AJSLP
Article
The Association Between Expressive
Grammar Intervention and Social and
Emergent Literacy Outcomes for
Preschoolers With SLI
Karla N. Washingtona
Purpose: To determine whether (a) expressive grammar
intervention facilitated social and emergent literacy outcomes
better than no intervention and (b) expressive grammar gains
and /or initial expressive grammar level predicted social and
emergent literacy outcomes.
Method: This investigation was a follow-up to a recently
published study exploring the impact of grammatical language
intervention on expressive grammar outcomes for preschoolers
with specific language impairment (SLI). Twenty-two 3- to
5-year-old preschoolers received ten 20-minute intervention
sessions addressing primary deficits in grammatical morphology.
Participants’ social and emergent literacy skills were not
targeted. Twelve children awaiting intervention, chosen from
the same selection pool as intervention participants, served
as controls. Blind assessments of social and emergent literacy
outcomes were completed at preintervention, immediately
postintervention, and 3 months postintervention.
A
pproximately 7% of young children are identified with
specific language impairment (SLI; Tomblin et al.,
1997). These children experience an unexpected
failure to acquire language skills at a typical rate (Leonard,
1998). This delay in development occurs despite the presence
of typical nonverbal cognitive abilities; hearing, emotional,
or behavioral regulation; and neurological function needed
to support language acquisition (Bishop, 1997; Plante, 1998;
Stark & Tallal, 1981). It is well established that children with
SLI are limited in their grammatical skills development
(Leonard, Camarata, Pawlowska, Brown, & Camarata, 2006,
2008). However, other core areas of impairment for some,
but not all, children with SLI include social and literacy skills
a have shown that inter-relationships exist between language
University of Western Ontario, London, Ontario, Canada
Correspondence to Karla N. Washington, who is now with the
University of Cincinnati: washink2@ucmail.uc.edu
Editor: Carol Scheffner Hammer
Associate Editor: Teresa Ukrainetz
Received March 7, 2011
Revision received August 23, 2011
Accepted September 8, 2012
DOI: 10.1044/1058-0360(2012/11-0026)
American Journal of Speech-Language Pathology • Vol. 22 • 113–125 • February 2013 • A American Speech-Language-Hearing Association
Results: Only intervention participants experienced significant
gains in social and emergent literacy outcomes and main-
tained these gains for 3 months postintervention. Expressive
grammar gains was the only single significant predictor of these
outcomes.
Conclusions: Expressive grammar intervention was asso-
ciated with broad impacts on social and emergent literacy
outcomes that were maintained beyond the intervention period.
Gains in expressive grammar predicted these outcomes.
Social and emergent literacy skills were positively affected for
preschoolers with SLI during a grammatical language intervention
program.
Key Words: social skills, emergent literacy, specific
language impairment, expressive grammar intervention,
preschool outcomes
development (Cabell, Justice, Zucker, & McGinty, 2009;
Cabell et al., 2010; McCabe, 2005).
Because children with SLI typically receive grammatical
language intervention to address hallmark deficits in gram-
matical skills, researchers and practitioners in speech-language
pathology need to consider whether the intervention can also
simultaneously address other areas of development, such as
social and literacy skills, as part of a preventative program,
even for those preschoolers who do not evidence deficits in
these two areas (Washington, 2011; Washington & Warr-Leeper,
2006). In addition, we need to consider whether initial level
of language functioning for grammatical skills and /or gains
in grammatical skills achieved during intervention might
predict social and literacy outcomes. Earlier researchers
and social skills (Bishop, 1997; McCabe, 2005) and between
language and literacy skills (Cabell et al., 2009, 2010; Catts
et al., 2002). Thus, exploring the inter-rationships between
language skills and social and literacy skills is important to
identifying the nature of these relationships in preschoolers
with SLI.
Information on outcomes and predictors of outcomes is
critical to demonstrating the broad impact of grammatical
skills (e.g., initial expressive grammar level and expressive
113
grammar gains) on other co-occurring areas of development
in which children may demonstrate ongoing deficits in
the later school years. The existence of social and literacy
deficits in the later school years for older children with SLI
supports the investigation of these co-occurring areas of
development during the preschool years (Baker & Cantwell,
1987; Catts, 1993; Catts et al., 2002; Fujiki, Spackman,
Brinton, & Hall, 2004).
Social Skills in Preschoolers with SLI
A study based on parent and teacher reports of 3- to
5-year-old preschoolers’ social and behavioral competencies
found that some preschoolers with SLI experienced greater
social and behavioral difficulties compared with preschoolers
without language disorders (McCabe, 2005). Good social
skills are needed for the developing child because they
provide access to well-needed social interactions necessary
for continued language development, later academic success,
and the formation and maintenance of friendships, one of the
most basic human needs (Liiva & Cleave, 2005; Washington,
2007). Evidence for the ongoing deficits in social skills
functioning in older children with SLI also supports the
investigation of these skills in the preschool years (Baker &
Cantwell, 1987; Durkin & Conti-Ramsden, 2007; Fujiki,
Spackman, Brinton, & Hall, 2004).
Emergent Literacy Skills in Preschoolers with SLI
Researchers have examined the emergent literacy skills
of preschoolers, which lead to future reading, writing, and
spelling abilities (Catts, 1993; Catts, Bridges, Little, & Tomblin,
2008; Catts et al., 2002; Scarborough, 2001; Whitehurst &
Lonigan, 2001). It has been reported that almost half of
kindergartners with language impairment will demonstrate
reading disability by the second grade (Catts et al., 2002).
Similar to social skills development, not all preschoolers
with SLI demonstrate deficits in literacy development, in-
cluding alphabet knowledge and print concepts (Cabell et al.,
2009, 2010; Justice, Bowles, & Skibbe, 2006).
