See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/273039515

The Relationship Between Phonological Awareness and Reading

Article in Language Speech and Hearing Services in Schools · October 2005

DOI: 10.1044/0161-1461(2005/029)

CITATIONS READS
186 1,576

3 authors:

Tiffany P Hogan Hugh W Catts

MGH Institute for Health Professions Florida State University
114 PUBLICATIONS 4,708 CITATIONS 107 PUBLICATIONS 12,889 CITATIONS

SEE PROFILE SEE PROFILE

Todd D. Little
Texas Tech University
341 PUBLICATIONS 35,479 CITATIONS

SEE PROFILE

All content following this page was uploaded by Todd D. Little on 27 April 2015.

The user has requested enhancement of the downloaded file.

 LSHSS

Clinical Forum

The Relationship Between

Phonological Awareness and Reading:
Implications for the Assessment of
Phonological Awareness
Tiffany P. Hogan
Hugh W. Catts
Todd D. Little
University of Kansas, Lawrence

S peech-language pathologists (SLPs) are actively

involved in the development of literacy skills
and in the remediation of literacy problems
(American Speech-Language-Hearing Association [ASHA],
2001). In particular, SLPs have an important role in the
The assessment of phonological awareness during
preschool and kindergarten provides critical insight into the
skills that children use to learn to read (Adams, 1990).
Lonigan, Burgess, and Anthony (2000) demonstrated that
phonological awareness, when compared to many other
assessment of phonological awareness due to their knowl- predictors, was the most stable and robust indicator of later
edge of phonetics and phonological disorders (Catts, 1991). reading in a group of children who were followed from late
Phonological awareness has been shown to be a primary preschool into kindergarten and first grade. In another data
factor underlying early reading achievement (Ehri, et al., set, Catts, Fey, Zhang, and Tomblin (2001) found that a
2001). Additionally, deficits in phonological awareness have kindergarten measure of phonological awareness was one of
been linked to reading disabilities (Lyon, Shaywitz, & five factors that predicted the presence of a reading
Shaywitz, 2003). disability in second grade. Numerous other studies have

ABSTRACT: Purpose: Speech-language pathologists (SLPs) information to the prediction of 4th-grade reading. Addi-
use phonological awareness assessments in many ways. tionally, a reciprocal relationship was found between
This study examines the usefulness of these assessments in phonological awareness and word reading, with kindergar-
kindergarten and 2nd grade. ten phonological awareness predicting 2nd-grade word
Method: Measures of phonological awareness and letter reading and, conversely, 2nd-grade word reading predicting
identification were administered in kindergarten, and 4th-grade phonological awareness.
measures of phonological awareness, phonetic decoding Clinical Implications: Phonological awareness assessment
(i.e., nonword reading), and word reading were adminis- provides information about reading in kindergarten but
tered in 2nd and 4th grades to a sample of 570 children loses its predictive power at 2nd grade. At that time,
participating in a longitudinal study of reading and phonological awareness and word reading become so
language impairments. highly correlated that phonological awareness does not add
Results: A path analysis indicated that kindergarten information to the prediction of 4th-grade reading.
measures of phonological awareness and letter identification
provided information to the prediction of 2nd-grade KEY WORDS: phonological awareness, assessment,
reading. In 2nd grade, measures of reading offered reciprocal relationship, prediction, early reading

