Comparative effectiveness of phonological awareness and oral language intervention for children with low emergent literacy skills.

Fielding-Barnsley, Ruth ; Hay, Ian

Introduction

Early identification of reading difficulties is paramount if teachers are to help change the current situation where 16% of children are falling into the category of 'disabled readers' in Australia. (Louden et al., 2000). In the United States, The National Assessment for Educational Progress (NAEP, 1998) identified 60% of students scoring below proficient in fourth grade. Early identification and the resulting remediation of reading difficulties have many advantages over leaving children to 'grow out' of their problems (Pressley, 2006; Torgesen, 1998). One such advantage is reducing the children's negative feelings associated with their constant failure in the classroom (Hay, 2000; Muter, 2003). Muter maintained that early identification also allowed for a 'purer' child profile to be established, because older children often acquired compensatory and avoidance behaviours that can mask their reading difficulties.

It has been demonstrated that there is a strong link between phonological awareness and early reading. Children's awareness of the phonological units of speech, particularly rhyme (Bradley & Bryant, 1983) and alliteration (Byrne & Fielding-Barnsley, 1995) have been found to have significant effects for early reading. Knowledge of the alphabet is also one of the best predictors for a successful transition into early reading (Adams, 1990).

Whilst there has been a plethora of research on the benefits of phonological awareness intervention for struggling readers (see Lovett, Steinbach & Frijters (2000) for a summary), there has been significantly less on the effects of language intervention (Muter, Hulme, Snowling, & Stevenson, 2004). The NICHD (2005) suggests that there is a vital link between early oral language and decoding skills. While phonological awareness, the understanding of the sound structure of language, is necessary, it alone is not sufficient for the successful acquisition of reading (Byrne, Fielding-Barnsley & Ashley, 2000). Hay, Elias, Fielding-Barnsley, Homel, and Frieberg, (2007) demonstrated that the inclusion of a structured language programme along with a shorter phonological awareness programme produced significant gains in children's reading ability; particularly if those children came from low (socio-economic status) SES homes. Children who fail to respond to reading intervention are often those with weak oral language (Al Otaiba & Fuchs, 2006). Hart and Risley (1995) made the link between weak oral language and low SES in their seminal study where it was found that children from high socio-economic status (SES) families heard around 487 utterances per hour, compared to 178 utterances per hour for children from families on welfare. By the time they were aged 4 years, the high SES children had been exposed to around 44 million utterances, compared to 12 million utterances for the lower SES children.

In particular, the evidence suggests that many children with significant reading and learning difficulties/disabilities have deficits in both phonological awareness and language skills (Saada-Robert, 2004; Snowling, 2005). Whilst language delays are considered a cause of reading delays, the children's lack of reading skills also have an ongoing negative influence on the children's vocabulary and language development (Catts & Kamhi, 2005). This reciprocal relationship between language and reading has significant implications for the type and range of screening and interventions teachers provide to children in the beginning school years.

Several researchers have reported on the beneficial effects of language instruction on measures of comprehension (Oakhill, Cain, & Bryant, 2003; Muter et al., 2004) and more recently Bowyer-Crane et al. (2008) have reported differential effects between an oral language and phonology with reading based intervention. In the Boyer-Crane et al., study the Phonology+Reading group showed an advantage over the Oral Language group on measures of literacy and phonology and the Oral Language group showed an advantage on measures of vocabulary and grammatical skills.

Whilst Hay et al. (2007) have demonstrated that a language plus phonological awareness intervention has enhanced decoding (reading) scores we do not know if the intervention has the same effect on measures of encoding (spelling). A recent study concluded that whilst a language and phonological awareness intervention delivered in pre-school demonstrated positive short term effects it did not generalise to measures of encoding (spelling) in Grade 2 (O' Connor, Arnott, McIntosh & Dodd, 2009).

In the following study the terms 'decoding' for early reading and 'encoding' for spelling are used interchangeably. The term 'decoding' is understood to describe the process of changing written letters into spoken sounds to make meaning of written text; or 'breaking the code'. 'Encoding' is the exact mirror image of this process and is understood to describe the process of changing spoken language into the symbols of written language. Both processes are used to discover the alphabetic principle and therefore valid measures of early reading.