The development of preschoolers’ print concepts (e.g.,
words and knowledge of book conventions) is important,
as it is a predictor of future reading success. Print concepts
represent the understanding of the forms, functions, and
conventions of print (Justice et al., 2006). However, this area
is under-researched compared with the vast research on
phonological awareness (Justice et al., 2006). Additional
information is needed to determine how print concepts could
benefit from grammatical language intervention provided for
children with SLI during the preschool years. This informa-
tion may help to establish the nature of the relationship be-
tween oral language skills and print concepts development, as
well as effective methods to support development in both areas.
Grammatical Language Intervention
There is growing evidence supporting the worth of inter-
vention for grammatical development in preschoolers with
SLI (Leonard et al., 2006, 2008; Washington, Warr-Leeper,
& Thomas-Stonell, 2011). The repeated opportunities and
systematic support provided during therapy addressing
114 American Journal of Speech-Language Pathology • Vol. 22 • 113–125 • February 2013
morphosyntax, including the requirement to attend and listen
critically to grammar, could enhance control of attention and
responsiveness and thus lead to gains in untargeted domains.
Evidence for the spreading effects of morphosyntax therapy
on phonology has been demonstrated (e.g., Tyler, Kerry,
Haskill, & Tolbert, 2002). However, a study by Frome Loeb,
Stoke, and Fey (2001) showed little evidence to support the
claim that FFW-Language treatment, which included train-
ing for morphosyntax, had a positive impact on prereading
and reading skills in a small sample of 5- to 8-year olds.
Furthermore, the observed positive changes in pragmatics
skills could not be confidently linked to the intervention.
Substantiation for the spreading effects of grammatical
language intervention on untargeted domains such as social
and literacy skills is not yet apparent in the published peer-
reviewed literature. For preschoolers with SLI, grammatical
language intervention could affect other key developmental
areas that might benefit from the attention and consistency
provided during therapy. As a consequence of ongoing needs
in later social and literacy functioning, it becomes important
to investigate the possible benefits of grammatical language
intervention on the development of these skills for preschoolers
with SLI. Particular features of the grammar program could
help this population of children socialize better with others
and attend to print, thus demonstrating the broad and positive
impact of this type of therapy.
In an earlier study, the effects of paper-based and computer-
assisted expressive grammar intervention (EGI) were com-
pared with no intervention (NI) in a sample of thirty-four
3- to 5-year old preschoolers with SLI who were enrolled in
educational settings (Washington et al., 2011). During in-
tervention, multiple repeated and guided opportunities to
produce grammatically correct sentences were provided and
embedded within socialization development opportunities,
as well as print awareness and book knowledge activities.
Results demonstrated that EGI participants experienced sig-
nificantly greater expressive grammar outcomes compared
with NI participants immediately postintervention and at
3 months postintervention.
The repeated practice opportunities that required atten-
tion and critical listening to grammar could have helped
preschoolers to have better control and focus for specific
social skills and print concepts. The specific social skills that
are of interest include interactions with others, using play
and leisure time to engage with others, and demonstrations of
sensitivity and responsibility with others. The print concepts
include letter and word awareness and book knowledge.
For the remainder of this article, these two variables will be
named social skills and print concepts.
The preschoolers did not demonstrate inordinately low
abilities at preintervention for the social skills and print
concepts of interest. Consequently, it was believed that these
preschoolers could benefit from the nonspecific and inci-
dental social and print aspects of the grammatical language
intervention provided. As both the EGI and NI preschoolers
were enrolled in educational settings during the program,
opportunities to attend to social skills and print concepts
were being provided. If the EGI preschoolers demonstrated
greater improvements in these skills compared with the NI
preschoolers, who were similarly enrolled, then the gains of
the EGI preschoolers could be attributed to the grammatical
language intervention provided. Evaluation of these types
of outcomes is needed to establish the broad-based and pos-
itive effects of grammatical language intervention. With the
growing need to establish the outcomes of speech-language
pathology interventions, evidence supporting the wide-range
impact of our services is considered beneficial (Threats, 2003).
The Current Study
There is no available peer-reviewed published evidence
documenting the incidental positive effects, including main-
tenance of skills, for grammatical language intervention on
social skills and print concepts development in preschoolers
with SLI. The primary goal of this investigation was to
determine whether the EGI offered in the Washington et al.
(2011) study provided enrichment that resulted in improved
social skills and print concepts outcomes in the sample of
preschoolers. Outcomes were established at preintervention,
immediately postintervention, and at 3 months post-
intervention. The author sought to determine whether clini-
cally meaningful changes in performance (i.e., at least one
SD gain above preintervention scores at postintervention and
at 3 months postintervention) occurred and were maintained
for EGI or NI participants. Bain and Dollaghan (1991)
highlighted that amount of change (e.g., outside the pretest
range) is one factor that needs to be considered in order to
conclude that change is clinically significant. Another factor
is the functional impact of the change.
The effectiveness of the two EGI conditions, one paper-
based and the other computer-assisted, on social skills and
print concepts was not compared, as the author did not have a
rationale for why the two conditions might have different
outcomes. In the earlier study, the two EGI conditions resulted
in similar expressive grammar outcomes (cf. Washington et al.,
2011). The author explored the results and confirmed the
suspicion that there also were no differences in social skills
and print concepts outcomes for the two groups of EGI
preschoolers. Ultimately, combining the two interventions
provided a larger sample and better power to determine dif-
ferences between EGI and NI outcomes.
The secondary goal of the current study was to establish
the impact of initial expressive grammar level and /or
expressive grammar gains on preschoolers’ social skills and
print concepts outcomes. Exploring the nature of the rela-
tionships among initial expressive grammar level, gains in
expressive grammar, and social skills and print concepts
outcomes might offer some insight into relationships be-
tween oral language skills and other co-occurring areas of
development. The following research questions and were
addressed:
1. Does EGI that includes socialization development
opportunities, along with print awareness and book
knowledge activities, facilitate improved social skills
outcomes that are maintained after the intervention,
compared with NI?
2. Does EGI that includes socialization development
opportunities and print awareness and book knowledge
activities facilitate improved print concepts that are
maintained after the intervention, compared with NI?
