Phonological-based assessment and teaching within a first year reading program in New Zealand.Introduction
Children enter school with very large differences in early literacy-related experiences and language competencies (literate cultural capital) and these differences have a strong influence on learning to read and write at school. Research also suggests that the learning environments contribute to this variability (Gilmore, 1998; Shonkoff & Phillips, 2000; Snow, Burns & Griffin, 1998; Tunmer, Chapman & Prochnow, 2003; Tunmer, Prochnow, Greaney & Chapman, 2007). Furthermore, the longer it takes to identify and remediate these literacy problems in school, the more costly and difficult it becomes later. For this reason it is important that schools have early identification procedures (including relevant and effective intervention strategies) available soon after children enter school, in order to avoid later literacy-related learning difficulties.
There are many positive aspects about the way reading is taught in the first two years of school in New Zealand. Many of these positive aspects were discussed in the recent Education Review Office (ERO) report Reading and Writing in Years 1 and 2 (2009). However, there are also some problematic issues that have continued to act as barriers to advancement particularly in relation to attempts at closing the literacy (under)achievement gap that consistently manifests itself in the international literacy surveys such as the Progress in International Reading Literacy Study (PIRLS). Although New Zealand has a centralised education system and follows a national curriculum, there is a widening achievement gap between the top readers and those at the 'tail end' of the achievement spectrum. While there has been much international scientific-based research over the past three decades investigating the causes of low reading achievement, there has been (and continues to be) a Ministry of Education reluctance to accept many of the findings from this research to inform instructional policies particularly for those students who show early literacy learning difficulties. One example of this reluctance is the ministry's unwillingness to accept that the national multi-million dollar Reading Recovery (RR) program has failed to make any significant impact on closing the literacy achievement gap. In fact since the introduction of RR in the early 1980s, the literacy achievement gap in New Zealand has continued to widen (Greaney, 2004; Tunmer & Chapman, 2004a; Chapman, Greaney & Tunmer, 2007; Tunmer et al. 2007). This is ironic given that RR was initially designed to address the early literacy difficulties of the very students who later form the basis of the literacy achievement gap statistics. If schools require state-funded assistance for students with reading difficulties after one year at school it appears that RR is currently the only 'program of choice' for this group. Furthermore, the Observation Survey assessment tool designed for RR is now also accepted as the benchmark assessment for measuring the literacy progress of 6-year-olds after one year at school.
Review of the Literature
Children from low-income backgrounds are particularly susceptible to early reading difficulties because they often lack the necessary preschool exposure to the literacy experiences that promote the development of literate cultural capital (Blachman, 2000, Hart & Risely, 1995, Nicholson, 2003).
On three measure of literate cultural capital used in the PIRLS study (Mullis, Martin, Gonzales & Kennedy, 2003) that included, Early Home Reading Activities, Index of Home Education Resources and Parent's Attitudes Towards Reading, it was shown that the percentage of New Zealand children who fell into the highest category was high but the difference in future reading achievement between students in the high and low categories was also very large compared to most other countries. Furthermore, it has been argued that this difference in reading performance between the high
and low groups may also be attributed to the New Zealand reading approach used in most primary schools. Tunmer et al. (2007) state for example that 'the New Zealand approach to teaching literacy is generally beneficial to children with an abundance of literate cultural capital at school entry (typically middleclass children), but highly disadvantageous to children with limited amounts of literate cultural capital (typically children from low-income backgrounds), hence the relatively high level of disparity between New Zealand readers in later grades' (pp. 30-31).
Examples of literate cultural capital that enhance literacy development in young children at school entry include oral language and vocabulary (promoted by verbal interaction), basic understanding of book and print concepts (promoted by adult story book reading), knowledge of the alphabet (letter names and sounds) and basic phonological sensitivity that is promoted by exposure to simple rhymes, poems, nursery rhymes and songs and by playing simple language games like 'I spy'.
Letter-name knowledge has been shown to be one of the best predictors of beginning reading achievement and Foulin (2005) suggests that letter-name knowledge may contribute to early literacy development in several different ways. The alphabetic principle (i.e., awareness that printed letters and spelling patterns represent sounds) may be enhanced from knowledge of letter names. Furthermore, because many of the individual letter names also represent a constituent sound, (e.g., a, b, d, e) this knowledge acts as a precursor to more general letter-sound knowledge. A further advantage of letter-name knowledge is that it facilitates phonological sensitivity particularly when children are also exposed to alphabet texts and games. Adams, Foorman, Lundberg & Beeler (1998) also maintain that 'before children can make any sense of the alphabetic principle, they must understand that those sounds that are paired with the letters are one and the same as the sounds of speech' (p. 1). The ability to be consciously aware of these sounds is known as phonemic awareness, but the research evidence demonstrates that such awareness does not develop naturally for many children (Neilson, 2009). In support of this claim Adams et al. (1998) state for example that 'without direct instructional support, phonemic awareness eludes roughly 25% of middle class first graders and substantially more of those who come from less literacy-rich backgrounds' (p. 1).