LANGUAGE, SPEECH, AND HEARING SERVICES IN SCHOOLS • Vol. 36 • 285–293 • Hogan et al.:
October 2005 Assessment
© American of Phonological Awareness
Speech-Language-Hearing Association 285
0161–1461/05/3604–0285
documented the robust relationship between early phono- predict word reading in second grade beyond a measure of
logical awareness and subsequent reading achievement letter identification. Because most kindergarten children
(Calfee, Lindamood, & Lindamood, 1973; Lonigan, et al., cannot decode words, a measure of letter identification was
2000; Torgesen, Wagner, & Rashotte, 1994; Wagner et al., used in this grade as an indication of literacy experience.
1997). Indeed, letter identification has been found to be highly
Once children begin reading, however, the best indicator predictive of later word reading (see Scarborough, 1998,
of current and future reading may simply be reading itself for a review). We hypothesized that both letter identifica-
(Bell, McCallum, & Cox, 2003). This possibility has led tion and phonological awareness would be significant
reading researchers to question the usefulness of phonologi- predictors of second-grade word recognition.
cal awareness assessments once a certain level of reading Second, we sought to determine if phonological aware-
achievement has been attained. Wagner and his colleagues ness, measured in second grade, would predict word
(Wagner et al., 1997) considered this issue using a large, reading in fourth grade beyond a measure of second-grade
longitudinal data set. They examined the amount of word reading. We predicted that second-grade phonological
information that a measure of phonological awareness could awareness would provide very little or no significant
add to the prediction of reading once a measure of current information toward the prediction of fourth-grade word
word reading and vocabulary was considered. Results reading once second-grade word reading was known.
indicated that from kindergarten to second grade, phono- Related to this question, we also determined if a measure
logical awareness predicted 23% unique variance in later of second-grade nonword reading (i.e., phonetic decoding)
word reading; from first to third, 8%; and from second to would predict fourth-grade word reading beyond a measure
fourth, only 4%. The authors concluded that phonological of second-grade word reading. Similar to a measure of
awareness measures in the primary grades offered a small phonological awareness in kindergarten, phonetic decoding
but statistically significant amount of information to the provides insight into the skills that children use to read
prediction of future word reading beyond that provided by words (Adams, 1990; Bell et al., 2003). We predicted that a
a measure of current word reading. However, in a later second-grade measure of phonetic decoding would predict
review of this work, Torgesen (1999) concluded that the fourth-grade word reading beyond a measure of second-
limited amount of information gained from the assessment grade word reading.
of phonological awareness beyond second grade may not Finally, this study extends the work of Wagner et al.
warrant the use of a phonological awareness assessment (1997) and others in several ways. First, we used a large,
given the amount of time needed to administer, score, and well-selected sample of children. Data from such a study
interpret such an assessment. add to the generalizability of findings to the population at
The reduction in the amount of information offered by large. Second, in our study, we evaluated the unique
phonological awareness assessments once reading is variance associated with phonetic decoding and phonologi-
underway may be explained, at least in part, by the cal awareness in word reading. Previous studies have
reciprocal relationship between phonological awareness and combined phonetic decoding with other word reading skills
reading. Initially, phonological awareness influences and have not allowed for the comparison of the unique
reading; but, once reading is underway, the process of contribution of phonetic decoding and phonological
learning to read influences phonological awareness. In awareness to word reading.
support of the reciprocity between reading and phonological
awareness, research has shown that reading instruction with
an emphasis on decoding printed words highlights the
sound structure of language and facilitates children’s METHOD
performance on tests of phonological awareness (Lundberg
& Hoien, 1991; McGuinness, McGuinness, & Donohue, Participants
1995; Perfetti, Beck, Bell, & Hughes, 1987). Because of
The participants in this investigation were a subsample
this relationship, phonological awareness may become so
of children who had taken part in an epidemiologic study
highly correlated with word reading that it may offer little
of language impairments in kindergarten children (Tomblin,
unique information to the prediction of reading once a
1995). The epidemiologic study used a stratified cluster
measure of reading is available. At such time, tests of word
sample of more than 7,000 children, stratified by residential
reading may provide a majority of the information when
setting (i.e., rural, urban, suburban) and cluster sampled by
predicting future reading, leaving no information to be
school building. Out of this sample, 328 children with
accounted for by phonological awareness.
language impairment and/or nonverbal impairments in
kindergarten consented to participate in a follow-up
longitudinal investigation of language and reading develop-
ment (Tomblin, 1995).1 Additionally, a random sample of
STUDY QUESTIONS
1
The present study investigated the usefulness of phono- Of the 328 children, 123 children had language impairment only (i.e.,
specific language impairment), 103 children evidenced nonverbal impair-
logical awareness assessments in the prediction of reading
ments only, and 102 children showed language and nonverbal impairments
in the early school grades. First, we sought to determine if (i.e., nonspecific language impairment). For a detailed account of criteria for
phonological awareness, measured in kindergarten, would classification of these impairments, see Tomblin et al. (1997).