Washington: Grammar Intervention and Social and Literacy Outcomes
3. Do preschool children’s initial expressive grammar level
and /or gains in expressive grammar predict social skills
outcomes, including maintenance of performance?
4. Do preschool children’s initial expressive grammar level
and/or gains in expressive grammar predict print concepts
outcomes, including maintenance of performance?
Corresponding hypotheses were made for each research
question: (a) social skills will significantly improve and will
be maintained for intervention participants only; (b) print
concepts will significantly improve and will be maintained
for intervention participants only; (c) initial expressive gram-
mar level and /or gains in expressive grammar will signif-
icantly predict social skills outcomes, including maintenance
of performance; and (d) initial expressive grammar level
and/or gains in expressive grammar will significantly pre-
dict print concepts outcomes, including maintenance of
performance.
Method
Only preschoolers from the Washington et al. (2011)
study participated in this project. An abridged version of
all procedures and protocols related to that study, including
measures utilized, is outlined in this article. The current
study represents a post hoc investigation of the spreading
effects of grammatical language intervention on these pre-
schoolers’ social skills and print concepts.
Participants
Participants were a convenience sample of 34 preschoolers
on a waitlist for intervention. Participants were recruited
from a government-funded preschool speech and language
initiative located in southwestern Ontario, Canada. As parents
accessed services, their speech-language pathologists (SLPs)
invited them to participate. The preschoolers’ ages ranged
from 3 years to 5 years (M = 52 months; SD = 5 months) at
preintervention. The participants were from urban and rural
regions, were monolingual English speakers, and were
identified by parents as Caucasian (n = 32), Asian (n = 1),
or other (n = 1). There were 27 boys and seven girls. All
preschoolers were enrolled in structured educational envir-
onments at the outset of intervention.
Preschoolers met the diagnostic criteria for SLI with
primary expressive grammar deficits. These criteria included:
(a) normal hearing sensitivity; (b) parent or SLP report of
absence of oral motor, neurological deficits, or pervasive devel-
opmental disorder; (c) normal receptive language skills (i.e.,
standard scores of 85 or greater) at the word level, as measured
by the Peabody Picture Vocabulary Test—IIIB (PPVT–IIIB;
Dunn & Dunn, 1997), and sentence level as measured by the
Clinical Evaluation of Language Fundamentals—Preschool
(CELF–P; Wiig, Secord, & Semel, 1992); (d) normal non-
verbal IQ (i.e., standard score of 85 or greater) as measured
by the Kaufman Brief Intelligence Test—2 (KBIT–2; Kaufman
& Kaufman, 2004) and (e) expressive language skills at or
below the 10th percentile as measured by two measures—
one a language test, the Structured Photographic Expressive
Language Test—Preschool (SPELT–P; Werner & Kresheck,
115
1983), and the other a language sample collected using devel-
opmental sentence scoring (DSS; Lee, 1974) procedures.
Although not a criterion for enrolment in the study, all parents
reported that they were already engaged in emergent literacy
activities at home (i.e., adult–child shared book-reading
activities).
After consent to participate was obtained, 22 children
were randomly selected for placement in the EGI group. The
remaining 12 children from the same intervention waitlist
served as the NI group.
Results of univariate analyses of variance (ANOVAs)
revealed no statistically significant differences between EGI
and NI participants for age, F(1, 32) = 3.75, p = .062, h2 = .11;
nonverbal cognitive skills, F(1, 32) = 0.90, p = .350, h2 = .03;
receptive language skills at the word level, F(1, 32) = 3.24,
p = .081, h2 = .09, and sentence level, F(1, 32) = 0.85, p = .365,
h2 = .03; and expressive language skills from a standardized
language test, F(1, 32) = 3.58, p = .067, h2 = .10, and a
spontaneous language sample, F(1, 32) = 3.05, p = .090,
h2 = .09. Participants were also not significantly different in
their preintervention social skills, F(1, 32) = 1.20, p = .281,
h2 = .04, and print concepts, F(1, 32) = 1.54, p = .223, h2 = .05.
See Table 1 for participant characteristics.
Procedures
Assessment protocol. Preschoolers were individually tested
using language and cognitive measures pursuant to the goals
of the Washington et al. (2011) study. The language mea-
sures included (a) the PPVT–IIIB (Dunn & Dunn, 1997),
used to evaluate receptive language skills at the word level;
(b) the receptive portions of the CELF–P (Wiig, Secord, &
Semel, 1992), used to evaluate receptive language skills
at the sentence level; (c) the SPELT–P (Werner & Kresheck,
1983), used to evaluate expressive grammar skills on a
standardized test; and (d) DSS (Lee, 1974), used to analyze a
45-min spontaneous language sample obtained during play
with a clinician. Both the SPELT–P and DSS procedures
were selected so that language skills in two distinct contexts
could be evaluated. The SPELT–P represented a standard-
ized evaluation of language skills, and the language sample
represented an evaluation of language skills with no prompting
for correct responses. Inter-rater reliability for the scoring
of the SPELT–P and DSS was completed by graduate
speech-language pathology students. For the SPELT–P and
DSS, inter-rater reliability was greater than 90%. Reliability
for DSS included point-to-point comparisons for word
transcription, appropriate DSS sentences, DSS category, and
DSS scoring. Nonverbal IQ was evaluated using the matrices
subtest of the KBIT–2 (Kaufman & Kaufman, 2004).
The DSS-scored spontaneous language samples were
used to establish initial expressive grammar level and ex-
pressive grammar gains. Use of spontaneous language samples
is considered a “gold standard” for assessing language skills
in children and has been suggested as being more repre-
sentative of children’s language skills compared with lan-
guage evaluations completed in more formal contexts where
optimal support is provided (Miller, 1996; Sealey & Gilmore,
2008). DSS z-scores for spontaneous language samples
collected at preintervention were used to represent initial
expressive grammar level. The DSS language samples total
scores were used to determine expressive grammar gains.
