of Phonological Awareness as a Predictor of Basic
The importance of phonological awareness as a foundation for the
development of basic reading skills has been demonstrated in recent studies and
reports (Ehri, 1991, 1994; Morais, Bertelsen, Cary, & Alegria, 1987; Share,
Jorm, MacLean, & Matthews, 1984; Snow, Burns, & Griffin, 1998;
Stanovich, 1994). Further, additional studies (Bradley & Bryant, 1983;
Bruck, 1992, 1993; Fawcett & Nicholson, 1995; Fletcher, Shaywitz,
Shankweiler, Katz, Lieberman, Stuebing, Francis, Fowler, & Shaywitz, 1994;
Fox & Routh, 1980; Lindamood, Bell, & Lindamood, 1992; Lyon, 1995) have
demonstrated the importance of PA in the assessment and remedial instruction of
disabled readers. And finally, PA has been a focus of substantial inquiry in
government reports (National Reading Panel, 2000) and reviews (Ehri, Nunes,
Willows, Schuster, Yaghoub-Zadeh, & Shanahan, T. 2001). As such, for
clinics and reading centers in which reading abilities and disabilities are
assessed, the understanding of the impact of PA on basic reading should be a
critical element of any reading assessment protocol. Accordingly, the purpose
of this paper is to report two investigations conducted at a
Defining Phonological Awareness
Phonological awareness (PA) has been defined by literacy authorities (Adams, 1990; Snow, et al., 1998; Stanovich, 1994; Torgesen, 1996) as the ability to work explicitly with sound elements smaller than the syllable. PA involves the conscious ability to manipulate individual speech sounds within words (Lundberg, Frost, & Petersen, 1988). More recently, ability in PA has been associated with six specific linguistic tasks that have become the target of assessment and instruction (Ehri, et al., 2001; National Reading Panel, 2000). These include the following:
1. Phonemic isolation – ability to recognize individual sounds in words. Example: “Tell me the first sound in ginger.” (/j/)
2. Phoneme identity – ability to recognize the common sound in different words.
Example: “Tell me the sound that is the same in peach, pile, and pickle? (/p/)
3. Phoneme categorization – ability to recognize the word with an odd sound among a sequence of three to four words.
Example: “Which word does not belong? bug, rug, bun, bus.” (rug)
4. Phoneme blending - ability to listen to individual sounds and to combine them into a recognizable word.
Example: “What word is /k/ /æ/ /t/?” (cat)
5. Phoneme segmentation – ability to break a word into it sounds by tapping or counting the sounds.
Example: “How many phonemes are in rain? (3: /r/ /ā/ /n/)
6. Phoneme deletion – ability to recognize what word remains when a specified phoneme is removed. Phoneme deletion is also refered to as elision.
Example: “What is stall without the /s/?” (tall)
Two of these tasks, phoneme blending and phoneme deletion are assessed as part of the phonological awareness cluster on the Comprehensive Test of Phonological Awareness (CTOPP) (Wagner, Torgesen, & Rashotte, 1999) and were the focus of the present studies.
The first study, Opatrny (1999), investigated the correlation of two subtests from the CTOPP (Wagner, et al., 1999) with two subtests from the Woodcock Diagnostic Reading Battery (WDRB) (Woodcock, 1997). The subjects for the first investigation were fifty-three children enrolled in a university reading center intervention program.
In the second investigation (Opatrny, French, & Cochran, 2000), a regression analysis was conducted to ascertain the predictability of significant relationships of the six subtests from the CTOPP with the Total Reading cluster from the Woodcock Diagnostic Reading Battery. In this second investigation, thirty existing diagnostic case records from an existing data base of client records were reviewed.
These studies are justified for the following three reasons. First, just as teachers conduct action studies to evaluate their teaching methods, clinical diagnosticians must also assess their practice. One way in which this can be done is by studying the tests used in assessment. The second reason for the studies stems from the technical development of the CTOPP. The authors completed several studies in which the CTOPP was correlated to formal reading tests. One study compared the CTOPP to the Woodcock Test of Reading Mastery. However, no previous test had been conducted to correlate the CTOPP to the Woodcock-Johnson Achievement Test or to the Woodcock Diagnostic Reading Battery, which contains the reading subtests of the WJAT. Finally, the studies were justified in that they provided the clinical staff of the participating university reading center with information needed to provide better reading interpretations and diagnoses.