Children who enter primary school with higher levels of cognitive and phonological abilities profit more from reading instruction and learn to read sooner than children who enter school without such skills (Bowey, 2005; Elbro & Scarborough, 2004; Tunmer, Chapman & Prochnow, 2006). In a longitudinal study Tunmer et al. (2006) found that on school entry measures of letter name knowledge, phonemic awareness and receptive vocabulary accounted for nearly half the variance in reading comprehension scores in year 7. Further support for the importance of early phonemic awareness as a determinant of later reading success comes from Adams et al. (1998) when they state that 'faced with an alphabetic script, a child's level of phonemic awareness on entering school is widely held to be the strongest single determinant of the success she or he will experience in learning to read, or conversely, the likelihood that he or she will fail' (p. 2).
While there are various components that influence literate cultural capital, there is evidence to suggest that the most at-risk readers have very limited phonemic awareness and decoding skills (Torgesen, 2004). These particular students require more explicit, intensive and systematic instruction in these skills than is usually provided within most regular junior school classes. Tunmer and Prochnow (2009) maintain for example that 'for children experiencing difficulty in developing the ability to perceive intuitively the systematic connections between speech and print as a by-product of more general reading, more intensive instruction in phonemic awareness and alphabetic coding skills is likely to be necessary' (pp. 172-73).
A major shortcoming when identifying children with literacy learning problems has been a failure to specify factors known to be causally related to learning to read (Tunmer & Greaney, 2010). Poor literacy achievement has often been attributed to disability rather than ineffective instruction. In a 3 year longitudinal study Vellutino, Scanlon and Jaccard (2003) investigated the extent to which ineffective instruction may be a possible cause of later persistent reading difficulties. The children who had significant reading problems in first grade were provided with explicit one-on-one remedial instruction during one semester that involved combinations of, context reading practice, vocabulary development, phonemic awareness, phonetic decoding and writing. The results showed that at the conclusion of the intervention, 67% of the children were within normal range on several phonological processing measures. Some of the earliest studies on the assessment and teaching of phonological-based skills in young children were conducted by Bradley and Bryant (1978; 1983). These studies demonstrated the positive correlations between early rhyme skills and later reading and spelling development. In a more recent study involving the teaching of phoneme blending and segmentation skills to kindergarten children, Yeh (2003) found that significant progress in these skills was made after only nine hours of instruction. In a similar study, Bailet, Repper, Piasta & Murphy (2009) developed 18 (30 minute) explicit and sequential lessons with 220 kindergarten children that focused on the teaching of letter names, letter sounds, syllable segmentation, rhyme alliteration and onset-rime awareness. The results showed significant gains in emergent literacy skills after only nine weeks of instruction.
This paper discusses a study that focused on incorporating a series of phonological-based literacy assessments and instructional activities within a regular New Zealand year one (i.e., first year) class. Children in New Zealand begin school on or soon after their fifth birthday and the intervention group were aged between 5 years and 5.8 years at the beginning of the study (mean age 5.4 years).
Aims of the study
There were three aims of the study. The first aim was to assess the phonological-based skills of a control group of predominantly Maori and Pasifika children who had been at the school for between 12 and 20 months and had received only the regular class literacy program. The second aim was to assess some phonological-based skills in a class of predominantly Maori and Pasifika new entrant students who had recently started school. The third aim was to present a series of explicit teaching activities that focused on the development of phonological-based skills for these new entrant students as supplementary tasks within the regular reading program, and to assess the effects of these activities on the students' literacy progress at age 6 years.
Rationale for the study
The following reasons formed the rationale for the study:
1. The six year Observation Survey (Clay, 2005) is currently the first and only formal literacy assessment used in most New Zealand year 1 classes. Because this assessment tool assesses literacy progress after one year of school, there is the possibility that children who are at risk of developing literacy learning difficulties are not identified early enough.
2. Because there are no specific phonemic awareness assessment components in the Observation Survey, teachers are not alerted to the likely problems and implications of such components for particular students who may be at risk of developing reading difficulties.
3. There has been a lack of research-informed literacy professional development for teachers of year 1 children in New Zealand, as most of the Ministry of Education literacy professional development initiatives have tended to focus on years 3 and above.
4. The recently-developed Reading and Writing Standards (Ministry of Education, 2009) and Literacy Learning Progressions (Ministry of Education, 2010) do not include a focus on the effective assessment and teaching of phonological-based early literacy skills within the first year of schooling.
The research design
The study involved two phases. The first phase involved an investigation of the reading-related knowledge and skills of a group of 26 students who had been at school for up to 20 months (age range 6 years to 6.8 years) and had received the regular in-class literacy program over this period. The data from this investigative phase formed the control group data for the study. As well as the regular six year Observation Survey (Clay, 2005) data, several other phonological-based assessment measures were administered individually to each student. The school setting was an urban city with a predominantly Maori and Pasifika roll.
The intervention phase involved the assessment and teaching of a group of 15 children over a 10 week period during the second term of their first year of school. Each child was administered a series of phonological-based assessments on three occasions. The first (i.e., pre-test) assessments occurred early in their first year of school followed by a second (i.e., post-intervention) assessment immediately following the conclusion of the intervention. The final assessments (i.e., post-tests) occurred as each student reached their 6th birthday. The regular Observation Survey assessment was also administered to each child on or soon after their sixth birthday.
The literacy-related assessments for controls (phase one)
The following assessment data were administered to the non-intervention control group.
Observation Survey (Clay, 2005)
This assessment tool includes the following components:
Letter name knowledge (n=54), Concepts about Print (n=24), a list of 15 sight words, a measure of writing vocabulary and a Hearing and Recoding Sounds in Words sub-test (n=37). Standard administration and scoring procedures were used. A measure of current book reading level was also obtained from the class teachers' records.