These scores represented the notion of “correct” production
for the entire sentence and not just attempts at creating
grammatically correct sentences. Expressive grammar gains,
along with social skills and print concepts outcomes, were
considered for change scores for preintervention to post-
intervention (immediate gains) and postintervention to 3 months
postintervention (continued growth).
Social skills and print concepts measures served as the
primary outcome measures for this study. Registered SLPs or
graduate students in speech-language pathology supervised

TABLE 1. Participants’ preintervention characteristics.

Total sample EGI NI

Characteristic (n = 34) (n = 22) (n = 12)

Age range 3;6 to 4;11 3;11 to 4;10 3;6 to 4;11

Gender
Female 7 6 1.
Male 27 16 1.
Receptive-wd 101.85 (4.88) 103.18 (4.75) 99.42 (4.30)
Receptive-sent 101.21 (6.95) 102.05 (8.33) 99.75 (2.90)
Expressive-standardized language test 10.26 (3.97)a 11.18 (3.03)a 8.58 (4.99)a
Expressive-spontaneous language 4.84 (.82)a 5.02 (.90)a 4.51 (.54)a
Nonverbal IQ 109.15 (10.10) 110.36 (10.97) 106.92 (8.24)
Social skills 103.26 (18.52) 105.55 (19.25) 99.08 (17.10)
73.04 (11.28)a 74.59 (11.16)a 70.17 (11.41)a
Print concepts 105.03 (14.16) 105.82 (15.73) 103.58 (11.23)
8.26 (4.13)a 8.91 (4.14)a 7.08 (4.01)a

Note. Means are reported, with SDs in parentheses. Standard scores are reported for all tests (M = 100, SD =15), with the exception of expressive
language measures. Expressive-standardized language test = Structured Photographic Expressive Language Test—Preschool (SPELT–P),
and Expressive-spontaneous language sample = DSS. Receptive-word (Receptive-wd) = Peabody Picture Vocabulary Test—III (PPVT–IIIB),
Receptive-sentence (Receptive-Sent) = Clinical Evaluation of Language Fundamentals—Preschool (CELF–P), Nonverbal IQ = Kaufman Brief Intelligence
Test—2 (KBIT–2), Social skills = Vineland Adaptive Behavior Scales (VABS), Print concepts = Preschool Word and Print Awareness test (PWPA).
a
Raw score.

116 American Journal of Speech-Language Pathology • Vol. 22 • 113–125 • February 2013

by registered SLPs administered these measures. Examiners
were trained on the measures to be used and were blind to
assessment time point and group assignment. To evaluate
social skills, the Socialization domain of the Vineland Adaptive
Behavior Scales (VABS; Sparrow, Balla, & Cicchetti, 1984)
Interview Edition Survey Form was completed. The VABS
is an interview-based standardized assessment of everyday
adaptations for four major domains, including Communica-
tion, Daily Living Skills, Socialization, and Motor Skills,
for children from birth to 18 years. The VABS is also an
established norm-referenced measure with acceptable psy-
chometric properties and can be used for progress monitor-
ing of a client’s level of functioning (Sparrow et al., 1984).
The Socialization domain provides information on a func-
tional skill that is observed in the client’s day-to-day life.
Individuals who know the client best and can provide in-
formation about level of functioning are most preferable for
completing the VABS domains (Sparrow et al., 1984).
Preschoolers’ parents acted as respondents for their chil-
dren’s social skills in day-to-day life. Using structured face-
to-face informal interviews, parents (32 mothers, two fathers)
described their children’s social skills in the following order
according to the three Socialization subdomains: (a) inter-
personal relationships—how does their child interact with
others; (b) play and leisure—how does their child spend his
or her play and leisure time; and (c) coping skills—how does
their child demonstrate sensitivity and responsibility to others.
Items were scored according to parental responses, with
scoring response options of usually, sometimes, or partially;
no/never; no opportunity; or don’t know. Using the VABS
interview protocol, basal (seven consecutive items scored
usually) and ceiling (seven consecutive items scored no/never)
rules were followed. Preschoolers’ ages were used to establish
the 20- to 30-min interview starting point.
The Preschool Word and Print Awareness test (PWPA;
Justice & Ezell, 2001) was used to evaluate print concepts.
The PWPA is an individually administered measure of
children’s print concepts that can be used to evaluate progress
in print concepts development in preschoolers (Justice &
Ezell, 2001; Justice et al., 2006; Justice, Kaderavek, Fan,
Sofka, & Hunt, 2009). Children’s knowledge of 14 concepts
about print (e.g., directionality) was completed during a
clinician–child shared book-reading activity using the com-
mercially available picture book Nine Ducks Nine (Hayes,
1990). The procedure used in this study adhered to the
protocol previously outlined in Justice et al. (2006) and Clay
(1979). This procedure is noncontrived and elicits informa-
tion about print concepts in a manner similar to parent
shared-reading activities. First, preschoolers were told,
“We are going to read this book together and I need you to
help me read.” Second, the examiner embedded a series of
print concepts tasks into the shared book-reading activity,
which included requests such as, “Show me the front of the
book,” “Show me the name of the book,” and “Where do I
begin to read?” Presentation of print concepts tasks was
the same for all preschoolers, and responses were scored as
correct, alternate correct, or incorrect.
Reliability for the VABS and PWPA. Double scoring for
parental responses on the VABS and preschoolers’ performance
on the PWPA was completed randomly and independently
Washington: Grammar Intervention and Social and Literacy Outcomes
for 21 (i.e., 20%) of the 102 assessment sessions. The sessions
were digitally recorded to facilitate rescoring of the original
testing. Each session selected was from a different participant
and was equally selected across groups and assessment time
points. Two trained research assistants were recruited to
perform this task. For the VABS, point-by-point agreement
in scoring averaged 98% (range, 94%–100%). For the
PWPA, point-by-point agreement in scoring averaged 99%
(range, 88%–100%).