286 L ANGUAGE, SPEECH, AND HEARING SERVICES IN SCHOOLS • Vol. 36 • 285–293 • October 2005
those children without language impairments was recruited. deletion of the first sound in a one-syllable word (e.g.,
The final longitudinal sample included 604 children (328 “Say fat without the /f/”). In second and fourth grades,
with language impairment; 276 unimpaired). All of the nine items were added to increase the task difficulty to a
participants, regardless of language or nonverbal abilities, grade-appropriate level. Four of these new items required
were monolingual English speakers with normal hearing deletion of the final sound in a one-syllable word (e.g.,
and no history of significant emotional or neurological “Say find without the /d/”), and five required deletion of
disorders. Furthermore, no child had been diagnosed with a middle sound from a one-syllable word (e.g., “Say
autism or mental retardation at the beginning of the wives without the /v/”). In each grade, the task was
longitudinal study. Over the course of the longitudinal discontinued after six consecutive errors. To quantify each
study, 34 children left the study, leaving 570 children with participant’s performance on the phonological awareness
complete data sets through fourth grade. task in kindergarten, raw scores were converted to z
These 570 children comprised the sample for the present scores based on the mean and standard deviation from the
study.2 Due to the participant selection procedure previously original study sample (N = 604). This procedure was also
described, the sample contained higher percentages of used in second and fourth grade (N = 570). The kinder-
children with language and nonverbal impairments than garten version of this task may be found in its entirety in
those found in the original epidemiologic study. Therefore, Catts et al. (2001).
we employed a weighting procedure, described in the Letter identification. Because relatively few kindergarten
analysis section below, to ensure that our results were children can decode nonwords (Wagner et al., 1997), the
representative of the original epidemiologic sample. Letter Identification subtest of the Woodcock Reading
Mastery Tests—Revised (WRMT–R; Woodcock, 1987) was
used in kindergarten as an early estimate of alphabetic
Materials knowledge and literacy experience. In this task, the
In kindergarten, participants were administered tests of participants were asked to name upper and lower case
phonological awareness and letter identification, and in letters printed in various fonts. Standard scores were
second and fourth grades, participants were administered assigned using the grade-based assessment norms from the
tests of phonological awareness, phonetic decoding, and test manual because letter-name knowledge is largely
word reading. Table 1 provides a summary of the assess- dependent on instruction (Adams, 1990).
ments described below and the grades at which these Word reading. To assess word reading in second and
assessments were administered. fourth grades, the Word Identification subtest of the
Phonological awareness. The phonological awareness WRMT–R was administered to each of the participants. In
task was a measure of syllable/phoneme deletion (Catts et this task, the participants orally read real words, decreasing
al., 2001) that was adapted from Rosner’s Auditory in frequency of occurrence from highly frequent words
Analysis Test (Rosner & Simon, 1971). In this task, such as “go” to increasingly less frequent words such as
participants are asked to delete a syllable or phoneme from “quench.” Again, because of reliance on instruction,
a word and say the remaining sound sequence. In kinder- standard scores were assigned using the grade-based
garten, the task consisted of 3 practice items and 21 test assessment norms from the test manual.
items. Thirteen of the items required deletion of the initial Phonetic decoding. In second and fourth grades, the
syllable in either a compound word (e.g., “Say baseball Word Attack subtest of the WRMT–R was administered to
without the ‘base’”) or a two-syllable word (e.g., “Say measure phonetic decoding. This task required participants
baby without the ‘ba’”). The remaining eight items required to orally decode nonwords increasing in length and
complexity. The first and least complex item on the subtest
requires the child to read the nonword “ree.” An example
2
Although our sample included missing data for 34 children due to attrition of a more complex item is “untroikest.” Standard scores
from kindergarten to second grade, a multiple EM imputation procedure was
also employed as a secondary analysis to estimate these missing data. The
were assigned using the grade-based assessment norms
results of the study were unchanged when using the data set containing the from the test manual because instruction plays a major part
full sample of 604 children. in learning to phonetically decode words (Adams, 1990).

Table 1. Assessments used to measure phonological awareness, letter identification, phonetic

decoding, and word reading.

Grade Construct Assessment

K, 2nd, 4th Phonological awareness Catts Deletion Task (Catts et al., 2001)
K Letter identification Letter Identification subtest of the Woodcock Reading
Mastery Tests—Revised (WRMT–R; Woodcock, 1987)
2nd, 4th Phonetic decoding Word Attack subtest of the WRMT–R
2nd, 4th Word reading Word Identification subtest of the WRMT–R

Note. K = kindergarten; 2nd = second grade; 4th = fourth grade.

