future research
Implications from Study one and Study two
The ultimate aim of this thesis was to further our understanding of the nature of reading comprehension and spelling difficulties in children with Down syndrome living in a multilingual society. This thesis also seeks to offer research findings that have implications for language and literacy intervention research.
First, all the children with Down syndrome tested for reading comprehension presented with reading comprehension deficits despite varying levels of reading accuracy. Speech-language pathologists need to be aware that poor reading comprehension may go unnoticed in children with Down syndrome who appear, superficially, at least, to read well. Thus, assessments of reading should include measures of text reading comprehension, to identify those children who may have 'hidden' difficulties (Nation & Norbury, 2005). Second, speech-language pathologists should choose spelling tests best suited to the needs of the specific individual. Children could potentially be given credit for spelling approximations (e.g., spelling the first or last sound in a word correctly). Clinically, it may be more beneficial to use spelling approximations in assessing children with Down syndrome as it provides greater prescriptive analysis of the types of strategies used to aid their spelling. For example, does the child use a logographic (i.e., visual) strategy? Or does the child use a sound-symbol strategy (i.e., saying the sound aloud and writing the corresponding letter)? Intervention strategies could then be tailored for the individual based on their learning strengths.
Singapore is a multilingual and multicultural country, with the Ministry of Education's bilingual policy requiring that all children from Primary 1 (at the age of 7) learn English and one of the other three official languages (Mandarin, Malay or Tami). Indeed, most Singaporean children begin learning two languages in kindergarten, if not at home (Dixon, 2010). Spelling instruction for most kindergarten children in Singapore typically comprises of rote learning of letter name sequences (Rickard Liow, 1999), while a phonological based spelling instruction begins in Primary 1 (Ministry of Education, 2007). Understandably, English is a language that requires both processes, since numerous irregular words cannot be decoded according to consistent rules (Seymour, 2006). Thus, children's spelling instruction may develop via a combination of whole word recognition (i.e., logographic) or alphabetic (i.e., phonological) processing in a deep orthography such as English. The spelling strategies that children use also depend on early linguistic experience. The long-term store of phonological representations that support literacy skill development is language specific (Stackhouse & Wells, 1997). Hence, repeated aural exposure to a particular language (such as one's first language or second language) provides an important foundation for spelling. For instance, if cross-linguistic transfer between Mandarin-English bilinguals (who speak English as a second language) occurs, visual-logographic strategies developed for reading and writing Chinese characters are typically used for spelling English words (Rickard Liow, 1999). This impacts on children's strategies as children may be obliged to rely more heavily on logographic processing than phonological processing or spelling in English (Richard Liow & Poon, 1998).
A study by Yeong and Rickard Liow (2010) found that the linguistic experience of some second language (L2) learners of English may adversely affect subsequent spelling development relative to children whose first language (L1) was English. Spelling development might be non-optimal in bilinguals whenever the phonological systems of their languages are as different as English and Mandarin. For instance, poorly developed phonological awareness resulted in Mandarin - L1 children relying more heavily on visual-logographic foundation than a phonological-alphabetic foundation to build an orthographic foundation in English (Rickard Liow & Tng, 2003). In this light, spelling methods could be tailored accordingly for particular kinds of bilingual children with Down syndrome. For example, for Mandarin - L1 learners of English, focus could be placed on strengthening phonological awareness skills. The majority of participants recruited for this thesis spoke English 100% of the time. Only 8 children with Down syndrome spoke another language 2% to 40% of the time at home, but spoke English 100% of the time at school. It is therefore hoped that this research will encourage others to pursue bilingual literacy research with children with developmental disabilities. This issue aside, another equally important implication arising from this thesis is the importance of designing effective interventions for spelling and reading interventions that involve underlying cognitive and linguistic sub-processes.
