Emergent Literacy Intervention For Vulnerable Pres
Emergent Literacy Intervention For Vulnerable Pres
Emergent Literacy Intervention For Vulnerable Pres
net/publication/10569379
CITATIONS READS
169 2,338
5 authors, including:
Some of the authors of this publication are also working on these related projects:
R01HD097189 Parent Self-Regulation, Parent-Child Coregulation, and Harsh Discipline View project
All content following this page was uploaded by Sy-Miin Chow on 31 December 2013.
Laura M. Justice
Sy-Miin Chow
Cara Capellini
Kevin Flanigan
Sarah Colton
University of Virginia, Charlottesville
This study determined the relative efficacy of knowledge over the entire 12-week intervention
an experimental explicit emergent literacy program; growth was significantly greater during
intervention program for preschoolers experienc- the experimental explicit intervention program
ing multiple risk factors. Using an alternating compared to the comparison program. An
treatment research design, children completed examination of individual differences and
two 6-week waves of intervention in small intervention outcome showed oral language
groups; one wave featured the experimental skills and literacy orientation to predict emergent
explicit intervention program, whereas the other literacy performance at the end of the program.
featured a comparison program. Emergent
literacy assessment was conducted at pretest Key Words: emergent literacy, language
and at the end of each wave. Results indicated disorders in children, intervention, at-risk,
significant widespread gains in emergent literacy preschool
E
mergent literacy refers to the foundation upon which describes children’s ability to implicitly and explicitly
children’s conventional reading and writing abilities represent spoken language as comprising discrete and
are built. More specifically, this term is used to recurrent sound elements (e.g., phonemes, syllables,
describe the behaviors, skills, and concepts of young words; Bradley & Bryant, 1985; Torgesen & Davis, 1996;
children that develop into and precede conventional Whitehurst & Lonigan, 1998; Wise, Ring, & Olson, 1999).
literacy (Kaderavek & Sulzby, 1998b; Sulzby, 1985). For Knowledge across both domains is acquired gradually and,
most children, the emergent literacy foundation is acquired for many children, incidentally during the course of early
within the period preceding formal literacy instruction, childhood.
transcending birth to about 6 years of age. Prediction studies following children from preschool or
Generally speaking, this foundation comprises two kindergarten into elementary school have consistently
distinct but highly interrelated areas of development: shown that performance on an array of emergent literacy
written language awareness and phonological awareness tasks reliably predicts children’s later literacy achievement
(Justice & Ezell, 2001b; van Kleeck, 1998). These domains (e.g., Badian, 1982; Catts, Fey, Zhang, & Tomblin, 1999,
refer, respectively, to children’s acquisition of knowledge 2001; Stuart, 1995; for review, see Scarborough, 1998).
about the orthography and the phonology of one’s own Children performing well on emergent literacy tasks
language. Written language awareness describes the generally have superior conventional literacy outcomes
implicit and explicit knowledge children acquire concern- relative to children demonstrating lower levels of perfor-
ing the fundamental properties of print, such as the mance. Prediction studies have also shown indices of both
relationship between print and speech and the functions written language and phonological awareness to contribute
and forms of particular written language units (e.g., letters, uniquely to conventional literacy outcomes (Badian, 1982,
words, punctuation marks; Hiebert, 1981; Justice & Ezell, 1986; Stuart, 1995; for review, see Scarborough, 1998).