Research questions

The research questions are driven by the theory, particularly by the concept that phonological awareness may be underpinned by oral language skills (Burgess & Lonigan, 1998) and phonological processing skills, such as encoding (spelling) may be largely dependent on the development of good oral language skills (Storch & Whitehurst, 2002). This research therefore asks the question, will an oral language intervention result in similar gains to phonological intervention on a measure of spelling?

The second research question aims to build on the research by Hart and Risley (1995) who link low oral language with SES and Hay et al. (2007) who demonstrated that a phonological awareness with oral language intervention resulted in increased literacy growth for children from low SES backgrounds. The research aims to answer specifically the influence of parents' educational and family literacy levels on outcome measures of early literacy, namely spelling.

Method

Participants

Participants were 457 students from Year One in 9 schools varying in Socio Economic Status (SES) in the Brisbane region of Queensland, Australia. Two schools were classified as high SES, five schools as average (middle) SES and two schools as low SES as determined by Australian Government income census data. Two hundred and thirteen students were female and 244 were male. Their mean chronological age in months at the time of screening was 70.6 (SD = 4.01). Children with a known intellectual disability were excluded from the study.

Approval for this study was received from the relevant university ethics committee, school authorities, as well as the principals, children's teachers, parents and caregivers.

Design and procedure

All 457 children received a battery of tests in the second term of their first year of formal schooling (Year 1 in Queensland). As a result they had received some literacy instruction. The children were settled and familiar with classroom routines. Children were tested individually in their schools, in a quiet environment, by trained research assistants who were also registered teachers. During the 1st round of screening all children received a letter knowledge and rhyme detection test in one session lasting approximately 30 minutes. Measures at the level of the phoneme were not administered due to floor effects in a pilot test. The tests were administered in the same order for each child.

During the 2nd round of screening, from the population of N = 457,a second cohort of 141 children was selected based on their alphabet and rhyme scores and the need to representatively sample across all of the schools involved in the project. The 141 of children (31%) who scored at or below the cut off scores of < (less than) or= to 15/26 for alphabet recognition and <or = to 5/10 for rhyme detection were then tested with an expressive and receptive vocabulary test in one session lasting approximately 45 minutes. The cut off scores were selected to identify children potentially at risk of developing difficulties in early literacy acquisition. The rhyme detection test was constructed in a way that allowed for children to score 3 correct answers by chance therefore a score of 5/10 was actually very low. The tests were administered in the same order for each child, with speed of word naming collected while completing the expressive vocabulary measure. The parent/s of each child also completed a family reading survey relating to each child in the study.

The sample of children selected for intervention was further reduced by taking their expressive and receptive language scores into consideration. Only those children who scored <or = to 70/100 on the HPNT or <or = to 70 (Standardised score) on the Peabody Picture Vocabulary Test (PPVT) were included in the intervention programme. Therefore only children who were identified as being at risk in both phonological awareness and language were included in the intervention sample. This resulted in a total of 96 children, 21% of the original sample of 457 children. The rationale for the cut off points for all pre-test measures was to have a cohort who would be an approximate representation of the Louden et al. (2000) study where 16% of the sample fell into the category of 'disabled readers'.

A group of 29 at-risk children acted as a classroom control and received no intervention. A group of 31 not at-risk children acted as class controls, 2 children were randomly selected from each class. This resulted in a total of 156 participants in the study who received all pre and post test measures. In addition 301 children from the original sample were included in the measure of spelling.

A measure of encoding (spelling) was selected as a proxy for early reading based on the findings of Morris and Perney (1984) who illustrated the strong relationship between early encoding (spelling) and reading. Also we gained much richer data from a partial scoring system adapted from a version by Liberman, Rubin, Duques and Carlisle (1985). Whereas a measure of reading is usually scored as being either correct or incorrect, this measure of encoding (spelling) awarded scores for partially correct encoding. (Please refer to the post testing for a full description of this test.)

Instruments

Tests Administered to all Children (N= 457)

Screening 1st round--The Phonological Abilities Test (PAT) (Muter, Hulme & Snowling, 1997).

The Phonological Abilities Test (PAT) consists of a number of phonological awareness subtests of which rhyme detection and a test of letter knowledge were selected for this study due to floor effects at the level of the phoneme.

Letter Knowledge: In this test the children were asked to supply either the name or the sound of each of the 26 lowercase letters of the alphabet, which were presented in random order on individual cards. Scoring was based on the total number of letters identified, whether by name or by sound.