Burt word reading test
The Burt word test (Gilmore, Croft & Reid, 1981) is a New Zealand standardised measure of isolated word reading ability. A list of 110 words of increasing difficulty is presented for the student to read aloud. Testing continues until 10 consecutive errors are made and no corrective feedback is given.
The Adams et al. (1998) assessment tool was used to assess sub-components of phonemic awareness. The phonemic awareness subcomponents included: detecting rhymes, counting syllables, matching initial sounds, counting phonemes, comparing word lengths and representing phonemes with letters. Each subcomponent measure contained practice items followed by five assessment task items. The maximum score for the phonemic awareness test was 30.
Phoneme segmentation task
A modified version of a test developed by Tunmer, Herriman & Nesdale (1988) was used to measure phonemic segmentation ability. The students used counters to represent the sounds in orally presented pseudo-words of varying length (e.g., ek, fip dilt). A correct score was given if the student was able to place the relevant number of counters that represented each phoneme for each target pseudo-word. The maximum score for the phoneme segmentation task was 24.
A pseudo-word reading test is a measure of phonological processing ability, and it is this ability that determines the student's ability to decode both real words and pseudo or non-words by using familiar letter-sounds and spelling patterns knowledge. According to Moats (2000) 'skilled reading presents a paradox: students who can most easily make sense of text are also those who can most easily read nonsense' (p. 9). There is a correlation between the ability to read pseudo-words (non-words) and real words, and a pseudo-word reading test is therefore an effective measure of decoding (phonological processing) skill.
Thirty monosyllabic pseudo-words from section 3 of the decoding skills test (Richardson & Di Benedetto, 1985) were used to measure knowledge of letter-sound patterns. The pseudo-words were presented in order of increasing difficulty ranging from simple consonant-vowel-consonant patterns (jit, dut) to blends, digraphs and vowel variations (prew, fruice). Two scoring procedures were used. A score was given for total words read (n=30) and for correct phonemes read (n=101).
Literacy related assessments for the intervention group (phase two)
As well as the assessments used for the controls, the intervention group also received measures of letter sound knowledge and letter writing ability (i.e., the ability to correctly write each letter unassisted).
Following the pre-testing, a 10 week in-class intervention was undertaken that involved a series of semi-structured daily lessons that focused on teaching some phonological-based skills to both the whole class and small groups of students. The aim was to allow the regular class teacher (and a teacher aide) to initially observe these lessons and to eventually take over the responsibility for including the tasks as supplements to their regular literacy program. Each teaching session was taken on a daily basis for four days per week for 10 weeks during the second school term in 2010. The whole class component of each lesson was generally taken over a 30 minute period, although small group and individual repeat lessons were often extended for another 15 minutes for those students who required additional teaching and/or practice. This mirrors what often occurs within the regular classroom where additional small group teaching may occasionally be necessary.
The lesson activities and tasks
A series of resources and games were purchased and/or designed for use with both the whole class and for individuals and small groups to teach the following phonological-based skills.
Letter names and sounds
To encourage the development of letter name knowledge the alphabet song was taught to the class. A set of laminated letters was made as a necklace for each child (who had scored low on the letter name pre-test). The necklace was worn and taken home and the parents were encouraged to ask their children to say the letter names. Each child was also given a commercial pack of alphabet cards that contained matching pictures of objects beginning with the relevant sounds. Again these were sent home to encourage parents to assist with their learning. For further development of letter name and sound knowledge the students were given counters and cards on which they were asked to 'place counters on something beginning with either a letter name' (or a sound).
Those students who had scored poorly on the letter writing task in the pre-test were each given an exercise book and were asked to write each unknown letter (e.g., a letter a day) many times. This task was completed when each student was able to correctly write the hitherto unknown letters unassisted. Periodic and random 'mini' tests' were used to check progress. Again these exercise books were taken home daily to allow the students to further practise their letter writing tasks.
Phonemic awareness activities
Bingo cards containing pictures of common objects (e.g., cat, dog) were used to teach phonemic awareness. For example, rhyme awareness was taught by asking the students to place counters on any pictures that rhymed with a given noun (e.g., 'cover a picture that rhymes with hen, log, van'). Syllable counting was taught by asking the students to place counters on pictures that were presented via syllables (e.g., 'cover mo/tor/bike/, kit/ten'). Phoneme blending was taught by asking the students to cover pictures after the items were presented via phonemes (e.g., 'cover d/o/g, v/a/n'). Initial phoneme deletion was taught by asking students to name objects (e.g., cake, dog, boat etc) followed by deletion of the initial sound (e.g., ake, og).
Linking phonemic awareness to spelling
To develop the link from phonemic awareness to spelling (by encouraging phoneme-grapheme awareness) the students were presented with a series of pictures of CVC nouns (e.g., cat, van, dog, pen etc) and were asked to first, place the relevant number of counters into spaces to represent each phoneme. The second task was to write the target word but again using the counters to reinforce each sound before it was written.
Linking the phonological-based skills to context reading and writing
During the regular class guided reading lessons the students were continually encouraged to use their phonological-based skills to assist with word identification. This aspect was modelled within regular whole class shared and guided reading lessons to assist the students to use the phonological-based skills for word identification during context reading.
Results for phase one
Table 1 presents the results from phase 1. This involved assessing the nonintervention control group of 27 children who had been at school for between 12 and 20 months. The assessment data from this phase included the regular Observation Survey (Clay, 2005) and several phonological-based assessments outlined previously.