Hogan et al.: Assessment of Phonological Awareness 287

Procedures constant that was equal to the expected prevalence of these
children (3.5%) divided by their actual prevalence in our
Test administration. Testing was conducted by trained sample (7.7%; constant = .454). A similar procedure was
examiners with undergraduate or graduate degrees in used to weight the scores of other participants based on
speech-language sciences/pathology or education. The their specific characteristics. (For further details concerning
battery of tests was completed during two 2-hr sessions at the weighting procedure and evidence of its effectiveness,
each grade level—kindergarten, second grade, and fourth see Catts, Fey, Zhang, & Tomblin, 1999; Tomblin, Zhang,
grade. Buckwalter, & O’Brien, 2003.)
Weighting of scores. Table 2 shows the distribution of
weighted scores for the measures in our analysis (N = 570).
The phonological awareness tasks at each grade are
presented as raw scores for ease of interpretation; the letter RESULTS
identification, word reading, and phonetic decoding tasks
Path Analysis
are represented by standard scores, with a mean of 100 and
a standard deviation of 15.3 As noted above, the sample of Path analysis was used to analyze the data because of
children used in this study had a higher prevalence of its ability to examine complex relationships between
children with language impairments than the general multiple measures (Pedhazur, 1997). Path analysis is similar
population. To improve the representativeness of our data, to regression analysis with one main exception. In path
we used weighted scores that took into consideration analysis, an estimate of measurement error for each
prevalence rates for language impairments and other measure can be included by using an estimate of the
characteristics in the general population; these data were reliability of that measure. This error estimate allows for a
taken from the original epidemiologic study (discussed in more robust test of the relationships between measures
detail in Tomblin et al., 1997). Based on these data, each when comparing analyses that assume no error in the
participant’s scores were weighted according to the likeli- measurements. Similar to regression analysis, path analysis
hood that a participant with his or her gender, language, determines the amount of unique variance that one measure
and nonverbal IQ profile would have been part of the accounts for in another. Whereas in regression, this unique
representative sample seen in the epidemiologic study. For variance is represented by a partial correlation, in path
example, the epidemiologic study estimated that boys with analysis, this unique variance is represented by a path
a language impairment and low nonverbal IQ composed coefficient. Using path analysis, models of both direct and
3.5% of the general population. In our sample (N = 570), indirect influence are constructed to represent hypothesized
however, these children composed 7.7%. To ensure that relationships among measures. Once a model shows a good
participants from this group did not contribute dispropor- fit to the data, based on the chi-square fit statistic, various
tionately to our results, their scores were adjusted by a relationships within the model may be examined.
Our proposed model, shown in Figure 1, involved
measurement at three time points. In kindergarten, measures
3
Table 2 shows that our sample (after weighting scores) performed above the of phonological awareness and letter identification were
normative mean (i.e., 100) on the Letter Identification subtest in kindergarten
and the Word Identification subtest in second grade. However, the sample
entered into the model. In second and fourth grade,
performed below the mean on the second- and fourth-grade measures of measures of phonological awareness, phonetic decoding,
phonetic decoding and on the fourth-grade Word Identification subtest. The and word reading were entered. Initially, all measurements
latter finding may be the result of our sample receiving reading instruction were proposed to be related to the measurements directly
that involved less emphasis on phonetics than that found in the WRMT–R
preceding them in time. Each specified relationship is
normative sample.
indicated by a line ending in an arrow, which represents
the direction of the relationship. For example, kindergarten
Table 2. Weighted descriptive statistics on all study variables
at kindergarten, second, and fourth grades. phonological awareness was proposed to account for
second-grade phonological awareness, phonetic decoding,
and word reading. As shown in Figure 1, three lines
M SD Max Min originate from kindergarten phonological awareness
predicting second-grade phonological awareness, phonetic
decoding, and word reading, respectively. Double arrowed
Kindergarten
lines represent the covariance associated with measurements
Phonological awareness 8.51 6.28 21.00 0.00
Letter identification 103.55 13.93 145.00 43.00 co-occurring in time (e.g., phonological awareness and
letter identification in kindergarten).
Second grade For each specified relationship, a path coefficient is
Phonological awareness 21.16 5.29 30.00 0.00 obtained and examined for significance using a z test. This
Phonetic decoding 94.26 16.79 129.00 44.00
Word reading 103.90 19.08 149.00 32.00 statistic was used to determine if the path coefficient was
significantly different from 0. If the path coefficient was
Fourth grade not significantly different from 0, then the path was
Phonological awareness 24.11 3.50 30.00 0.00 removed from the model. In Figure 1, only one path was
Phonetic decoding 93.79 16.22 133.00 28.00
removed because it was not statistically significant. That
Word reading 96.97 15.46 130.00 32.00
path is represented by a dotted line. Paths may also be

288 L ANGUAGE, SPEECH, AND HEARING SERVICES IN SCHOOLS • Vol. 36 • 285–293 • October 2005
 Figure 1. Path analysis of sample (N = 570).

Kindergarten Second Grade Fourth Grade

Word .77 Word

Reading Reading
.34
.21
.63
.88
.44 .15
.65
Letter .31 Phonetic Phonetic .28
.47
Identification Decoding Decoding

.19 .15

.69 .43
.61
.37 ns
.39 .09

Phonological .49 Phonological .42 Phonological

Awareness Awareness Awareness

Note. ns = not statistically significant.

added to a model if they are found to be significant. In our freedom (p = 0.17). This statistic indicated that the data did
model, no paths were added beyond the ones initially not significantly deviate from the proposed model and that
specified. an excellent to outstanding fit of the model to the data was
found (Joreskog & Sorbom, 2003).
Model Statistics
Study Questions
The path model was tested using the covariance matrix
associated with our measurements employing the LISREL The results for each study question will be described
8.54 (Joreskog & Sorbom, 2003) program with maximum using a simplification of the model shown in Figure 1 (see
likelihood estimation. The correlations are shown in Table Figure 2). First, we sought to determine if phonological
3, with the split-half reliability for each measure shown on awareness, measured in kindergarten, would predict word
the diagonal. These reliabilities served as an estimate of reading in second grade beyond a measure of letter
measurement error. Model fit was assessed using the identification. In line with our hypothesis, we found that a
minimum fit function chi-square statistic (Joreskog & kindergarten measure of phonological awareness accounted
Sorbom, 2003). The chi-square statistic is the most familiar for unique variance in second-grade word reading (β = .37;
and stringent model statistic. Our final model (i.e., Figure path 2) beyond that accounted for by letter identification
1) had a chi-square value of 10.34, with 7 degrees of (β = .44; path 1).