Rhyme detection: The 10-item rhyme detection subtest from the PAT was used to measure children's ability to detect rhyme in spoken words. There were two practice items and corrective feedback was given for the first four items. For the ten items with pictures the children were required to indicate which of three words rhymed with a named stimulus word (e.g., fish, gun, hat) with (e.g., cat).

Tests administered to at risk children

Vocabulary--Receptive

Peabody Picture Vocabulary Test- Revised (PPVT-R; Dunn & Dunn, 1981). The PPVT- R is a standardised measure of receptive vocabulary skills. Participants are presented with an easel page that contains four numbered pictures. The goal of this task is to select the picture that matches the word that was orally presented by the administrator. The test manual reports internal consistency coefficients that range from 0.67 to 0.88 (median = 0.80) for Form L and from 0.62 to 0.86 (M = 0.81) for Form M. Immediate retest alternate from reliability coefficients range from 0.73 to 0.91 (M = 0.82) and delayed retest alternate form reliability coefficient range from 0.52 to 0.90 (M = 0.78).

Vocabulary--Expressive

Hundred Picture Naming Test (HPNT) (Fisher & Glennister, 1992). The HPNT was used to measure expressive vocabulary. This test consists of 100 line drawings of everyday objects such as: butterfly, scissors and kite. The participants are asked to name each of the items and this test is also timed. Test occasion correlation = 0.83.

Stony Brook Family Reading Survey (Whitehurst, 1993). This adapted questionnaire (Fielding-Barnsley & Purdie, 2002) consists of 6 questions relating to each family's reading related activities in the home and 2 questions related to each parent's level of education. Examples of reading related questions are: How many picture books do you have in your home and how often does your child ask to be read to? Each question has a range of five possible answers.

Intervention

The intervention sessions consisted of 16 x 30 minute lessons over a period of 8 weeks. The lessons were conducted by trained research assistants in small groups of 4 to 5 children outside of their regular classrooms.

Phonological awareness was taught using Hatcher's (2000) 'Sound Linkage', an integrated programme for overcoming reading difficulties and Byrne and Fielding-Barnsley's (1991) 'Sound Foundations' Programme.

Forty-nine children at 5 of the 9 schools received this intervention. The intervention was comprised of systematic and explicit instruction in phonological awareness beginning with understanding the terms beginning, middle and end, followed by learning about syllables, rhyme and phonemes. These activities enable the instructors to work directly on improving children's ability to analyse the sounds of speech, and to relate this knowledge to the process of learning to read and spell. 'Sound Foundations' was used to complement instruction in phonemic awareness. This kit consists of large posters depicting words beginning or ending with the same sound, for example the beginning /s/ poster depicts the sun, sea, seal, seahorse etc., and the ending /s/ poster depicts a bus, glass, octopus etc. The snap and domino card games included in Sound Foundations were also used to reinforce this learning. The alphabet was not included in this instruction as this was being taught in the classroom.

Forty seven children at 5 of the 9 schools received Marion Blanks' four levels of dialogue instruction. Blank's four levels of dialogue are: (i) dealing with directly supplied information (matching the experiences); (ii) classification (selective analysis of the experience); (iii) reorganisation (reordering the experience); and (iv) abstraction and inference (reasoning about the experience. Marion Blank and her colleagues (Blank & Franklin, 1980; Blank, Rose, & Berlin, 2003) have proposed four levels of dialogue complexity, where the children are active participants in the learning interchanges. In such a communicative context the teacher initiates and shapes the dialogue so that the students respond at a more appropriate and advancing level of linguistic complexity.

Referring to Table 1 which identifies the four levels and the sub-elements, at the lowest level of complexity or first level, the student is required to respond to language concerning salient perceptions (e.g., to the question, what is this?), with a focus on vocabulary development, moving to the second level of the organisation of information stage where the key questions investigate how objects, events, or issues are classified. This organisation and classifying of the information particularly helps facilitate students' encoding and retrieval of information into and from their long-term memory. This skill is particularly relevant to the current study where the children are demonstrating their ability to retrieve information in order to encode (spell) words. The third level is focussed on reorganising or adding to the information, based on what an individual already knows of the topic, which is the linking of the information to higher order reasoning (e.g., what else do you know about?). This level of dialogue relates to the current study where the children use their knowledge of orthography to recognise similarities in word families. The fourth level deals with the abstract, and at this level the language demands involve reflecting upon or restructuring perceptions (e.g., the question, what do you think will happen if ...?). Again this level of dialogue assists the children in the current study to utilise their knowledge of phonemic awareness in for example deleting the /s/ sound in nest to produce the word net.