Table 1 presents the results for the non-intervention group's Observation Survey data and the phonological-based measures. This group of students had received only the regular class reading instruction during their first 12-20 months of school. Because letter name knowledge is the preferred (and usually only) measure of alphabet knowledge assessed when the Observation Survey is administered, a measure of letter sound knowledge was also included in this study. The results indicate that while letter name knowledge was generally acceptable (mean 48), there were lower scores for the letter sounds knowledge (mean 43.0). This lower letter sounds knowledge was also reflected in the low scores for both the Burt word reading measure (mean 11.9) and the book reading level (mean 6.5). An average reader's raw score on the Burt test at age 6 years is 20 words and the benchmark book reading level for a student who has been at school for 12 months is level 12. The control group's scores on both these reading measures indicate that these students were already at serious risk of developing reading difficulties. Furthermore, the mean scores for the phonemic awareness sub-task measures indicate that, despite having attended school for at least 12 months, the control group had still not developed satisfactory levels on these measures. Similarly the very low mean scores on both the phoneme segmentation task (10.4) and the pseudo-word reading task (1.2) further indicated that the regular in-class literacy programs that these students had received in their first year of instruction have had minimal effect on the development of these important literacy-related skills.
Results for phase two (pre-tests)
The phonological-based assessment measures used in phase one were also administered to the intervention students who had been at school for between 1 and 8 months. Because students in New Zealand start school on or soon after their 5th birthday, the age range in any one class and at any time, varies accordingly, and so at any point in time, there will be a range of ages in any first year class. However, it was interesting to note that even after up to eight months of instruction, the pre-test mean scores for most of the measures were still very low, suggesting that the regular class program was not adequately addressing these phonological-based learning needs. Following the pre-test assessments, the in-class phonological-based intervention was carried out daily over a 10 week period. Tables 2 and 3 present the results from the pre-and post-tests for the intervention group.
The data in Table 2 show that at pre-test the students could recognise approximately 24 letter names and 17.8 letter sounds. This represents less than half the letter names and less than one third of the sounds. The Burt word test mean score at pre-test was 1.0 and the mean book reading level was 1.8.
The results from the phonemic awareness measure ranged from 1.3 (word length) to 0.0 for the spelling task. Both the phoneme segmentation task (mean 1.1) and the pseudo-word reading task (mean 0.0) further indicated that the intervention group had low levels in these important literacy-related areas of knowledge at pre-test.
Two further spelling measures (alphabet letter writing and simple VC and CVC word spelling) were used in phase two of the study and the results in Table 3 show that at pre-test, the students were able to correctly write 10 of the 26 letters of the alphabet. The mean score for spelling the 10 VC words (e.g., an, up) was 0.3 and the CVC words (e.g., cat, dog) was 0.0. These results reflect a very low level of sound-letter awareness.
Results for phase two post-tests (including comparisons with controls)
The post-test data will be discussed in two sections. The first section will discuss comparative data between the two groups at age 6 years. The second section will present data that further highlights the intervention group's progress from pre-test through to post-test on the monitored measures.
The data presented in Table 4 shows that at age 6 years, the intervention group had out-performed the controls on every measure.
While some of the differences may have been small, these are none-the-less, interesting results. The intervention group out-performed the controls on all the Observation Survey sub-test measures. Furthermore, this group's mean Burt word score of 18.2 (versus 11.9 for controls) suggests that the effects of the phonological-based intervention had positively impacted on the students' ability to transfer the skills to word reading. Similarly, the higher book reading level of the intervention group (8.5 versus 6.4) also suggests positive transfer effects.
Similar trends for the phonological-based measures (at post-test) also indicate that the intervention group out-performed the controls. In particular, the intervention group's pseudo-word reading (sounds) almost doubled the control group's mean score on this measure (40.0 versus 22.3). This result suggests that the intervention group were more prepared to attempt many of the pseudo-words presented in this test whereas the controls tended to make no attempt at all on many of the items. The increased awareness of how letters and spelling patterns map on to print that the intervention group had developed as a result of the intervention, has positively impacted on their willingness to attempt to decode pseudo-words. Such a confidence was not apparent with the controls.
Progress tracking for intervention group
There were three assessment points (i.e., pre-test, post-intervention and posttest), but because of the age variation of the intervention group, the final posttest assessment point (at age 6 years) did not occur for up to 6 months after the conclusion of the intervention for some of the students. However, this was a positive outcome because it allowed a measure of longer term transfer effect to be considered.
The following figures present progress monitoring data for several of the measures including: letter names knowledge, letter sounds knowledge, letter writing ability, phonemic awareness, basic VC and CVC word spelling, Burt word reading and Book reading level. The figures also show the nonintervention control group's mean post-test scores at age 6 years for the same measures for comparison. Because this control group was only assessed at one point, their data on the figures are indicated in the ovals. The results for each of these measures will now be presented and discussed.
The data in Figure 1 show that at pre-test, the intervention group's mean letter name knowledge score was 24.0 and at post-intervention was 49.8. The mean score at age 6 years for this measure was near ceiling at 52.7 for the intervention group compared to 48.0 for the controls at the same age. The difference between these two groups for the letter names was minor, but was nonetheless, higher for the intervention group.
The same pattern of results occurred for the letter sounds knowledge.