2000, 2002; Lomax & McGee, 1987; Whitehurst & Taken together, the importance of young children’s written
Lonigan, 1998). Phonological awareness, in contrast, language and phonological awareness for later literacy
American Journal of Speech-Language Pathology • Vol. 12 • 320–332 • August 2003 • © American Speech-Language-Hearing Association
320 American Journal of Speech-Language Pathology • Vol. 12 • 320–332 • August 2003
1058-0360/03/1203-0320
achievement argues for the need for increased focus on in kindergarten, encompassing such diverse areas as
preschool emergent literacy intervention as a proactive vocabulary knowledge, grammatical understanding, and
model for preventing reading difficulties (Snow, Burns, & narrative comprehension (Catts et al., 1999). Preschool and
Griffin, 1998). Effective early intervention programs kindergarten children experiencing more severe and/or
should result in substantial, widespread growth across both widespread language difficulties appear to be at relatively
emergent literacy domains for those children most vulner- greater risk for poor literacy outcomes (Bishop & Adams,
able for emergent and conventional literacy difficulties. 1990). Findings such as these unequivocally argue the need
for increased emphasis on a preventive model of literacy
intervention for children with LI.
Children Vulnerable for Literacy
Difficulties
The majority of children, by virtue of being immersed Poverty
in a literate society, acquire emergent literacy concepts and As important as oral language proficiency is to emer-
skills relatively effortlessly during the course of early gent literacy development, the latter also appears to be
childhood. Nevertheless, some children appear particularly strongly mediated by the frequency with which children
vulnerable for experiencing difficulties achieving an are formally and informally exposed to language and
adequate emergent literacy foundation (for an overview of literacy in the home, school, and community (Frijters,
key risk factors, see Snow et al., 1998; also see Catts et al., Barron, & Brunello, 2000; Leseman & de Jong, 1998;
2001). Of relevance to the research reported here are two McCormick & Mason, 1986; Neuman, 1999; Ninio, 1980;
circumstances associated with a significant increase in risk Purcell-Gates, 1996). Children who seldom interact with
status for emergent literacy difficulties: oral language written language (e.g., through parent–child shared
impairment (LI) and poverty. storybook reading experiences) have more difficulty
acquiring emergent literacy knowledge compared to peers
with more frequent literacy opportunities (e.g., Raz &
Oral LI Bryant, 1990; Wells, 1985). Limited exposure to oral and
There is a reciprocal and robust association between written language is a circumstance encountered relatively
young children’s oral language proficiency and emergent often by young children reared in low-socioeconomic
literacy development (Boudreau & Hedberg, 1999; status (SES) households, and it is a situation that may
Chaney, 1992; Dickinson & Tabors, 1991; Lonigan, contribute to the relatively low levels of emergent literacy
Burgess, & Anthony, 2000). Prediction studies have shown skill observed in low-SES children (e.g., Feitelson &
that an array of discrete oral language proficiency indices, Goldstein, 1986; McCormick & Mason, 1986; Ninio, 1980;
including measures of vocabulary and grammar, consis- Teale, 1986; Whitehurst et al., 1994).
tently serve as moderate to robust predictors of conven- Indeed, converging evidence has provided considerable
tional literacy outcome (for review, see Scarborough, 1998, documentation that both emergent and conventional
2000). Moreover, the prediction strength becomes increas- literacy skills of children from low-SES households differ
ingly powerful when several measures of language are in comparison to those of their peers from middle- and
combined into a composite index of language proficiency upper-SES households (e.g., Bowey, 1995; Chaney, 1994;
(Lonigan et al., 2000; Scarborough, 1998). Dickinson & Snow, 1987; Fernandez-Fein & Baker, 1997;
It is thus not particularly surprising that young children Justice & Ezell, 2001b; Lonigan et al., 1999; Warren-
exhibiting significant difficulties with oral language Leubecker & Carter, 1988; White, 1982). Dickinson and
proficiency are at increased risk for delayed attainment of Snow (1987) contrasted the performance of young children
emergent literacy knowledge and also are relatively more from low- and middle-SES households on a series of
likely than their typically developing peers to experience written language awareness tasks, finding that middle-SES
poor conventional literacy outcomes (Bird, Bishop, & children performed significantly better than low-SES
Freeman, 1995; Bishop & Adams, 1990; Boudreau & children on measures of print production, book reading
Hedberg, 1999; Catts, 1993; Magnusson & Naucler, 1990, concepts, and environmental print decoding. Lonigan and
1993). Preschool children with LI consistently show colleagues (1999) more recently reported similar findings
depressed performance relative to their peers on an array of when comparing the emergent literacy performance of low-
emergent literacy tasks addressing both written language SES children in Head Start to that of children in a childcare
and phonological awareness (e.g., Boudreau & Hedberg, serving middle-SES families. Children in Head Start
1999; Gillam & Johnston, 1985; Magnusson & Naucler, demonstrated relatively low levels of skill on measures of
1993). Likewise, studies of school-age children with alphabet knowledge, letter-sound knowledge, book reading
reading difficulties have revealed a notably high frequency concepts, and environmental print decoding.