Post intervention testing

A total of 156 children received the full Phonological Abilities Test (PAT) and the Hundred Picture Naming Test (HPNT). In addition, all children plus the remaining 301 children in all classes received the Fielding-Barnsley/ Hay spelling test. The 156 children comprised 49 from the phonological awareness intervention, 47 from the oral language intervention, 29 classroom at-risk controls and 31 not at-risk classroom controls.

The PAT consists of six subtests: four tests of phonological awareness (two rhyming and two segmentation tasks), a test of speech rate and a test of letter knowledge. The six tests are:

1. Rhyme detection which requires the child to select the word which rhymes with the stimulus word for an array of three choices.

2. Rhyme production in which the child is given 30 seconds in which to produce as many rhyming responses as possible to a stimulus word.

3. Word completion (syllables and phonemes) in which the child 'finishes off' the final syllable or phoneme in a series of one or two syllable words.

4. Phoneme deletion (beginning and end sounds) in which the child removes or deletes either the first or final phoneme of a single syllable word.

5. Speech rate which requires the child to repeat the word 'buttercup' 10 times as quickly as possible while the examiner records the time taken.

6. Letter knowledge in which the child is asked to supply either the name or sound for each of the twenty- six letters of the alphabet.

The Fielding-Barnsley/ Hay (2006) spelling test and scoring system are amended versions of the Liberman, Rubin, Duques, and Carlisle (1985) spelling test. The children were asked to spell 10 words in writing (dog, man, one, said, blue, come, plug, went, limp, tree) and 4 pseudowords (ig, sut, frot, yilt) presented in the same order. The words were said once, repeated in a sentence, and spoken a third time. The pseudowords were spoken three times. A correct spelling, such as 'come' earns the maximum of 6 points. Letter reversals were ignored. The rest of the scoring system is based on the number of phonemes represented; the scoring system values accurate segmentation. A spelling that represents all phonemes with conventionally acceptable letters, such as 'cum' or 'kum' for come earns 5 points, and one that represents all phonemes but includes more distantly related letters such as 'cam', is given 4 points. If the attempt includes more than one but not all phonemes, such as 'cm' or 'ku' for 'come' or 'pug' for 'plug', earns 3 points. One phoneme with a conventional letter, such as k, is given 2 points, and one phoneme with a related letter, such as g or n for come, earns 1 point.

Results

The full cohort of children was first investigated for gender differences in rhyme and alphabet knowledge, with outcomes not significantly influenced by gender (rhyme F (1,442) = 1.58, sig = 0.209; alphabet F (1,450) = 0.162, sig = .687).

Correlations of language and early indicators of reading

Investigating the relationship between the early literacy variables, the highest correlation was r = 0.68 between expressive and receptive language (HPNT and PPVT), see Table 2. There are also high inter-correlations between the children's alphabet knowledge and their expressive and receptive language scores. Unlike alphabet knowledge, rhyme did not correlate with expressive and receptive language, however, there was a significant moderate correlation between children's alphabet and rhyme knowledge, r = 0.32. The children's speed of word naming on the HPNT was negatively correlated with both alphabet knowledge and the language measures, demonstrating that children who achieved higher scores on alphabet and language measures took less time to respond to the HPNT pictures, that is they had a higher level of automaticity associated with the naming of objects. For example, based on the obtained plotted regression square line of 0.05 an alphabet score of 19 equals 200 seconds, and an alphabet score of 5 equals 800 seconds. This finding demonstrates that if a word is in the child's lexicon or long term memory it is retrieved more quickly.

Home reading environment

Children's abilities to do the rhyme tasks were uncorrelated with their home reading behaviours. In contrast, children's alphabet knowledge was highly associated with home book reading as was children's receptive and expressive language development. The highest correlations were associated with the children's alphabet knowledge, the amount of home reading, and the years of education of the children's primary caregivers, see Table 3.

Intervention outcome using the Fielding-Barnsley/ Hay (2006) spelling results

In the first of the following set of analyses of variance (ANOVAs), a measure of children's spelling results was entered as the outcome variable, and type of intervention (phonological, or language) plus two relevant control groups (class control not at risk or at-risk control) was entered as the predictor variable.