Table 3. Correlations and split-half reliabilities for the sample (N = 570).

Variable 1 2 3 4 5 6 7 8

1. Phonological awareness, K (.93)

2. Letter identification, K .48 (.94)
3. Phonological awareness, 2nd .58 .43 (.86)
4. Phonetic decoding, 2nd .54 .50 .71 (.91)
5. Word reading, 2nd .59 .62 .70 .88 (.97)
6. Phonological awareness, 4th .48 .38 .67 .63 .63 (.83)
7. Phonetic decoding, 4th .52 .46 .68 .84 .82 .69 (.89)
8. Word reading, 4th .55 .56 .67 .83 .90 .65 .87 (.91)

Note. Split-half reliabilities for each measure are in parentheses on the diagonal; all correlations are significant at p < .05.

Hogan et al.: Assessment of Phonological Awareness 289

 Figure 2. Path coefficients for kindergarten phonological awareness and letter identification and
second- and fourth-grade phonological awareness, phonetic decoding, and word reading extracted
from our path analysis shown in Figure 1.

Kindergarten Second Grade Fourth Grade

Word .77 Word

Reading 33 Reading
.44
.15

11 .37
ns
44
22
Letter Phonetic Phonetic
Identification Decoding Decoding
55

Phonological Phonological Phonological

Awareness Awareness Awareness

Second, we examined whether phonological awareness, reading. This finding would shed light on the limited
measured in second grade, would predict word reading in information offered by second-grade phonological aware-
fourth grade beyond a measure of second-grade word ness to the prediction of fourth-grade word reading by
reading. As predicted, a second-grade measure of phono- indicating that initially, phonological awareness influenced
logical awareness added no information (β = ns; path 5) to word reading and then, word reading influenced phonologi-
the prediction of fourth-grade word reading beyond that cal awareness. Figure 3 contains data pertinent to this
provided by the second-grade measure of word reading question. The results show that our hypothesis was con-
(β = .77; path 3). firmed. Kindergarten phonological awareness and second-
Next, we determined if a measure of second-grade grade word reading were more strongly correlated (β = .37;
phonetic decoding would predict fourth-grade word reading path 1) than kindergarten letter identification and second- AQ
beyond a measure of second-grade word reading. We grade word reading (β = .19; path 2; ∆χ2(1, n = 570) = 14.52,
predicted that a measure of phonetic decoding would p < .05). In contrast, second-grade word reading and
provide significant information to the prediction of fourth- fourth-grade phonological awareness were correlated (β =
grade word reading beyond that provided by a second-grade .21; path 3), whereas second-grade phonological awareness
measure and this prediction was validated. Second-grade and fourth-grade word reading were not significantly
phonetic decoding predicted a small but significant amount correlated (β = ns; path 4).
of variance in fourth-grade reading (β = .15; path 4) Finally, SLPs assess and treat children who have, or are
beyond second-grade word reading (β = .77; path 3). suspected to have, deficient speech, language, and/or
Considering that phonological awareness contributed reading skills. In this study, we examined our questions
significant information to the prediction of word reading using a sample of children with a wide range of skills,
from kindergarten to second grade but not from second to from high to low language functioning. We acknowledge
fourth, we examined the potential reciprocity between that the majority of the children in our sample will not
phonological awareness and word reading across these likely be evaluated by an SLP because the majority of our
grades to better understand our results. We hypothesized sample exhibited typical reading/language development. In
that kindergarten phonological awareness would be more an attempt to better approximate the children most likely to
strongly related to second-grade word reading than kinder- be seen by an SLP, we reexamined our study questions in
garten letter identification would be to second-grade two subsamples of below-average readers; one subsample
phonological awareness. We expected that the inverse included those who scored below the 40th percentile on the
would be shown from second to fourth grade; that is, Oral Reading Accuracy Index of the Gray Oral Reading
second-grade word reading would be more strongly related Tests—Third Edition (Wiederholt & Bryant, 1994), a
to fourth-grade phonological awareness than second-grade measure of word reading, and the other subsample included
phonological awareness would be to fourth-grade word a more impaired group who scored below the 25th

290 L ANGUAGE, SPEECH, AND HEARING SERVICES IN SCHOOLS • Vol. 36 • 285–293 • October 2005
 Figure 3. Path coefficients for kindergarten phonological awareness and letter identification and
second- and fourth-grade phonological awareness and word reading extracted from our path analysis
shown in Figure 1.