The data in Figure 2 show that the mean letter sounds knowledge score for the intervention group at pre-test was 17.8. At post-intervention the mean score for this measure increased to 47.8 and continued to improve to 52.3 at post-test at age 6 years, compared to 43.0 for the controls at the same age. The most interesting finding from Figure 2 was the rate of increase for the intervention group between the pre-test and post-intervention assessment points. These results demonstrate the effectiveness of the intervention which included an explicit teaching focus on the letter sounds with those students who had scored poorly on the pre-test for this measure. The letter sounds knowledge for many students in the control group had still not fully developed even after up to 20 months of instruction.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
Similarly the data in Figure 3 illustrate an increased performance over a short timeframe for letter writing ability following the explicit teaching of these skills in the intervention.
At the outset of the intervention (i.e., pre-test) the mean score for the letter writing task was 10.0 and this increased to 23.5 letters at post-intervention and 24.0 at age 6 years. This measure was not used for the control group. However letter confusions and reversals often remain throughout the first year of schooling and the explicitness of the teaching within the intervention demonstrated that such confusions may be greatly reduced in a relatively short timeframe with explicit teaching.
Six phonemic awareness subtasks (Adams et al. 1998) were administered to both the intervention and control groups and the results for these measures are presented in Figure 4.
The data in Figure 4 show that at pre-test the mean score for the intervention group for the phonemic awareness measures was 4.6 but had increased to 14.6 at post-intervention and to 19.5 at age 6 years. The control group mean score for the phonemic awareness measures at age 6 was only 2.0 even though this group had received up to 20 months of schooling. These results suggest that the explicit phonological-based teaching tasks had impacted positively on the development of phonemic awareness. Furthermore, this awareness continued to develop after the completion of the intervention.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
A similar pattern of progress was evident for the phoneme segmentation task.
The data in Figure 5 show that the intervention group mean score at pretest for the phoneme segmentation task was 1.0 but at age 6 years was 14.8 compared with 10.0 for the controls.
Two reading measures (Burt words and book reading levels) were also used in the study. The data in Figure 6 show that the intervention group's mean Burt word score increased from 1.0 at pre-test to 9.8 at post-intervention and 18.2 at post-test. At age 6 the control group's mean Burt score was 11.9.
A raw score of 20 words for the Burt word test represents an 'average' score for a 6-year-old, so the data in Figure 6 indicate that while the students were still performing below par, the intervention group were performing at significantly higher levels than the controls and were in fact, not far below average. These results indicate that the intervention had a positive effect on the word reading ability of the intervention group. These results are further confirmed in the book reading level scores presented in Figure 7.
The mean book reading level scores for the intervention group at pre-test was 1.8 and this increased to 5.8 at post-intervention. By age 6 the mean book reading level for this group was 9.3 compared to 6.5 for the controls. A proxy measure for 'average' book level at age 6 years is level 12, so the data in Figure 7 show that both groups are still performing below par. However, as with all the measures, the intervention group has out-performed the controls at age 6 years.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
The data for the final measure (basic spelling) are presented in Figure 8. This measure was not administered to the control group.
The data in Figure 8 show that at pre-test the intervention group's mean score for basic spelling of VC and CVC words was 3.0. At post-intervention this increased to 8.2 and at age 6 was 11.0. This data suggests that the phonological-based intervention impacted positively on the students' ability to segment the sounds in basic words and to use this knowledge to assist with spelling.
This study has highlighted several key issues related to the assessment and teaching of early literacy skills to students in their first year of school. The first issue concerns the importance of establishing relevant research-based assessments to enable teachers to assess the literacy-related skills and knowledge of students as soon as possible after school entry. This earlier assessment is in opposition to the current status quo procedure in which most schools in New Zealand leave the first major literacy assessment (i.e., Observation Survey) until the students' sixth birthday and after 12 months at school. Delaying the initial major assessment for 12 months represents a 'wait-to-fail' approach. However, although the Observation Survey has been the accepted benchmark literacy assessment measure in New Zealand for most 6-year-olds for nearly thirty years, there is a lack of phonemic awareness measures in this tool. While the Hearing and Recording Sounds in Words sub-test focuses on spelling letter-sound patterns, it is not a genuine measure of phonemic awareness. This is because many of the target words in this test may be successfully spelled from memory without the necessity of the respondent to consciously reflect on the separate phonemes. Some words that appear in this sub-test such as I, at, a, am, to, up, the, is, and, me, on and go for example, are likely to be recalled (and spelled) as high frequency sight words. Such familiarity would therefore enable students to spell most of them without the need for any conscious reflection on their separate phonemic elements in isolation. Furthermore, genuine phonemic awareness assessment tasks usually demand oral responses and this isn't required in the Clay test. In support of the claim that more phonological-based assessments are required at the outset of schooling Cunningham (1999) states that 'If children lack phonemic awareness when they come to school, it is the responsibility of the school to see to it they have every opportunity to develop it' (p. 70). However, in the absence of any appropriate phonemic awareness measures, teachers are unlikely to be aware of such weaknesses in their students and will therefore be even less likely to develop appropriate instructional programs to address such weaknesses.