The ANOVA showed a significant main effect for group (F(3, 115)=4.757, p<0.01). Group differences in relation to the spelling outcome are shown in Table 4.

As indicated in Table 4, the phonological intervention group did better than the language experimental group and significantly better than the at-risk control group (p<0.01).

A series of follow-up ANCOVAs (analysis of covariance & variance) was utilised to examine the effect of various pre-existing conditions that might be deemed to be influential. This analytic approach is equivalent to combining analysis of variance and regression approaches within a single procedure. The question of interest is whether type of intervention is still statistically influential under these conditions.

The first of these ANCOVAs included pre-test measures of rhyming knowledge and alphabetical knowledge as covariates. Under these conditions, alphabet knowledge (F(1,113)=32.269) and group (F(3,113)=2.845, p<0.5) both predicted FBH spelling outcomes at statistically significant levels, while the effect of rhyming knowledge was statistically non-significant.

With pre-existing rhyming and alphabet knowledge taken into account, the phonological intervention surpassed other treatment groups and did significantly better than the at-risk control group (p<0.05).

Family background

The effect of family background was examined by including two Family Reading Survey (FRS) items that measured parents' years of schooling (Secondary, Post-secondary) completed by the two spouses, and school SES as covariates in an ANCOVA. The outcomes suggested that neither years of school nor SES predict spelling outcomes. Under these conditions, the effect group was marginally significant (p=0.062).

The effect of home reading was examined by an ANCOVA that included a covariate that combined four FRS items related to the frequency with family members read picture books to the child, how often the child asked to be read to, how often that child looked at books by him/herself, and how often the adult accompanied the child to the library. As with family background, this measure of pre-existing literacy predicted spelling outcomes at non-significant levels, in contrast with type of intervention (F(2,49)=3.939, p<0.05), where either phonological or experimental language participants did better than those in the at-risk control group (p<0.05).

Discussion

The overall finding is that the most significant differences in measures of spelling are related to the type of intervention rather than factors related with family literacy practices or parents' education levels. This concurs with the research conducted by Noble, Farah and McCandliss (2006) who stated that 'regardless of background children with higher phonological awareness abilities are decoding successfully' (p. 140) and Van Steensel (2006) and White (1982) who both argued that it was the social, language and literacy enhancement associated with higher SES homes that were the important factors; not the parental occupations or family income. This finding has important ramifications for early childhood educators who may need to adjust their teaching programmes to cater for these differences.

The low literacy levels at school entry reported in the current study are certainly cause for concern and should alert teachers to the fact that these children will require systematic and explicit teaching of phonological awareness and language in order to overcome any potential risks associated with low levels of pre-literacy skills at school entry. The message is that these children are capable of closing the gap between being a poor reader and a good reader, or in this case between being poor spellers and good spellers. It must be emphasised that it is the intervention that has the most significant effect rather than the family background or family literacy practices.

It is critical that teachers are aware of the important factors, such as knowledge of the alphabet, that do have a significant impact on measures of literacy (Adams, 1990). The current study revealed that alphabet knowledge was significantly correlated with the measures of spelling. However, knowledge of rhyme was not significantly correlated with measures of spelling. Other studies have also reported non-significant results for knowledge of rhyme and outcome measures of reading (Muter et al., 2004).

The current study supports other studies (Byrne & Fielding-Barnsley, 1995) in the finding that phonological awareness intervention supports early measures of reading and spelling but what is not so commonly known is the effect of early language intervention on measures of early literacy such as spelling. The current study supports the findings of Burgess and Lonigan (1998) and Storch and Whitehurst (2002) that phonological awareness should be combined with early language intervention. We know that phonological awareness and language support each other (Hay et al., 2007; Nation & Snowling, 2004) and that both are necessary in order to support early readers; particularly those at risk due to entering school with low levels of language. It has also been noted that oral language supports 'learning to learn' (NICHD, 2005) and this could explain why the oral language intervention in this study was almost as effective as the phonological awareness intervention. It could be hypothesised that some children cannot access the language related to instruction in phonological awareness such as, 'what is the beginning sound in this word?' The potential interference of semantic complexity of instruction is clarified in Marion Blank's oral language intervention. Speece and Cooper (2002) have provided evidence that language training predicts growth of phonological skills in kindergarten children. They concluded that the ability to understand the meaning component of language is a significant indicator of single-word reading in first grade. The emphasis on vocabulary growth in the oral language component of the current study supports the hypothesis that oral language supports the development of early literacy skills such as spelling. A similar study (Hill & Launder, 2010) failed to demonstrate a relationship between oral vocabulary and reading achievement but it could be suggested that they may have found a correlation had they used a measure of spelling as demonstrated in the current study.