of earlier deficits in diverse areas of oral language. One These same patterns of comparatively poor performance
prospective study of second graders with reading problems when comparing low-SES children to their more advan-
found that many of these children as toddlers had exhibited taged peers have also been observed when looking at
significant difficulties with oral language development phonological awareness performance; children’s perfor-
(Scarborough, 1990). A more recent study found 57% of mance on measures of phonological awareness has
183 children characterized as poor readers in second grade consistently been shown to be influenced by SES (e.g.,
to have exhibited difficulties in receptive language ability Chaney, 1994; Fernandez-Fein & Baker, 1997; Nittrouer,
Variable M SD Range
Note. Speech production = percentile rank from the Goldman-Fristoe Test of Articulation (Goldman
& Fristoe, 1985). Receptive language and expressive language = standard scores from the Auditory
Comprehension and Expressive Communication subtests, respectively, of the Preschool Language
Scale–Third Edition (PLS-3; Zimmerman et al., 1992); total language = Total Language score from
the PLS-3. Normative references for the PLS-3 are M = 100, SD = 15.
impairment, 6 showed receptive and/or expressive LI scheduling block (allowing for only three 30-min sessions)
with speech production impairment, and 6 showed during the typical school day.
receptive and/or expressive LI without speech production Following group assignment, children then completed a
difficulty. 12-week emergent literacy intervention conducted in two
consecutive 6-week waves, with the order of waves
counterbalanced across the two groups. Children com-
General Procedures pleted the intervention waves in small groups of 6 children
A within-subjects alternating treatment research design each; Group A (6 children) was a single group, whereas
involving two intervention programs (experimental Group B (12 children) was randomly subdivided further
explicit, comparison) served as the framework for this into two groups of 6 children.
investigation. At the start of the study, children were Equivalency for the two groups was examined post hoc
administered an emergent literacy pretest and then were across the following variables: chronological age, speech
randomly assigned to one of two groups: Group A (n = 6) production (percentile rank from the GFTA), and language
or B (n = 12). The uneven group composition resulted from proficiency (Total Language score from the PLS-3).
scheduling and design constraints, in which the 18-child Determination of group equivalency was conducted via a
sample needed to receive two interventions in a 1.5 hr series of independent samples t tests with an alpha level of
1 (A) 103 104 68 Typically developing language and speech production skills
2 (A) 82 90 79 Typically developing language and speech production skills
3 (A) 87 97 63 Typically developing language and speech production skills
4 (A) 65 62 55 Receptive/expressive language impairment; typical speech production skills
5 (A) 94 81 19 Expressive language impairment; speech production impairment
6 (A) 67 68 3 Receptive/expressive language impairment; speech production impairment
7 (B) 71 84 19 Receptive language impairment; speech production impairment
8 (B) 80 62 3 Receptive/expressive language impairment; speech production impairment
9 (B) 65 64 60 Receptive/expressive language impairment; typical speech production skills
10 (B) 70 73 44 Receptive/expressive language impairment; typical speech production skills
11 (B) 73 64 NA Receptive/expressive language impairment; speech production impairment suggested
12 (B) 74 75 68 Receptive/expressive language impairment; typical speech production skills
13 (B) 66 81 67 Receptive/expressive language impairment; typical speech production skills
14 (B) 84 81 15 Expressive language impairment; speech production impairment
15 (B) 103 95 15 Typically developing language skills; speech production impairment
16 (B) 98 107 99 Typically developing language and speech production skills
17 (B) 98 105 44 Typically developing language and speech production skills
18 (B) 69 72 43 Receptive/expressive language impairment; typical speech production skills
Note. Receptive language and expressive language = standard scores from the Auditory Comprehension and Expressive Communication
subtests of the Preschool Language Scale–Third Edition (Zimmerman et al., 1992). Speech production = percentile rank from the Goldman-
Fristoe Test of Articulation (Goldman & Fristoe, 1985).