Defining basic reading. For the purpose of these investigations, the definition of basic reading was taken from the Basic Reading Skills Cluster of the Woodcock Diagnostic Reading Battery (Woodcock, 1997). Specifically, this includes letter-word identification and word attack. On the Woodcock Diagnostic Reading Battery, word identification involves the recognition of sight words, such as get, was, part, presented in isolation. Word attack involves the pronunciation of nonsense words (letter combination or pseudo words) such as tiff, zoop, and dright.
Elision 7.58 2.73
Blending Words 10.58 2.67
Phonological Awareness 91.00 11.83
Letter-Word Identification 82.60 14.91
Word Attack 81.30 15.00
* Subtest average means on the CTOPP = 10.0 (+/- 3). Cluster means on the CTOPP as well as the subtest and cluster scores on the WDRB = 100 (+/- 15).
As shown in Table 1, the children assessed in the first study displayed low average abilities in phonological awareness and below average abilities in basic reading. This is consistent with the fact that the students were enrolled in a program to enhance their basic reading skills.
Next, correlations between subtests were calculated. These results are presented in Table 2.
** p < .05.
As shown in Table 2, significant correlations were seen among all tests and clusters. It was anticipated that correlations between CTOPP and WDRB tests would be significant. In further reviewing the data, it was found that elision correlated most strongly with word attack, whereas sound blending correlated most strongly with letter-word identification.
Discussion. The results are consistent with the concurrent correlation study reported in the CTOPP examiner’s manual (Wagner, et al., 1999). However, in that study, the CTOPP was correlated with the Woodcock Reading Mastery Tests-R. As in the present study, the correlations were significant, elision correlated to word attack (.74, p < .01) and to word identification (.53, p < .001). Blending words correlated significantly to word attack (.32, p < .01) but not to word identification.
In developing assessment protocols for intervention programs, it is important to utilize tests that are valid and reliable in the results they generate. In the first study, consistent with the studies previously cited, phonemic awareness was shown to be a significant factor in basic reading development. Thus, the supposition that PA should be assessed in intervention programs was confirmed, as it would appear the subtests from the CTOPP used in this study proved to be not only valid, but also a reliable indicator of word-level reading skills in these reading center children.
The second study (Opatrny, et al., 2000), in which existing diagnostic records were reviewed, sought to extend the first. Specifically, the objective of this second study was to use regression analysis to ascertain which CTOPP subtests most impacted the total reading score on the WDRB. The WDRB uses four core subtests to calculate the Total Reading Score. These are Letter-Word Identification, Word Attack, Vocabulary, and Reading Comprehension. The core battery of the CTOPP consists of six subtests. In addition to the phonological awareness subtests of elision and word blending, the assessment includes two tests of rapid naming (digits and letters) as well as two phonological memory subtests (memory for digits and non-word repetition).
In order to select profiles for inclusion in the review, sixty case files were selected at random from diagnostic cases completed in 1999 and 2000. The first criterion for inclusion in the analysis was the age of the children assessed matched the children’s ages included in Study #1. The second criterion for inclusion was that the client was tested with both the CTOPP and the WDRB. Those in which one or the other was not administered were excluded. The first thirty files that met these criteria were selected.
Means and standard deviations were calculated as in the first study. Then, using the SPSS computer application, a step-wise regression analysis was completed using the WDRB total reading score as the dependent variable and the CTOPP subtests as the independent variables.
Subtest/Cluster Mean* Standard
Elision 8.17 2.81
Blending Words 7.90 2.41
Memory for Digits 8.97 3.07
Rapid Digit Naming 8.16 2.32
Non-Word Repetition 8.80 1.93
Rapid Letter Naming 8.16 2.15
Letter-Word Identification 93.10 12.93
Word Attack 92.10 14.67
Reading Vocabulary 95.60 13.45
Reading Comprehension 97.63 13.36
Total Reading 94.33 12.54
* Subtest expected means on the CTOPP = 10.0 (+/- 3).