Whilst there remain several questions as to the importance of oral language and its relationship to beginning reading we know that oral language is a better predictor of comprehension in later years and that phonological awareness predicts decoding (Bishop & Snowling, 2004). It can therefore be deduced that both skills are important to the ultimate goal of reading for understanding and should be part of the early teaching repertoire.

Limitations

One of the limitations of this study was that the 2 interventions were not combined, that is that each intervention group received either the phonological awareness intervention or the language intervention so we are unable to confirm that phonological awareness and language may have a multiplicative effect on outcome measures of spelling.

Another limitation is that the in-class control group did not receive any intervention and therefore this could have resulted in a Hawthorne effect for both intervention groups.

Summary

This study has highlighted the importance of explicit and systematic instruction in phonological awareness and oral language for children at the beginning stages of reading. Whilst this research highlighted a group of children identified with low language and phonological skills at school entry it does not preclude teaching these skills to all children. Measures of spelling, especially those that credit approximation, may be a better indicator of children's acquisition of the alphabetic principle than reading.

Acknowledgement

This article is based on research funded by the Australian Research Council Discovery grant DP0666577 to Professor Ian Hay, Associate Professor Ruth Fielding Barnsley and Professor Adrian Ashman.

References

Adams. M.J. (1990). Beginning to read: Thinking and learning about print. Cambridge, MA: Massachusetts Institute of Technology.

Al Otaiba, S., & Fuchs, D. (2006). Who are the young children for whom best practices in reading are ineffective: An experimental and longitudinal study. Journal of Learning Disabilities, 39(5), 414-431.

Bishop, D., & Snowling, M. (2004). Developmental dyslexia and specific language impairment: Same or different? Psychological Bulletin, 130 (6), 858-886.

Blank, M., & Franklin, E. (1980). Dialogue with preschoolers: A cognitive-based system of assessment. Applied Psycholinguistics, 1, (2) 122-150.

Blank, M., Rose, S.A., & Berlin, L.J. (2003). Preschool Language Assessment Instrument: The language of learning in practice (2nd Ed.). Austin, TX: Pro Ed.

Bowyer-Crane, C., Snowling, M.J., Duff, F.J., Fieldsend, E., Carroll, J.M., Miles, J., Gotz, K., & Hulme, C. (2008). Improving early language and literacy skills: Differential effects on an oral language versus phonology with reading intervention. Journal of Child Psychology and Psychiatry, 49, 422-432.

Bradley, L., & Bryant, P.E. (1983). Categorizing sounds and learning to read--a causal connection. Nature, 301, 419-421.

Burgess, S.R., & Lonigan, C.J. (1998). Bidirectional relation of phonological sensitivity and prereading abilities: Evidence from a preschool sample. Journal of Experimental Child Psychology, 70(2), 117-141

Byrne, B., & Fielding-Barnsley, R. (1995). Evaluation of a program to teach phonemic awareness to young children: A 2- and 3-year follow-up and a new preschool trial. Journal of Educational Psychology, 87(3), 488-503.

Byrne, B., & Fielding-Barnsley, R. (1991). Sound Foundations. Sydney, Australia: Peter Leyden Educational Publishers.

Byrne, B., Fielding-Barnsley. R., & Ashley, L. (2000). Effects of preschool phoneme identity training after six years. Outcome levels distinguished from rate of response. Journal of Educational Psychology, 92(4), 659-677.

Catts, H.W., & Kamhi, A.G. (2005). The connection between language and reading disabilities. Mahwah, NJ: Erlbaum.

Dunn, L.M., & Dunn, L. (1997). Peabody Picture Vocabulary Test, Version 3. Circles Pines, MI: American Guidance Service

Fielding-Barnsley, R., & Purdie, N. (2002). Developing pre-literacy skills via shared book reading: The assessment of a family intervention program for preschool children at risk of becoming reading disabled. Australian Journal of Learning Disabilities, 7(3), 13-19.