Table 3 depicts changes in children’s scores from pre- to collapsed across the entire 12-week period, without regard
posttest on the five emergent literacy measures. Visual to children’s performance levels at the start of each wave.
inspection of these data suggests growth over time for all This is a particularly important consideration in the present
areas examined. Statistical analyses supported this observa- study given the alternating treatment design, in which
tion, with the test statistic showing a significant main effect children’s experiences prior to a particular wave may have
for time based on Wilks’s criterion, F(5, 25) = 11.84, p = influenced their growth within that wave. For instance,
.000. In contrast, there was no significant finding for Group A children began the comparison program follow-
program, F(5, 25) = 0.708, p = .623, nor was there a Time × ing 6 weeks of the experimental explicit program, whereas
Program interaction, F(5, 25) = 2.153, p = .092. Group B children experienced the opposite schedule.
These findings indicate that children made substantial A second set of analyses, using a series of paired-
gains in collective consideration of the emergent literacy samples t tests, was thus conducted to determine the extent
measures over the 12-week intervention period. Univariate of change for the five dependent measures during each
tests for the main effect for time showed significant growth intervention wave. We took children’s preexisting differ-
for each of the five measures: alphabet knowledge, ences in performance levels at the start of each wave as a
F(1, 29) = 25.37, p = .000; print awareness, F(1, 29) = baseline level against which performance at the end of the
10.54, p = .003; name writing, F(1, 29) = 4.385, p = .045; intervention wave was compared. These analyses con-
phonological segmentation, F(1, 29) = 25.868, p = .000; trolled for variability in performance levels at the start of
and rhyme production, F(1, 29) = 12.454, p = .001. each wave within and across individual children. Table 4
depicts children’s scores on each dependent measure at the
beginning and end of the two 6-week intervention waves.
Effectiveness of the Experimental Results showed that children’s scores on each emergent
Explicit Approach to Intervention literacy measure significantly improved from the beginning
Data were also analyzed to compare the magnitude of to the end of the 6-week experimental explicit intervention
growth in emergent literacy for the two intervention waves, program: alphabet knowledge, t(14) = 7.025, p = .000;
contrasting the experimental explicit program and the print awareness, t(17) = 3.302, p = .004; name writing,
comparison program. The question of whether the experi- t(17) = 2.365, p = .03; phonological segmentation, t(16) =
mental explicit approach exerted an advantage was 4.735, p = .000; and rhyme production, t(16) = 3.469,
partially addressed by the multivariate test described p = .003. The same set of comparisons was done to
previously, in which program served as a between-subjects characterize the extent of growth during the 6-week
factor. At the multivariate level, the test statistic indicated comparison program. During this wave, children showed
no significant main effect for program. However, in this significant growth only for phonological segmentation,
particular analysis, children’s gains across each wave were t(13) = 2.662, p = .02. Pre- to postwave changes for the
Alphabet knowledge 7.3 8.5 9.8 8.9 6.7 7.5 12.0 9.1
Print awareness 6.4 3.4 7.1 3.1 5.1 3.1 7.5 3.5
Name writing 3.5 1.5 3.8 1.6 2.9 2.4 4.0 1.4
Phonological segmentation 5.1 3.2 8.3 3.5 4.4 3.2 8.8 4.8
Rhyme production 0.9 2.6 1.9 3.3 1.4 2.9 4.1 4.2