A second concern is the level of general acceptance within the New Zealand Ministry of Education (through its continued funding support) that Reading Recovery is viewed as the most effective program for addressing the literacy learning needs of students from schools in low socio-economic areas, and in particular, Pasifika and Maori students. Although this program has been operating in schools since the early 1980s, the international literacy surveys have shown that New Zealand's gap between the top and bottom achievers continues to be one of the greatest in all the participating countries. Furthermore, these surveys also show that many of the lowest performing students are from low socio-economic area schools that often include many Maori and Pasifika students. However, it is even more concerning that the data from the RR annual statistics consistently show that these very same groups are generally less successful in RR and are often referred on to other specialist services. Lee (2009) noted for example that 'Boys, Maori and Pasifika students and those from low decile schools were more likely to be referred on for further support' (p. 31). These findings are not surprising given that the RR program lacks an emphasis on both the assessment and teaching of phonological-based skills (including phonemic awareness) even though the scientific research literature demonstrates the critical importance of such skills, especially for students at risk of developing literacy learning difficulties (Chapman, et al. 2007; Tunmer & Chapman, 2003; Tunmer & Chapman, 2004b; Chapman, Tunmer & Prochnow, 2001; Tunmer & Greaney, 2008; Iversen & Tunmer, 1993).
According to Tunmer and Chapman (2004b) there are two advantages in providing struggling readers with explicit systematic instruction in word identification strategies both within and outside the context of regular reading. First, explicit context-free instruction in word analysis (i.e. phonics) encourages full attention to the orthographic patterns that make up words. Second, explicit and systematic word-level instruction outside the context of regular reading helps struggling readers rely less on ineffective context-based cues such as looking at illustrations, predicting unfamiliar words from contexts or guessing. However, before the student is able to make any sense of phonics he must first have developed sufficient phonemic awareness. Without sufficient phonemic awareness, phonics is meaningless. The superior performances on the reading measures (Burt words, pseudo-word reading, and book level reading) shown by the intervention group at post-test, were likely to be due to both the phonemic awareness teaching tasks and the explicit word-level strategy training tasks that were included in the intervention phase of the study. Without the phonemic awareness tasks, the word-level strategy training would not have been as successful.
A design weakness relates to the small sample, especially in the intervention group. While the non-intervention control group included 26 children (who had been at school for up to 20 months), the intervention group contained only 15 children. This small sample was based on the total enrolments in one 'new entrant' class at the time of the study.
Second, although teacher effects may be seen to have influenced the positive outcomes of the results for the intervention group, it must be emphasised that while 2 students received post-tests immediately following the conclusions of the intervention (because they had their 6th birthday at this time), the remainder of the students (i.e., 13) did not receive their post-tests until up to 8 months following the conclusion of the study. It is also noteworthy that many of these 'late-tested' students had also received up to three different teachers since completing the study. These frequent changes would have also caused disruptions to their regular programs yet their post-test results still indicated positive effects.
There are several key issues arising from the study that may go some way to explaining some of the reasons for New Zealand's widening literacy (under) achievement gap problem that is repeatedly presented in the international literacy surveys over the last forty years. In each of these surveys the same results show that Maori, Pasifika and students from low socio-economic area schools form the bulk of those appearing in the tail of underachievement. Furthermore, since the introduction of RR in the early 1980s, the tail of underachievement has continued to lengthen even though at the same time, the Ministry of Education continues to support and fund RR as the 'program of choice' for the very students who later are represented in the tail statistics. The research evidence also suggests that RR is not necessarily any more effective than many other programs. For example, in their meta-analysis of one-to-one tutor programs Elbaum et al. (2000) noted that 'the findings of this metaanalysis do not provide support for the superiority of Reading Recovery over other one-to-one reading interventions. Typically, about 30% of students who begin Reading Recovery do not complete the program and do not perform significantly better than control students' (p. 617).
If we are serious about addressing the needs of the students who are at risk of developing early literacy learning difficulties, we need to first, identify at the outset of schooling, those students who are at risk of developing such difficulties. For this to occur, teachers need more appropriate research-based literacy assessments than are currently available. More appropriate instructional interventions are also required to enable teachers to more effectively teach those particular students who require more explicit phonological-based instruction than is usually available within most whole language classrooms.
Adams, M.J., Foorman, B.R., Lundberg, I., & Beeler, T. (1998). Phonemic awareness in young children. Baltimore: Paul H. Brookes Publishing.
Bailet, L.L., Repper. K.K., Piasta, S.B., & Murphy, S.P. (2009). Emergent literacy intervention for prekindergartners at risk for reading failure. Journal of Learning Disabilities, 42(4), 336-355.
Blachman, B.A. (2000). Phonological awareness. In M.L. Kamil, P.B. Mosenthal, P.D. Pearson & R. Barr (Eds.), Handbook of Reading Research, (Vol 3, pp. 483-502). Mahwah, NJ: Erlbaum.
Bowey, J.A. (2005). Predicting individual differences in learning to read. In M.J. Snowling & C. Hulme (Eds.), The Science of reading: A handbook (pp. 155-172). Oxford: Blackwell.
Bradley, L., & Bryant, P.E. (1978). Difficulties in auditory organization as a possible cause of reading backwardness. Nature, 271, 746-747.
Bradley, L., & Bryant, P.E. (1983). Categorizing sound and learning to read: a causal connection. Nature, 301, 419-521.
Chapman, J.W., Greaney, K.T., & Tunmer, W.E. (2007). How well is Reading Recovery really working in New Zealand? New Zealand Journal of Educational Studies, 42(1), 17-29.