Fisher, J.P., & Glennister, J.M. (1992). The Hundred Pictures Naming Test. Hawthorn: The Australian Council for Education Research.

Hart, B.H., & Risley, T.R. (1995). Meaningful differences in the everyday experience of young American children. Baltimore, MD: Paul H. Brookes.

Hatcher, P.J. (2000). Sound Linkage: An integrated programme for overcoming reading difficulties (2nd ed.). London: Whurr.

Hay, I. (2000). Gender self-concept profiles of students suspended from high school. Journal of Child Psychology and Psychiatry and Allied Disciplines, 41, (3) 345-352.

Hay, I., & Fielding-Barnsley, R. (2006). Enhancing the early literacy development of children at risk for reading difficulties. Australian Journal of Learning Disabilities, 11(3), 117-124.

Hay, I., Elias, G., Fielding-Barnsley, R., Homel, R., & Frieberg, K. (2007). Language delays, reading delays and learning difficulties: Interactive elements requiring multidimensional programming, Journal of Learning Disabilities, 40 (5), 400-409.

Hill, S., & Launder, N. (2010). Oral language and beginning to read. Australian Journal of Language and Literacy, 33 (3), 240-254.

Liberman, I.Y., Rubin, H., Duques, S., & and Carlisle, J. (1985). Linguistic abilities and spelling proficiency in kindergartners and adult poor spellers. In D.B. Gray & J.F. Kavanagh (Eds.), Biobehavioral measures of dyslexia (pp. 163-176). Parkton, MD: York Press.

Louden, W., Chan, L., Elkins, J., Greaves, D., House, H., Milton, M., Nichols, S., Rohl, M., Rivalland, J., & Van Kraayenoord, C. (2000). Mapping the territory. Canberra, Australia: Commonwealth Department of Education, Training and Youth Affairs.

Lovett, M.W., Steinbach, K.A., & Frijters, J.C. (2000). Remediating the core deficits of developmental reading disability: A double deficit perspective. Journal of Learning Disabilities, 33(4), 334-342.

Morris, D., & Perney, J. (1984). Developmental spelling as a predictor of first-grade reading achievement. The Elementary School Journal, 84 (4), 440-457.

Muter, V. (2003). Early reading development and dyslexia. London: Whurr Pub

Muter,V., Hulme, C., & Snowling, M. (1997). The Phonological Abilities Test. (PAT). London. The Psychological Corporation. Harcourt Brace & Company.

Muter, V., Hulme, C., Snowling, M., & Stevenson, J. (2004). Phonemes, rimes, vocabulary, and grammatical skills as foundations of early reading development: Evidence from a longitudinal study. Developmental Psychology, 40 (5), 665-681.

National Assessment for Educational Progress (NAEP) (1998). Reading report card for the nation and the states. National Center for Education Statistics, Washington, DC.

Nation, K., & Snowling, M.J. (2004). Beyond phonological skills: broader language skills contribute to the development of reading. Journal of Research in Reading, 27 (4) 342-356.

NICHD. Early Child Care Research Network. (2005). Pathways to reading: The role of oral language in the transition to reading. Developmental Psychology, 41(2), 428-442.

Noble, K.G., Farah, M.J., & Mc Candliss, B.D. (2006). Socio-economic background modulates cognition-achievement relationships in reading. Cognitive Development, 21(3), 349-368.

Oakhill, J., Cain, K., & Bryant, P.E. (2003). The dissociation of word reading and text comprehension : Evidence from component skills. Language and Cognitive Processes, 18(4) 443-468.

O'Connor, M., Arnott, W., McIntosh, B., & Dodd, B. (2009). Phonological awareness and language intervention in preschoolers from low socio-economic backgrounds: A longitudinal investigation. British Journal of Developmental Psychology, 27(4), 767-782.

Pressley, M. (2006). Reading instruction that works: The case for balanced teaching. (3rd ed.). New York: Guilford Press.

Roth, F.P., Speece, D.L., & Cooper, D.H. (2002). A longitudinal analysis of the connection between oral language and early reading. The Journal of Educational Research, 95 (5), 259-272.

Saada-Robert, M. (2004). Early emergent literacy. In T. Nunes, & P. Bryant (Eds.), Handbook of children's literacy (pp. 578-598). Dordrecht: Kluwer Academic.