Designing effective interventions
The purpose of reading is to enable the reader to construct meaning from text. For many individuals with Down syndrome, understanding the text may be difficult, and so reading activities simply become the practice of word calling (Morgan, Moni & Jobling, 2004). A review of the literature revealed limited empirical research about effective pedagogies and intervention strategies specifically relating to the spelling and reading comprehension of individuals with Down syndrome (Moni & Jobling, 2001;Â Fowler, Doherty & Boynton, 1995;Â Bochner, Outhred & Pieterse, 2001). Hence, an important direction for future work is to examine the utility of therapy approaches using well designed experimental treatment studies, to further elucidate the developmental mechanisms of reading comprehension and spelling acquisition (Nation & Norbury, 2005).
In spelling, children need to identify each phoneme in a word and assign a grapheme to each phoneme (Larkin & Snowling, 2008). This task can be particularly taxing for children with Down syndrome. Individuals with Down syndrome are relatively better visual learners, and may rely more heavily on whole-word or visual orthographic processes. Hence, spelling strategies such as chunking words at the syllable level (e.g., the word basket could be divided into 'bas' and 'ket') or morpheme level (e.g., the word dogs can be divided into two morphemes, 'dog' and '-s') may relieve the phonological short-term working memory load and serve as a spelling intervention strategy. However, further research using chunking as a spelling intervention is required.
It is also important to include memory training for children with Down syndrome. Strengthening phonological short-term working memory to improve spelling could be taught by teaching a rehearsal strategy (e.g., Broadley & MacDonald, 1993; Laws, MacDonald & Buckley, 1996). Phonological short-term working memory retains information for only a limited amount of time and decays rapidly. In order to overcome this and retain information longer, information must be periodically repeated, or rehearsed, either by articulating it out loud, or by mentally simulating such articulation. Information may then re-enter the phonological store and thus be extended. This could be accomplished through for example, exposure to reading. Early reading books are often repetitive and reading may provide the opportunity for frequent practice of words. For example, reading the sentences, "This is the house that Jack built. This is the malt that lay in the house that Jack built. This is the rat that ate the malt, that lay in the house that Jack built." (Opie & Opie, 1951) may aid children's acquisition of long-term memory representations (Laws, 2004). In this light, spelling may be enhanced through rehearsal and repetition strategies.
Reading strategies for children with Down syndrome could capitalise on their strengths in whole word recognition to foster the development of phonological decoding skills. Reading intervention could start with an early emphasis on whole-word reading and progress to include phonological awareness training (Buckley, 2003; Goetz et al., 2008). This comprehensive approach has also been supported for children with reading delays (Hatcher, Hulme & Snowling, 2004). Activities could be designed to emphasise phonological patterns in word sets with which the child has had previous success. For instance, if the child has had whole-word decoding success with several words beginning with the phoneme /b/, targeting these words as a group and emphasizing their phonological similarity may raise the child's phonological awareness, enhancing reading.
Phonological awareness intervention has been shown to effectively support reading development in children with Down syndrome. Kennedy and Flynn (2003) as well as Cupples and Iacono (2000) found that children with Down syndrome were able to make improvements in phonological awareness skills and utilise phonological awareness skills to aid in word decoding. Other researchers noted that phonological awareness intervention may enhance word recognition, reading comprehension and spelling performance in children with spoken language impairments. Specifically, this includes phoneme level activities (e.g., blending, phoneme segmentation and manipulation) as well as activities that strengthen children's phoneme-grapheme connections (Gillon, 2004).
Cologon, Cupples and Wyver (2007) compared two training intervention programs: a phonological awareness program and a reading comprehension program in 15 children with Down syndrome (aged 2;11 to 10;8) for 10 weeks. The phonological awareness intervention emphasised oral reading using decoding skills and blending tasks. The reading comprehension intervention comprised of selecting pictures to match action words and sentences. Children were assessed pre-intervention, immediately following completion of the intervention and six months post intervention. Results showed that children in both intervention groups demonstrated significant gains on measures of phonological awareness, letter-sound knowledge and passage comprehension. This suggests that phonological awareness skills may be an important factor contributing towards reading comprehension.