Chapman, J.W., Tunmer, W.E., & Prochnow, J.E. (2001). Does success in the Reading Recovery program depend on developing proficiency in phonological processing skills? A longitudinal study in a whole language instructional context. Scientific Studies of Reading, 5, 141-176.
Clay, M.M. (2005). An observation survey of early literacy achievement. Auckland, NZ: Heinemann.
Cunningham, P.M. (1999). What should we do about phonics? In: L. Gambrell, L. Morrow, S. Neuman & M. Pressley (Eds), Best practices in literacy instruction (pp. 68-89). New York: Guilford Press.
Education Review Office. (2009). Reading and writing in years 1 and 2. Wellington, NZ: Education Review Office
Elbaum, B., Vaughn, S., Hughes, M., & Moody, S. (2000). How effective are one-to-one tutoring programs in reading for elementary students at risk for reading failure? A meta-analysis of the intervention research. Journal of Educational Psychology, 92, 605-619.
Elbro, C. & Scarborough, H.S. (2004). Early identification. In T. Nunes & P. Bryant (Eds.), Handbook of children's literacy (pp. 339-359). Dordrecht, The Netherlands: Kluwer Academic.
Foulin, J.N. (2005). Why is letter-name knowledge such a good predictor of learning to read? Reading and Writing, 18, 129-155.
Gilmore. A., Croft, C., & Reid, N. (1981). Burt word reading test: New Zealand revision. Wellington, NZ: New Zealand Council for Educational Research.
Gilmore, A.M. (1998). School entry assessment. Wellington, NZ: Ministry of Education.
Greaney, K. (2004). First to fourth to thirteenth (and in all probability) still dropping?
New Zealand's international literacy results: some personal thoughts about the reasons for the gap. DELTA, 56(2), 53-64.
Hart, B., & Risley, T.R. (1995). Meaningful differences in the everyday experiences of young American children. Baltimore, MD: Brookes.
Iversen, S.A., & Tunmer, W.E. (1993). Phonological processing skills and the Reading Recovery programme. Journal of Educational Psychology, 85, 112-125.
Lee, M. (2009). Annual Monitoring of Reading Recovery: The data for 2008. Wellington, NZ: Ministry of Education.
Ministry of Education. (2009). Reading & Writing Standards for years 1-8. Wellington, NZ: Learning Media.
Ministry of Education. (2010). The Literacy Learning Progressions: Meeting the reading and writing demands of the curriculum. Wellington, NZ. Learning Media.
Moats, L.C. (2000). Speech to print: language essentials for teachers. Baltimore, MD: Paul H. Brookes.
Mullis, I.V.S., Martin, M.O., Gonzalez, E.J., & Kennedy, A.M. (2003). PIRLS 2001 international report. Boston, MA: International Study Centre, Lynch School of Education, Boston College.
Neilson, R. (2009). Assessment of phonological awareness in low-progress readers. Australian Journal of Learning Disabilities, 14(1), 53-66.
Nicholson, T. (2003). Risk factors in learning to read. In B. Foorman (Ed.), Preventing & remediating reading difficulties: Bringing Science to Scale (pp. 165-193). Timonium, MD: York Press.
Richardson, E., & Di Benedetto, B. (1985). Decoding skills test. Parkton, MD: York Press. Shonkoff, J., & Phillips, D.A. (Eds.), (2000). From neurons to neighbourhoods. Washington, DC: National Academy of Sciences.
Snow, C.E., Burns, M.S. & Griffin, P. (Eds.), (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press.
Torgesen, J.K. (2004). Lessons learned from research on intervention for students who have difficulty learning to read. In P. McCardle & V. Chhabra (Eds.), The voice of evidence in reading research (pp. 355-382). Baltimore, MD: Brookes.
Tunmer, W.E., Herriman, M.L., & Nesdale, A.R. (1988). Metalinguistic abilities and beginning reading. Reading Research Quarterly, 23, 134-158.
Tunmer, W.E., Chapman, J.W., & Prochnow, J.E. (2003). Preventing negative Matthew effects in at-risk readers: a retrospective study. In B. Foorman (Ed.), Preventing and remediating reading difficulties: Bringing science to scale. (pp. 121-163). Timonium, MD: York Press.
Tunmer, W.E., & Chapman, J.W. (2003). The Reading Recovery approach to preventive early intervention: As good as it gets? Reading Psychology. 24, 337-360.
Tunmer, W.E., & Chapman, J.W. (2004a). Why the reading achievement gap won't go away: Evidence from the PIRLS 2001 international study of reading achievement, DELTA, 56(2), 69-82.
Tunmer, W.E., Chapman, J.W. (2004b). Reading Recovery: distinguishing myth from reality. In R.M. Joshi (Ed.), Dyslexia: myths, misconceptions, and some practical applications (pp. 99-114). Baltimore, MD: International Dyslexia Association.
Tunmer, W.E., Chapman, J.W. & Prochnow, J.E. (2006). Literate cultural capital at school entry predicts later reading achievement: A seven year longitudinal study. New Zealand Journal of Educational Studies, 41, 183-204.
Tunmer, W.E., & Greaney, K.T. (2010). Defining dyslexia. Journal of Learning Disabilities, 43, 229-243.
Tunmer, W.E., Prochnow, J.E., Greaney, K.T., & Chapman, J.W. (2007). What's wrong with New Zealand's national literacy strategy? In R. Openshaw & J. Soler (Eds). Reading across international boundaries: history, policy and politics (pp. 19-42). Charlotte, NC: Information Age Publishing.