Snowling, M.J. (2005). Literacy outcomes for children with oral language impairments: Developmental interactions between language skills and learning to read. In H.W. Catts, & A.G. Kamhi (Eds.), The connection between language and reading disabilities (pp. 55-75). Mahwah, NJ: Erlbaum.

Storch, S.A., & Whitehurst, G.J. (2002). Oral language and code-related precursors to reading: Evidence from a longitudinal structural model. Developmental Psychology, 38(6), 934-947

Torgesen, J.K. (1998), Catch them before they fall: Identification and assessment to prevent reading failure in young children. American Educator, 22, 32-39.

Van Steensel, R. (2006). Relations between socio-cultural factors, the home literacy environment and children's literacy development in the first years of primary education. Journal of Research in Reading, 29,4,367-382.

White, K.R. (1982). The relation between socioeconomic status and academic achievement. Psychological Bulletin, 91, 461-481.

Whitehurst, G.J. (1993). The Stony Brook Family Reading Survey. Stony Brook. NY: Author.

Ruth Fielding-Barnsley & Ian Hay

University of Tasmania

Associate Professor Ruth Fielding-Barnsley heads the literacy team at The University of Tasmania. Previously Ruth has lectured at University of New England, Armidale and the Queensland University of Technology. Ruth's research interests are in the areas of family literacy, the acquisition of reading, teachers' knowledge of metalinguistics, dyslexia and asperger's syndrome. She was an advisor for the new Australian Curriculum: English.

Professor Ian Hay is the Dean of the Faculty of Education, University of Tasmania, Australia. He has published over 100 book chapters, refereed journal articles and other related articles, and delivered over 100 conference presentations. His research interests include children's reading development, students with literacy and academic difficulties, and the role of motivation in learning.

Table 1
Four levels of language complexity and proficiency related to teacher
discourse and questioning (Blank et al., 2003)

Level of       Language Complexity to the       Example of Teacher
Complexity &   experience                       Discourse
Proficiency

1              Directly Supplied Information    What do you see?
               (Matching experiences)
               Classification

2              (Selective analysis of           What colour is that?
               experience)

3              Reorganisation                   Re-tell me the story

4              Abstraction and Inference        What made it happen?
               (Reasoning about experience)

Table 2
Correlations between children's rhyme, alphabet, receptive and
expressive language and speed of word naming

                    Rhyme     Alphabet     PPVT       HPNT      Speed

Rhyme                 1
Alphabet            .32 **       1
PPVT (Receptive
  Language)          .02       .40 **       1
HPNT (Expressive
  Language)          .11       .58 **     .69 **       1
Speed of word
  naming             .07      -.24 **    -.31 **    -.33 **       1

** Correlation is significant at the 0.01 level (2-tailed)

Table 3
Correlation between children's home reading environment and children's
ability with rhyme, alphabet, expressive and receptive language

                                 Rhyme    Alphabet     PPVT      HPNT

Level of home book reading
  to child                        .13      .48 **     .43 **    .38 **
Started reading to your child     .08       .21 *     .33 **     27 **
Number of picture books in
  the home                        .11      .42 **     .47 **    .36 **
Level of child asking to be
  read to                         .02      .31 **     .40 **    .30 **
Level child looks at books
  by him/herself                  .02      .25 **       .17       .06
Years of schooling of primary
  care giver                      .05      .48 **     .38 **    .33 **
Years of schooling of spouse      .07      .38 **     .34 **      .22

** Correlation is significant at the 0.01 level (2-tailed)

* Correlation is significant at the 0.05 level (2-tailed)

Table 4
Effect of type of intervention on a measure of spelling outcome
(ordered by magnitude)

Dependent Variable:      Mean     Std Error   Lower 95%CI   Upper 95%CI
FBH Spelling test
results

Class control group      42.10      1.32         39.49         44.71
Phonological
  experimental group     39.95      1.14         37.69         42.21
Language experimental
  group                  38.66      1.28         36.13         41.18
At-risk control group    34.00      1.75         30.53         37.47

Table 5
Effect of type of intervention on spelling outcomes with rhyming
and alphabet knowledge taken into account

Dependent Variable:                Mean    Std Error   Lower    Upper
FBH Spelling test results                              95%CI    95%CI

Phonological experimental group   40.99      1.13      38.76    43.22
Language experimental group       40.02      1.16      37.73    42.32
Class control group               38.24      1.63      35.00    41.48
At-risk control group             35.80      1.62      32.60    39.00