Gallaher, van Kraayenorrd, Jobling and Moni (2002) also examined the effectiveness of a series of tutoring sessions over six months on the reading and spelling development of an individual (Abby, aged 19) with Down syndrome. Intervention during individual tutoring sessions focused mainly on phonological awareness skills (i.e., rhyming, initial phoneme identification and phoneme segmentation), sight vocabulary, and word decoding, as part of a LATCH-On (Literacy and technology-Hands On) program. Spelling intervention included identifying the initial phoneme in words. This greatly facilitated Abby's ability to supply the correct initial consonant for a word. By the end of the tutoring period, Abby spelled the majority of initial consonants correctly in words. In addition, phonological training in rhyming patterns using a set of predictable books was conducted, allowing Abby to anticipate patterns in sentences. This familiarity with words resulted in greater understanding of the text. It was observed that Abby often made semantic errors (e.g., substituting a word that might have made sense in a sentence but was not the actual word) in her reading, providing evidence that she showed understanding of what was read..
Along similar lines, Van Bysterveldt (2009) monitored the reading and spelling skills in Ben, a boy with Down syndrome (aged 5;2) over 3 years who participated in an 18 weeks integrated phonological awareness intervention program. The program involved a parent implemented home program to facilitate letter-sound knowledge via print referencing techniques during story book reading, as well as phoneme identification and matching tasks. Post intervention results at age 8 showed that Ben was able to transfer real word decoding skills to connected text, with sufficient accuracy to support comprehension of the text (i.e., answering three out of four comprehension questions correctly at Level 1) when assessed on the Neale Analysis of reading Ability (NARA; Neale, 1999). Ben's spelling also showed increases in the number of phoneme-grapheme matches and a growing ability to spell sounds in non-words. Thus, training phonological awareness skills may serve to boost word spelling approximations and support reading comprehension skills.
Ouellette (2006) suggested that reading comprehension skills may also benefit from vocabulary development teaching, focusing on in-depth word knowledge and semantic organization: teaching of word definitions, words with multiple meanings, varied contextual use of vocabulary items, and word relations to expand and organize the semantic system. This also seems ripe for an intervention study examining the effects of improving semantic skills and reading comprehension deficits in children with Down syndrome. Pany, Jenkins and Schreck (1982) assessed the effect of vocabulary instruction on reading comprehension in a group of students with learning disabilities (aged 10 to 12). Vocabulary training involved synonym instruction where an experimenter showed a printed target word and said a one word synonym (e.g., debris means trash), then stated a sample sentence containing the word (e.g., after the picnic, we put the debris in the garbage can). Each student individually stated the word and its synonym, whereupon the group repeated them in unison. Overall, results showed that vocabulary training involving synonym acquisition transferred to comprehension of single sentences.
In summary, the persistent nature of spelling and reading comprehension difficulties in individuals with Down syndrome suggests the need for sustained interventions (Van Bysterveldt, 2011). Future research should track the outcomes of reading comprehension and spelling interventions longitudinally to evaluate intervention strategies within a preventative framework. In addition, it would be beneficial to evaluate home and classroom based activities to promote literacy development in the Down syndrome population.
The body of research reported in this thesis contributes to what is known about the reading comprehension and spelling abilities of children with Down syndrome. First, this research provides evidence that particular cognitive and linguistic skills (namely phonemic awareness, receptive vocabulary and age) significantly predict reading comprehension in Down syndrome. Notably, the present research suggests that the spelling abilities of Singaporean children with Down syndrome may rely on different processes compared to typically developing children. Phonological short-term working memory, in particular, appeared to play a more important role than phonological awareness for spelling in children with Down syndrome compared with their vocabulary matched typically developing peers. In addition, this research offers findings that can be used to guide future assessment and intervention approaches of reading comprehension and spelling development in the Down syndrome population.
There is still much to learn about reading and spelling abilities in Down syndrome within a multilingual environment. It is hoped that this thesis encourages further research in understanding the nature of reading and spelling difficulties in children with Down syndrome with the ultimate goal of improving the literacy skills and enriching the lives of children with Down syndrome.