Assessment sessions. All testing was completed during
90-min assessment sessions at preintervention, immediately
postintervention, and 3 months postintervention, in a stan-
dard clinical setting. During the assessment sessions, breaks
were provided as appropriate, based on child cues. Each
assessment session followed the same protocol, with the
exception of the nonverbal cognitive skills evaluation. Lan-
guage measures were administered first, followed by the
cognitive measure, and finally the literacy measure. For
younger participants who were below the lower age limit of
4 years for cognitive testing using the KBIT–2, evaluation
was completed in a separate 10-min session once they were
of appropriate age. The social skills measure was completed
with participants’ parents in a co-occurring assessment session.
EGI. The 22 participants in the EGI group received gram-
matical language intervention targeting the goal of accurate
production of third-person singular present progressive
sentences. Each sentence element (i.e., noun phrase subject
slot, verb, noun phrase object slot) was elicited individually
from each participant. EGI preschoolers received ten 20-min
intervention sessions, once a week. The interventions were
designed with specific features to directly address expressive
grammar deficits. Half of the EGI participants received
computer-assisted intervention, and half did not.
The following procedure was used during EGI: (a) An
introduction to the expected routine was made, followed by
a 2- to 7-min practice block. (b) Each sentence element
(subject noun phrase, verb, object noun phrase) was elicited
individually (sentence breakdown) using the following
questions: subject: “Who do you want to play with?”; verb:
“What is s/he doing?”; object: “What does s/he want to play
with?” (c) Each session continued to follow the same pro-
cedure until 80% accuracy on average over two consecutive
sessions was achieved. (d) At the beginning of the sub-
sequent session, participants were asked only to engage in
sentence build-up (“Put it all together”). No further sen-
tence breakdown was required once the 80% criterion was
achieved. See Appendix for a sample intervention routine.
Eleven EGI participants received computer-assisted in-
tervention using a software program named My Sentence
Builder (Washington & Warr-Leeper, 2006). This program
uses a syntactic slot-filler approach in which visual rep-
resentation (i.e., picture support) for every sentence element
(e.g., nouns, verbs, grammatical morphemes) that is needed
to create a present progressive sentence is provided. With
SLP support and guidance, participants moved from screen
to screen while selecting elements needed to create a gram-
matically correct sentence. Thus, participants were able to
observe the location of the “who,” the “is doing,” and the
“what” (noun and verb pictures) along with grammatical
elements (is, ing, the; depicted in triangles) in a sentence. Using
117
this approach, there was a slow, deliberate construction of each
sentence component to provide these participants with in-
creased time to process and produce grammatical information.
The remaining 11 EGI participants received non–computer-
assisted intervention. Typical conventional language training
procedures were utilized in clinician–client dyads. In lieu
of computer time, participants in this intervention received
visual supports provided through turning the pages of books
along with objects in play and picture cards with actions
to facilitate grammatical productions in a drill–play format.
Emphatic stress was included to facilitate attention to sentence
components (e.g., subject, verb, object). This technique of
stressing language targets has been recommended to increase
the salience of unstressed grammatical targets (Fey, Long, &
Finestack, 2003).
Social skills and print concepts intervention features.
While directly addressing grammatical skills, the SLP pro-
vided guidance for listening, taking turns, manipulating toy
objects, and interacting with others. When preschoolers
made mistakes, the SLP helped them cope to avoid frustration
with the task and with mistakes. Thus, preschoolers were
learning how to behave appropriately and demonstrate social
skills that facilitated inclusion with others.
In addition, during intervention sessions, the SLP used
emphatic stress or visual representations to highlight mor-
phemes, including grammatical inflections. The SLP pointed
to words and letters while turning the pages of books and
moving from screen to screen in the computer program.
Because children went from screen to screen while talking
about pictures and words when using the computer program,
the activity could be considered similar to turning the pages
of a book. The SLP also highlighted specific book con-
ventions (e.g., directionality, orientation) while targeting
grammatical productions. Thus, preschoolers were required
to attend to these specific print concepts when producing
each target sentence element (i.e., subject, verb, object).
Intervention fidelity. The author of the current study pro-
vided all intervention services. To establish adherence to
intervention protocols, 20% of sessions (44 sessions) were
randomly recorded and scored for accuracy of the interven-
tion implementation by two graduate students in speech-
language pathology. Sessions were observed equally for both
intervention programs. A checklist of the critical intervention
elements that included the intervention procedure, identifi-
cation of the interventionist, the session number, group type,
length of the session, and the techniques used during the
session was used for scoring. The intervention SLP adhered
to the protocol 100% of the time for all 44 intervention sessions.
NI. Twelve participants who were on the intervention
waitlist did not receive EGI during the study, thus serving as
control participants. Similar to intervention participants,
the NI participants were enrolled in structured educational
settings during the intervention study. Following the com-
pletion of this study, these participants were offered a block
of grammatical language intervention.
Design and Data Analysis
A pre–post–follow-up design was used. ANOVAs were
used to compare the social skills and print concepts outcomes
118 American Journal of Speech-Language Pathology • Vol. 22 • 113–125 • February 2013
of the two groups of preschoolers. Significant differences in
outcomes were established using preset alpha levels (p < .05).
Planned follow-up tests were completed for ANOVAs with
F values that met the established significance level. An
adjusted preset alpha level ( p < .017), representing a
Bonferroni correction, was used in these follow-up tests.
Effect sizes were reported using eta-squared (h2) values.
Regression analyses were completed to examine the
relationships between (a) initial expressive grammar level and
social skills and print concepts outcomes and (b) expressive
grammar gains and social skills and print concepts outcomes.
Preintervention to postintervention (immediate gains) and
postintervention to 3 months postintervention (continued
growth) performance for expressive grammar gains and
social skills and print concepts outcomes were considered in
the regression analyses. The significance of the relationships
was determined using a preset alpha level (p < .05). Stan-
dard conventions (i.e., 0.10, 0.30, and 0.50), established to
indicate small, medium, and large coefficients, were used
to determine the magnitude of the regression coefficients
(Green & Salkind, 2011).