Kindergarten Second Grade Fourth Grade

Word Word
Reading Reading

.37 ns

33

Letter
Identification
11

44
22

.21
.19
Phonological Phonological Phonological
Awareness Awareness Awareness

percentile on the same measure. The results from these phonological awareness predicted word reading from
subsamples were essentially the same as those obtained kindergarten to second grade, whereas from second to
using our full sample. Only one difference was apparent in fourth grade, this relationship reversed; second-grade word
the analyses using the subsamples: Kindergarten phonologi- reading predicted fourth-grade phonological awareness. This
cal awareness and second-grade word reading were no finding was consistent with a reciprocal relationship
longer more strongly correlated than kindergarten letter between phonological awareness and word reading. These
identification and second-grade word reading. Overall, these results have several clinical implications.
findings indicate that our results are consistent with those First, our findings converge with a large body of research
of children at the lower end of the normal distribution (i.e., indicating that the measurement of phonological awareness in
those likely to be seen by an SLP); although these findings kindergarten adds useful information to the prediction of
do not directly determine if there is a level of word word reading (Ehri et al., 2001). This information is beyond
reading at which phonological awareness may still contrib- that which can be gained from other strong kindergarten
ute unique variance to its prediction. literacy predictors such as letter identification. Therefore,
measures of phonological awareness should be included
when assessing kindergarten children to determine future
reading outcomes and/or risk for reading disability. SLPs
DISCUSSION have the skills needed to assess and interpret measures of
phonological awareness in kindergarten and should play a
This study investigated the usefulness of phonological significant role in this process.
awareness assessments in the prediction of word reading Second, our results indicate that beyond kindergarten (at
during the early school grades. We found that a measure of least by second grade), a measure of phonological aware-
phonological awareness in kindergarten predicted second- ness may offer little unique information to the prediction of
grade word reading beyond a measure of letter identifica- word reading. We found that by second grade, the best
tion. This pattern was not the case from second to fourth predictor of word reading is word reading itself. Therefore,
grade, when a second-grade measure of phonological rather than use a measure of phonological awareness at this
awareness did not provide unique information to the time, a measure of word reading should be used to make
prediction of fourth-grade word reading beyond that predictions about future reading outcomes. Because we also
provided by second-grade measures of word reading and found that phonetic decoding provided unique information
phonetic decoding. In an attempt to understand the loss of beyond that obtained from word reading, a measure of this
unique information gained from phonological awareness in ability might also be included in assessments of reading
second grade, we examined the relationship between outcome. Such a measure provides useful information
phonological awareness and word reading. We found that concerning how children are using their orthographic