* Subtest expected means on the WDRB = 100 (+/- 15).
A review of the scores presented in Table 3 demonstrate that the children assessed at the reading center generally do score in the low average range on both the CTOPP and the WRDB. This is generally expected of children being assessed for reading disabilities as previously cited.
Since the first previous study (Opatrny, 1999) had demonstrated strong correlations between the CTOPP and WDRB, a regression analysis was completed to further ascertain which subtests of the CTOPP might predict the total reading score on the WDRB. A significant finding in this regard would help to focus the intervention of these students as well as allow for fewer subtests to be administered. In fact, the regression ANOVA was significant for two predictors—Elision (F=7.506, p = 0.011) and Blending Words (F=6.539, p = 0.005).
These results of the regression analysis are presented in Table 4, follows:
Dependent Variable: WDRB Total
As shown in Tables 4, the Elision and Blending Words subtests from the CTOPP were shown to be significant predictors for Total Reading. Of the two, the Elision subtest appears to be the subtest having the greater prediction value on the total reading variable.
1. Elision is a key element of phonological awareness.
Of the elements evaluated by the CTOPP measuring phonological awareness, this one element has been consistently powerful in identifying children who have reading difficulties.
2. Elision predicts success in reading nonsense words more than real words.
Since elision deals with manipulation of phonemes, it would make sense that children who can perform this task would do better at reading abstract pseudowords. Further study is needed to ascertain the range of influence that elision has over other reading skills and behaviors.
3. Elision may be an important element in the screening of reading success, albeit at the level of basic reading and broad measures of total reading.
Teachers continually ask how they can quickly assess young readers who may be at difficulty for reading failure. It would appear that using elision tasks may be one answer to this question.
In summary, although further research will be conducted to investigate these findings, it would appear, at least for now, that the use of elision as a key element in reading assessment and diagnosis is a sound practice.
Bradley, L. & Bryant, P. (1983). Categorizing sounds and learning to read: A causal connection. Nature, 301, 419-421.
Bruck, M. (1993). Component spelling skills of college students with childhood diagnoses of dyslexia. Learning Disability Quarterly, 16, 171-184.
Ehri, L. (1991). Development of the
ability to read words. In
Ehri, L. (1994). Development of the
ability to read words: Update. In
Ehri, L. C., Nunes, S. R., Willows, D. M., Schuster, B. V., Yaghoub-Zadeh, & Shanahan, T. (2001). Phonemic awareness instruction helps children learn to read: Evidence from the National Reading Panel’s meta-analysis. Reading Research Quarterly, 36 (3), 250-287.
Fawcett, A. & Nicholson, R.
(1995). Persistence of phonological awareness deficits in older children with
Fletcher, J. M., Shaywitz, S. E., Shankweiler, D. P., Katz, L., Lieberman, I. Y., Stuebing, K. K., Francis, D. J., Fowler, A. E., & Shaywitz, B. A. (1994). Cognitive profiles in reading disability: Comparisons of discrepancy and low achievement definitions. Journal of Educational Psychology, 86, 6-23.
Fox, B. & Routh, D. K. (1980). Phonemic analysis and severe reading disability in children. Journal of Psycholinguistic Research, 9, 115-119.
Lindamood, P. C.,
Morais, J., Bertelsen, P., Cary, L., & Alegria, J. (1987). Literacy
training and speech segmentation. In
Opatrny, C. (1999). Phonological
processing as it relates to linguistic dysfunctions. Unpublished senior
Opatrny, C., French, M. P., &
Cochran, L. L. (2000). Measuring the impact of phonological awareness on
basic reading skills. Paper presented at the American Reading Forum Annual
Share, D., Jorm, A., MacLean, R., & Matthews, R. (1984). Sources of individual differences in reading acquisition. Journal of Educational Psychology, 76, 1309-1324.
Snow, C., Burns, M., &
Wagner, R. K., Torgesen, J. K., &
Rashotte, C. A. (1999). The Comprehensive Test of Phonological Processing
Woodcock, R. W. (1997). Woodcock
Diagnostic Reading Battery Examiner’s Manual.