Raw scores or raw change scores, not standard scores,
were used to represent outcomes (cf. Hadley, Olsen, & Earle,
2005). Raw scores on the VABS and the PWPA were entered
into SPSS to complete the ANOVAs. To complete the re-
gression analyses, raw change scores were used.
Results
Social Skills Outcomes After EGI
EGI participants’ mean performance at postintervention
revealed scores that were at least one SD above preinter-
vention performance, indicating clinically meaningful changes
in social skills. This pattern was maintained at 3 months
postintervention. A similar outcome was not evident for NI
participants. See Table 2 for mean performance on the VABS
at each time point.
A mixed-model ANOVA was used to test for differences
between social skills scores of the EGI and NI groups of
preschoolers. Results from the ANOVA met the preset alpha
level of .05. A significant interaction was found between
time and group, F(2, 64) = 32.50, p < .001, partial h2 = .50.
Thus, the main effects were not interpreted. Instead, follow-up
tests using one-way univariate ANOVAs for VABS scores
at each time point were completed (Portney & Watkins,
2009). Findings at preintervention did not meet the preset
TABLE 2. Participants’ mean performance on the VABS.
Group Time n M (SD )
EGI Pre 22 74.59 (11.16)
Post 22 94.95 (7.08)
3 months post 22 95.68 (7.72)
NI Pre 12 70.12 (11.41)
Post 12 74.33 (12.35)
3 months post 12 75.30 (10.86)
Note. Scores represent participants’ raw score performance at each
time point.
alpha level of .017, F(1, 32) = 1.20, p = .281, h2 = .04. How-
ever, findings at postintervention, F(1, 32) = 38.68, p < .001,
h2 = .55, and at 3 months postintervention, F(1, 32) = 47.95,
p < .001, h2 = .60, met the set significance level for these
follow-up tests. Pairwise comparisons of means at post-
intervention and 3 months postintervention revealed that
EGI participants had higher means compared with controls.
See Figure 1 for preintervention, postintervention, and
3 months postintervention performance on the VABS.
Print Concepts Outcomes After EGI
EGI participants’ mean performance at postintervention
revealed scores that were at least one SD above preinter-
vention performance, indicating clinically meaningful changes
in print concepts. This pattern was maintained at 3 months
postintervention. A similar outcome was not evident for NI
participants. See Table 3 for mean performance on the PWPA
at each time point.
A mixed-model ANOVA was tested for differences be-
tween print concepts scores of the EGI and NI groups of
preschoolers. Results from the ANOVA met the preset alpha
level of .05. A significant interaction was found between
time and group, F(2, 64) = 14.84, p < .001, partial h2 = .32.
Thus, the main effects were not interpreted. Instead, follow-up
tests using one-way univariate ANOVAs for PWPA scores
at each time point were completed. Findings at preintervention
did not meet the preset alpha level of .017, F(1, 32) = 1.54,
p = .223, h2 = .05. However, findings at postintervention,
F(1, 32) = 10.12, p < .001, h2 = .24, and at 3 months post-
intervention, F(1, 32) = 28.56, p < .001, h2 = .47, met the set
significance level for these follow-up tests. Pairwise com-
parisons of means at postintervention and 3 months post-
intervention revealed that EGI participants had higher PWPA
mean scores than controls. See Figure 2 for preintervention,
postintervention, and 3 months postintervention performance.
Factors Predicting Social Skills Outcomes
Multiple regression analyses were completed to establish
the relationships among initial expressive grammar level,
FIGURE 1. Mean performance of social skills between the two
groups illustrated at the three assessment time points. VABS =
Vineland Adaptive Behavior Scales; EGI = expressive grammar
intervention; NI = no intervention.
Washington: Grammar Intervention and Social and Literacy Outcomes
TABLE 3. Participants’ print concepts mean performance on the
PWPA.
Group Time n M (SD )
EGI Pre 22 8.91 (4.14)
Post 22 13.32 (3.24)
3 months post 22 14.91 (1.90)
NI Pre 12 7.08 (4.01)
Post 12 8.83 (4.99)
3 months post 12 8.43 (5.14)
Note. Scores represent participants’ raw score performance at each
time point.
expressive grammar gains, and social skills. All partic-
ipants were included in these analyses. Raw change scores
for immediate gains (pre–post) and continued growth (post–
3 months post) on the VABS served as the dependent variables.
Given the sample size in this study, no more than three
predictor variables were selected.
Two-predictor and three-predictor regression models
were used. In the two-predictor models, preschoolers’ initial
expressive grammar level and immediate expressive gram-
mar gains served as predictor variables. In the three-predictor
models, preschoolers’ initial expressive grammar level,
immediate expressive grammar gains, and continued growth
in expressive grammar gains served as the predictor variables.
The regression model with VABS preintervention to
postintervention raw change scores demonstrating immedi-
ate gains, initial expressive grammar level, and immediate
expressive grammar gains was significant, R = .48, R2 = .23,
F(2, 31) = 4.71, p = .016. In this two-predictor model, initial
expressive grammar level did not significantly correlate with
VABS scores, r(32) = .17, p = .167. However, a signifi-
cant positive correlation was observed between immediate
expressive grammar gains and VABS change scores, r(32) = .45,
p = .004 (see Table 4). Expressive grammar gains were a
significant predictor of VABS change scores, b = .45, t(31) =
2.87, p = .007. The magnitude of the regression coefficient
(b) was medium. Initial expressive grammar level was not
FIGURE 2. Mean performance of print concepts between the two
groups illustrated at the three assessment time points. PWPA =
Preschool Word and Print Awareness Test.
119
TABLE 4. Correlations among regression variables.