Tunmer, W.E., & Prochnow, J.E. (2009). Cultural relativism and literacy education: Explicit teaching based on specific learning needs is not deficit theory. In R. Openshaw & E. Rata (Eds.), The politics of conformity in New Zealand (pp. 154-190). Auckland, NZ: Pearson.
Tunmer, W.E., & Greaney, K.T. (2008). Reading intervention research: An integrative framework. In G. Reid, A. Fawcett, F. Manis, & L. Siegel (Eds.), The Sage handbook of dyslexia (pp. 241-267). London: Sage.
Tunmer, W.E., & Greaney, K.T. (2010). Defining Dyslexia. Journal of Learning Disabilities, 43, 229-243.
Vellutino, F.R., Scanlon, D.M., & Jaccard, J. (2003). Towards distinguishing between cognitive and experiential deficits as primary sources of difficulty in learning to read: A two-year follow-up of difficult-to-remediate and readily remediated poor readers. In B. Foorman (Ed.), Preventing and remediating reading difficulties: Bringing science to scale (pp. 73-120). Baltimore, MD: York Press.
Yeh, S.S. (2003). An evaluation of two approaches for teaching phonemic awareness to children in Head Start. Early Childhood Research Quarterly, 18, 513-529.
Keith Greaney & Alison Arrow
Table 1. Raw score means and standard deviation for all measures for the non-intervention control group Mean SD Age (months) 75.5 OBSERVATION SURVEY Letter names (54) 48.0 8.5 Letter sounds * (54) 43.0 15.8 Concepts about Print (24) 12.7 2.8 Clay Words (15) 7.2 4.3 Writing Vocabulary 16.4 10.7 Hearing & Recording Sounds in Words (37) 24.6 10.7 Burt Words (110) 11.9 1.3 Book Reading Level 6.5 3.7 PHONEMIC AWARENESS Rhyme Detection (5) 1.9 1.9 Counting Syllables (5) 2.2 1.7 Initial Phonemes (5) 2.6 1.9 Phoneme Counting (5) 1.2 0.9 Word Length (5) 2.9 1.4 Spelling (5) 1.9 1.7 PHONEME SEGMENTATION (24) 10.4 0-19 PSEUDO-WORD READING Total words (30) 1.2 0-15 Total sounds (101) 22.3 0-86 * Not usually assessed Table 2. Raw score means and standard deviations for all measures for the intervention group as a function of time of testing Pre-test Post-test Mean SD Mean SD OBSERVATION SURVEY Letter names (54) 24.0 22.0 52.7 2.0 Letter sounds (54) 17.8 19.5 52.3 2.7 Concepts about Print (24) N/A 15.8 3.1 Clay words (15) N/A 10.3 3.7 Writing vocabulary N/A 17.7 10.6 Hearing & Recording Sounds (37) N/A 25.5 8.1 Burt Words (110) 1.0 2.2 18.2 9.8 Book reading level 1.8 1.2 8.5 5.4 PHONEMIC AWARENESS Rhyme detection (5) 0.4 1.3 3.7 1.9 Count syllables (5) 1.0 1.5 3.0 1.6 Initial phonemes (5) 0.9 1.4 4.5 1.4 Phoneme counting (5) 0.8 1.1 2.1 1.2 Word length (5) 1.3 1.2 3.0 4.5 Spelling (5) 0.0 0.0 2.8 1.7 PHONEME SEGMENTATION (24) 1.1 2.8 13.1 5.8 PSEUDO-WORD READING Total words (30) 0.0 0.0 3.0 4.5 Total sounds (101) 1.6 6.1 40.0 25.2 Table 3. Mean raw scores and standard deviations for letter writing and basic (VC & CVC) word spelling measures for intervention group as a function of time testing Pre-test Post-test Mean SD Mean SD Letter Writing (26) 10.0 9.6 23.1 2.9 BASIC WORD SPELLING VC Words (10) 03 0.61 6.5 2.9 CVC Words (10) 0.0 0.0 4.9 3.4 Table 4. Comparative means and standard deviations for the all measures for both groups at age 6 years Intervention Controls group Mean SD Mean SD OBSERVATION SURVEY Letter Names (54) 48.0 8.5 52.7 2.0 Letter Sounds (54) 43.0 15.8 52.3 2.7 Concept about Print (24) 12.7 2.8 15.8 3.1 Clay Words (15) 7.2 4.3 10.3 3.7 Writing Vocabulary 16.4 10.7 17.7 10.6 Hearing & Recording Sounds 24.6 10.7 25.5 8.1 Burt Words (110) 11.9 1.3 18.2 9.8 Book Level 6.4 3.7 8.5 5.4 PHONEMIC AWARENESS Rhyme Detection (5) 1.8 1.9 3.7 1.9 Counting Syllables (5) 2.2 1.7 3.0 1.6 Initial Phonemes (5) 2.6 1.9 4.5 1.4 Phoneme Counting (5) 1.2 0.9 2.1 1.2 Word length (5) 2.9 1.4 3.0 4.5 Spelling (5) 1.9 1.7 2.8 1.7 PHONEME SEGMENTATION (24) 10.4 6.7 13.1 5.8 PSEUDO-WORD READING Total Words (30) 1.1 2.8 3.0 4.5 Total Sounds (101) 22.3 23.2 40.0 25.2