Hogan et al.: Assessment of Phonological Awareness 291

knowledge and phonological awareness to read novel ending sounds, or have trouble blending sounds together to
printed words. form a word because he or she lacks the necessary phono-
Although this study specifically addressed the use of logical awareness to do so. This type of information is
phonological awareness assessments when predicting word relevant to determining the underlying nature of a reading
reading, the results could be extended to concurrent disability and/or assessing treatment progress.
assessments directed at determining the underlying nature Even in light of our results and the above discussion, it
of a reading problem and/or assessing treatment progress. is conceivable that phonological awareness probes may still
Using path analysis, we were only directly able to address be helpful to determine more specific intervention goals
issues of prediction. However, our model provides some and assess treatment progress in second grade and beyond
suggestions concerning the possible nature of concurrent for some children. For example, when planning specific
relationships. As shown in Figure 1, concurrent measures of intervention goals, an SLP may suspect that a child has
phonetic decoding and word reading were more highly difficulty segmenting sounds in initial blends based on the
related than were those involving phonological awareness types of words that the child incorrectly decoded on a
and word reading (.88 vs. .65 in second grade and .63 vs. phonetic decoding assessment (e.g., “blue” was read as
.28 in fourth grade). Of course, these data do not speak to “bue”). Further in-depth exploration of the child’s ability to
the unique contribution of concurrent measures of phonetic segment initial blends using a phonological awareness
decoding versus phonological awareness to word reading. probe of this skill will likely aid in intervention planning.
However, in an earlier study using these same data, we Likewise, an SLP working with a child to improve his
employed hierarchical regression analyses to examine ability to blend printed words containing stop consonants
concurrent relationships (Catts & Hogan, 2002). This study (e.g., “/b/ /o/ /t/ goes together to make ‘boat’”) may find
showed that concurrent measures of phonetic decoding that a probe of this skill offers important additional
accounted for a considerable amount of the unique variance information about the effects of treatment beyond that
in word reading, whereas phonological awareness added provided by a test of word reading or phonetic decoding.
little or no unique variance at second and fourth grades. These uses seem appropriate as long as the relationship
Before proceeding, it should be noted that our results between reading and phonological awareness is considered
concerning phonological awareness assessment may be and the phonological awareness assessment (or probe) is
dependent on the way in which we measured phonological not the primary assessment of reading outcomes for the
awareness. Recall that our measure was one involving reasons described above.
syllable/phoneme deletion. This measure was chosen Even though phonetic decoding assessments have
because of its close relationship to word reading ability typically been administered by reading specialists, it is not
(Torgesen et al., 1994). It is possible that if another outside of an SLP’s scope of practice to administer and
measure of phonological awareness was used (e.g., pho- interpret such an assessment (ASHA, 2001). Tests of
neme segmentation), the results could have differed. Further phonetic decoding measure children’s knowledge of English
research is necessary to address this issue. orthography as a phonetic transcription, and in some cases,
The results of our study suggest that at least by second specific nonwords additionally assess morphological
grade, measures of phonetic decoding may provide more knowledge (e.g., “gaked” and “mancingful,” from the
unique information about concurrent word reading than will WRMT–R Word Attack subtest). SLPs have phonetic
phonological awareness, as measured in this study. Again, transcription skills as well as knowledge of phonological
the reason for this finding may be the reciprocal relation- development. These skills and knowledge provide the
ship between phonological awareness and reading. This foundation for transcribing and analyzing decoding errors
relationship ensures that by second grade, measures of using information on sound contrasts, phonological pro-
phonetic decoding and phonological awareness tap some- cesses, and sound development. Reading specialists and
what similar skills and knowledge. That is, measures of classroom teachers greatly enhance their ability to under-
phonological awareness and phonetic decoding essentially stand decoding breakdowns through collaboration with
become overlapping assessments, each providing informa- SLPs. As such, SLPs should collaborate with reading
tion about orthographic and phonological knowledge and specialists and classroom teachers to enhance the under-
skills. However, because measures of phonetic decoding standing of word reading problems. This collaboration is
overlap more with word reading, such measures typically necessary to provide the most effective assessment and
will be a better choice for reading-related assessments than treatment for children with reading disabilities (Snow,
will measures of phonological awareness. A test of phonetic Scarborough, & Burns, 1999).
decoding provides information about how a child uses his
or her orthographic knowledge and phonological awareness
to decode novel words. For example, a measure of phonetic
decoding allows educators to determine if a child can ACKNOWLEDGMENTS
decode simple consonant-vowel-consonant words but has
This study was supported by grants from the National Institutes
trouble decoding more complex words—an ability directly of Health (P05-DC02726; T32-DC000052; P30-DC005803; P30-
related to early word reading. Additionally, such an HD02528). The authors would like to thank J. Bruce Tomblin,
assessment provides the opportunity to gain pertinent Marc Fey, Holly Storkel, Xuyang Zhang, Gail Gillon, Paula
information regarding the child’s phonological awareness. Buckwalter, Marlea O’Brien, Connie Ferguson, Jodi Schwartz, Amy
For example, the child may skip over a sound, leave off Kundel, Jacqueline Daniels, and Angela Muller for their valuable