DSS DSS VABS
pre–post post–3 months pre–post
DSS pre–post — .34* .45*
DSS post–3 months .34* — .37*
VABS pre–post .45* .37* — .14
VABS post–3 months .07 .02 .14
PWPA pre–post .19 .26 –.01
PWPA post–3 months .36 .03 .48**
DSS z-score .01 .25 .17
*p < .05. **p < .01.
found to be a significant individual predictor of VABS per-
formance, b = .17, t(31) = 1.07, p = .294.
The three-predictor regression model with VABS post-
intervention to 3 months postintervention raw change scores
(continued growth), initial expressive grammar level, imme-
diate expressive grammar gains, and continued growth
in expressive grammar gains was not significant, R = .07,
R2 < .001, F(3, 30) = 0.05, p = .985. The corresponding
correlations among predictor and dependent variables in this
regression model were also not significant (see Table 4). The
finding for this regression model suggests that the independent
variables, initial expressive grammar level, b = –.002, t(30) =
–0.01, p = .992, immediate expressive grammar gains, b = .07,
t(30) = 0.37, p = .715, and continued expressive grammar
gains, b = –.006, t(30) = –0.03, p = .978, did not predict
continued growth in participants’ social skills.
Factors Predicting Print Concepts Outcomes
The approach to regression analyses for social skills
was also followed for the regression analyses completed for
print concepts immediate gains and continued growth on
the PWPA. The two-predictor regression model with PWPA
preintervention to postintervention raw change scores
(immediate gains), initial expressive grammar level, and
immediate expressive grammar gains was not significant,
R = .24, R2 = .06, F(2, 31) = 0.95, p = .40. The corresponding
correlations among predictor and dependent variables in this
regression model were also not significant (see Table 4).
Neither initial expressive grammar level, b = .14, t(31) = 0.82,
p = .421, nor immediate expressive grammar gains, b = .19,
t(31) = 1.11, p = .277, predicted changes in preintervention
to postintervention print concepts.
The three-predictor regression model with PWPA post-
intervention to 3 months postintervention raw change scores
(continued growth), initial expressive grammar level, imme-
diate expressive grammar gains, and continued expressive
grammar gains also was not significant, R = .37 R2 = .14,
F(3, 30) = 1.81, p = .167. Initial expressive grammar level,
r(32) = .09, p = .304, and continued growth in expressive
grammar gains, r(32) = .03, p = .445, did not significantly
correlate with the dependent variable. There was, however, a
significant positive correlation between immediate expres-
sive grammar gains and PWPA change scores for continued
growth, r(32) = .36, p = .019 (see Table 4). Immediate ex-
pressive grammar gains was found to be the only significant
120 American Journal of Speech-Language Pathology • Vol. 22 • 113–125 • February 2013
VABS PWPA PWPA DSS
post–3 months pre–post post–3 months z-score
.07 .19 .36** .01
.02 .26 .03 .25
.34* .48** .17
— .03 .03 –.003
.34* — .28 .14
.03 .28 — .09
–.003 .14 .01 —
predictor of PWPA change scores, b = .36, t(30) = 2.14,
p = .031. The magnitude of the regression coefficient (b) was
medium. Neither initial expressive grammar level, b = .13,
t(30) = 0.72, p = .477, nor continued growth in expressive
grammar gains, b = .15, t(30) = –0.79, p = .439, predicted
continued growth in PWPA change scores.
Discussion
Two major findings were reported in the current study. As
hypothesized, compared with NI, EGI for preschool SLI was
associated with greater improvement of social skills—as
measured by parent report—and of print concepts outcomes,
as measured by a literacy test. The postintervention and 3
months postintervention scores achieved for social skills and
print concepts outcomes were at least one SD above pre-
intervention performance for intervention participants only.
The functional impact associated with changes in social
skills was that parents reported observing their children
socialize better with others in everyday “naturalistic” contexts
after intervention. Because print concepts were observed
and evaluated in a structured setting, the functional impact
associated with these changes was not as clear as compared
with that for social skills. However, scoring of specific items
on the PWPA required the examiner to observe the pre-
schoolers’ unprompted demonstration of print concepts (e.g.,
turning the book to the front to identify the title). This pro-
vided the opportunity to observe that preschoolers had an
increased ability to spontaneously engage in print-concepts
activities.
The improvements for intervention participants were of
statistical and clinical significance and, most important, were
maintained for 3 months postintervention. Bain and Dollaghan
(1991) highlighted the importance of such large changes.
Because the intervention and control preschoolers did not
demonstrate inordinately low abilities for social skills and
print concepts at preintervention, EGI offered greater enrich-
ment, not remediation, of these skills compared with NI.
Because preschoolers in both groups were enrolled in edu-
cational settings, which provided them with opportunities to
engage in social skills and attend to print concepts, the ad-
ditional benefits of receiving the grammatical language
intervention are even more important.
The second finding of the current study was that a signif-
icant predictive value of the two-predictor model containing
initial expressive grammar level and preintervention to post-
intervention expressive grammar gains was evident for social
skills immediate gains. No other models were found to be
predictive of either social skills or print concepts outcomes.
It is important to note, however, that expressive grammar
gains were the only significant single predictor of both social
skills and print concepts outcomes. There was a differential
effect of this predictor variable, with immediate expressive
grammar gains predicting immediate social skills and con-
tinued growth (i.e., maintenance of gains) for print concepts.
The hypotheses regarding the relationships between depen-
dent and independent variables were partially supported
on the basis of these results.
Social Skills Outcomes After EGI
EGI was associated with significantly greater social skills
outcomes compared with NI. These outcomes were evi-
dent immediately after intervention and were maintained for
3 months postintervention. Specific intervention features
may have been beneficial in facilitating the observed out-
comes for intervention participants.
During the intervention program, preschoolers had multiple
guided opportunities from the SLP for learning how to take
turns, listen and play appropriately, and interact with others.