292 L ANGUAGE, SPEECH, AND HEARING SERVICES IN SCHOOLS • Vol. 36 • 285–293 • October 2005
 contributions. We are particularly appreciative of the time and Pedhazur, E. J. (1997). Multiple regression in behavioral
effort provided by the parents and children involved in this study. research: Explanation and prediction (3rd ed.). Fort Worth, TX:
Harcourt.
Perfetti, C. A., Beck, I., Bell, L. C., & Hughes, C. (1987).
Phonemic knowledge and learning to read are reciprocal: A
REFERENCES longitudinal study of first grade children. Merrill-Palmer
Quarterly, 33, 283–319.
Adams, M. J. (1990). Beginning to read: Thinking and learning
about print. Cambridge, MA: MIT Press. Rosner, J., & Simon, D. (1971). The Auditory Analysis Test: An
initial report. Journal of Learning Disabilities, 4, 40–48.
American Speech-Language-Hearing Association. (2001). Roles
and responsibilities of speech-language pathologists with respect Scarborough, H. S. (1998). Early identification of children at risk
to reading and writing in children and adolescents. ASHA for reading disabilities: Phonological awareness and some other
Supplement, 21, 17–28. promising predictors. In B. K. Shapiro, P. J. Accardo, & A. J.
Capute (Eds.), Specific reading disability: A view of the
Bell, S. M., McCallum, R. S., & Cox, E. A. (2003). Toward a
spectrum (pp. 75–119). Timonium, MD: York Press.
research-based assessment of dyslexia: Using cognitive measures
to identify reading disabilities. Journal of Learning Disabilities, Snow, C. E., Scarborough, H. S., & Burns, M. S. (1999). What
36, 505–516. speech-language pathologists need to know about early reading.
Topics in Language Disorders, 20, 48–58.
Calfee, R. C., Lindamood, P., & Lindamood, C. (1973).
Acoustic–phonetic skills and reading: Kindergarten through Tomblin, J. B. (1995). Midwest collaboration on specific language
twelfth grade. Journal of Educational Psychology, 64, 293–298. impairment. Washington, DC: National Institute of Deafness and
Catts, H. W. (1991). Facilitating phonological awareness: Role of Other Communication Disorders.
speech-language pathologists. Language, Speech, and Hearing Tomblin, J. B., Records, N., Buckwalter, P., Zhang, X., Smith,
Services in Schools, 22, 196–203. E., & O’Brien, M. (1997). Prevalence of specific language
Catts, H. W., Fey, M. E., Zhang, X., & Tomblin, J. B. (1999). impairment in kindergarten children. Journal of Speech,
Language basis of reading and reading disabilities: Evidence Language, and Hearing Research, 40, 1245–1260.
from a longitudinal investigation. Scientific Studies of Reading, Tomblin, J. B., Zhang, X., Buckwalter, P., & O’Brien, M.
3, 331–361. (2003). The stability of primary language disorder: Four years
Catts, H. W., Fey, M. E., Zhang, X., & Tomblin, J. B. (2001). after kindergarten diagnosis. Journal of Speech, Language, and
Estimating the risk of future reading difficulties in kindergarten Hearing Research, 46, 1283–1296.
children: A research-based model and its clinical implementation. Torgesen, J. K. (1999). Assessment and instruction for phonemic
Language, Speech, and Hearing Services in Schools, 32, 38–50. awareness and word recognition skills. In H. W. Catts & A. G.
Catts, H. W., & Hogan, T. P. (2002, November). At what grades Kamhi (Eds.), Language and reading disabilities (pp. 128–153).
should we assess phonological awareness? Poster presented at Needham Heights, MA: Allyn & Bacon.
the annual convention of the American Speech-Language- Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1994).
Hearing Association, Atlanta, GA.
Longitudinal studies of phonological processing and reading.
Ehri, L. C., Nunes, S. R., Willows, D. M., Schuster, B. V., Journal of Learning Disabilities, 27, 276–286.
Yaghoub-Zadeh, Z., & Shanahan, T. (2001). Phonemic
Wagner, R. K., Torgesen, J. K., Rashotte, C. A., Hecht, S. A.,
awareness instruction helps children learn to read: Evidence
Barker, T. A., Burgess, S. R., et al. (1997). Changing relations
from the National Reading Panel’s meta-analysis. Reading
between phonological processing abilities and word-level reading
Research Quarterly, 36, 250–287.
as children develop from beginning to skilled readers: A 5-year
Joreskog, K. G., & Sorbom, D. (2003). LISREL [Computer longitudinal study. Developmental Psychology, 33, 468–479.
software]. Lincolnwood, IL: SSI.
Wiederholt, J. L., & Bryant, B. R. (1994). Gray Oral Reading
Lonigan, C. J., Burgess, S. R., & Anthony, J. L. (2000). Tests—Third Edition. Austin, TX: Pro-Ed.
Development of emergent literacy and early reading skills in
preschool children: Evidence from a latent variable longitudinal Woodcock, R. (1987). Woodcock Reading Mastery Tests—Revised.
study. Developmental Psychology, 36, 596–613. Circle Pines, MN: American Guidance Service.
Lundberg, I., & Hoien, T. (1991). Initial enabling knowledge and
skills in reading acquisition: Print awareness and phonological
segmentation. In D. J. Sawyer & B. J. Fox (Eds.), Phonological Received May 6, 2004
awareness in reading: The evolution of current perspectives (pp. Accepted November 15, 2004
74–95). New York: Springer-Verlag. DOI: 10.1044/0161-1461(2005/029)
Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2003). A
definition of dyslexia. Annals of Dyslexia, 53, 1–14. Contact author: Tiffany P. Hogan, Department of Speech,
McGuinness, D., McGuinness, C., & Donohue, J. (1995). Language, Hearing, University of Kansas, 3001 Dole Human
Phonological training and the alphabet principle: Evidence for Development Center, 1000 Sunnyside Avenue, Lawrence, KS
reciprocal causality. Reading Research Quarterly, 30, 830–852. 66045-7555. E-mail: tehogan@ku.edu

Hogan et al.: Assessment of Phonological Awareness 293

View publication stats