The SLP encouraged and directed each preschooler to wait for
his or her own turn before being permitted to produce sentence
components. Preschoolers also had to learn to listen to in-
structions in their entirety before completing a task. The SLP
also helped to model appropriate play behaviors for times
when a preschooler engaged with toys and then encouraged
the preschooler to demonstrate similar actions. In addition to
these aspects of the intervention, the repeated opportunities
for producing target components also provided a guided con-
text for preschoolers to learn how to cope with making a
mistake. Preschoolers in the control group did not have these
opportunities and also did not experience equal or better out-
comes than their intervention cohorts.
Older children with SLI show associated long-term dif-
ficulties in social skills development (Baker & Cantwell,
1987; Durkin & Conti-Ramsden, 2007; Fujiki et al., 2004).
Interventions that facilitate the development of language
skills by incorporating socialization activities during the pre-
school years might offer some benefits to children’s ongoing
development of social skills in the later school years.
Print Concepts Outcomes After EGI
EGI was also associated with significantly greater print
concepts outcomes compared with NI. SLP guidance for
critical intervention features—such as pointing to and talking
about print, and turning the pages of a book or moving from
screen to screen on a computer while showing preschoolers
how to construct and “read” a sentence—appeared to help
increase the children’s awareness of print and knowledge of
book conventions. For example, the SLP pointed to printed
grammatical forms, such as is, ing, and the, to help highlight
elements needed in a grammatically correct sentence. Thus,
the SLP directed children’s attention to the letters and words in
these grammatical forms, as well as increased their awareness
Washington: Grammar Intervention and Social and Literacy Outcomes
of the left-to-right orientation and directionality in reading,
thereby helping the children to produce grammatically correct
sentences. As the systematic support was repeated session to
session, intervention participants had multiple practice oppor-
tunities for exposure to directionality, orientation, letters, words,
and other print concepts that enhanced their literacy growth,
compared with that of control participants. Consequently, inter-
vention participants’ motivation toward, interest in, and con-
tinued understanding of the target print concepts may have also
increased more than it did for those who had no intervention.
Information on development of print concepts is needed
(Justice et al., 2006). The findings from the current study
expand the available knowledge about print concepts out-
comes for preschoolers with SLI. Because there are asso-
ciated long-term negative effects of having a language
impairment in kindergarten on later reading success, practi-
tioners may want to include materials and activities that
enhance print concepts as well as language skills.
Factors Predicting Social Skills and
Print Concepts Outcomes
A secondary focus of the current study was determining
the impact of initial expressive grammar level and /or ex-
pressive grammar gains on social skills and print concepts
outcomes over time. The specific time points investigated
were preintervention to postintervention (immediate gains)
and postintervention to 3 months postintervention (continued
growth). Of equal importance to the other findings in the
current study is that expressive grammar gains were the only
significant single predictor of social skills and print concepts
outcomes. There was a differential impact of the predictor
variable depending on the time period of the gains in ex-
pressive grammar as well as when the social skills and print
concepts outcomes were achieved. These findings are dis-
cussed separately in the upcoming sections.
Factors predicting social skills outcomes. The two-
predictor model containing initial expressive grammar level
and immediate expressive grammar gains in relationship to
preintervention to postintervention social skills outcomes
showed a significant predictive value. The results of the
regression indicated that the two grammar predictors explained
23% of the variance in social skills changes. As an inde-
pendent variable, immediate expressive grammar gains was
found to be a better predictor of social skills performance
than was initial language level. The direction of the corre-
lational coefficient was positive, suggesting that those pre-
schoolers with greater gains in expressive grammar skills
during the intervention period also experienced greater social
skills changes during the same time period. Furthermore,
those children with fewer gains in expressive grammar also
experienced fewer gains in social skills.
Previous researchers have highlighted the idea that oral
language skills are a principal means to achieve successful
social interactions (Bishop, 1997; Washington, 2007). EGI
participants’ language skills improved from preintervention
to postintervention more than those of NI participants.
The indirect effects of the added benefits of grammatical
language intervention for treatment participants are con-
sidered important to predicting immediate social skills changes.
121
Consequently, as oral language functioning for expressive
grammar improved, corresponding improvements in interper-
sonal interactions, play, and coping also were evident. Based
on this study, however, it is not possible to determine the
relative impact of improved grammar versus incidental activ-
ities that enhance social skills during intervention.
The three-predictor model that included grammar pre-
dictors and continued changes in social skills was not found
to have a significant predictive value. The predictors included
in the model also did not have individual effects on the
dependent variable. The lack of correlation between initial
expressive grammar level and changes in social skills for
continued growth could be a result of the limited variability
in initial expressive grammar level among the preschoolers
as well as their initial level of social skills. The lack of sig-
nificance found for the correlation between both types of
expressive grammar gains and continued growth in social
skills could also be due to a leveling off of the language
differences between EGI and NI participants. Specifically,
the magnitude and impact of those differences decreased.
Thus, all participants could have experienced language gains
large enough to facilitate the observed changes in social
skills at this time point.
Factors predicting print concepts outcomes. Unlike the
findings observed for the models of social skills changes,
neither the two-predictor model nor the three-predictor model
was found to have a significant predictive value for print
concepts changes. This means that the combined effect of the
various predictor variables did not have a significant rela-
tionship with the dependent variable. The limited variability
in initial expressive grammar level as well as the high initial
level of print concepts skills could explain the nonsignificant
correlations observed.
The benefit of conducting a standard multiple regression
is that the model’s predictive value and the value of each
independent variable, controlling for other independent
variables, can be determined (Green & Salkind, 2011; Portney
& Watkins, 2009). It was found that expressive grammar
gains as a single independent variable was the best and only
significant predictor of print concepts changes. Immediate
expressive grammar gains predicted continued growth in print
concepts. Preschoolers who experienced greater immediate
gains in expressive grammar also experienced greater con-
tinued growth in print concepts. In addition, preschoolers
who experienced lesser gains in expressive grammar also
experienced lesser continued gains in print concepts.
Learning through observation and interaction in early
literacy experiences promotes the early development of print
concepts and other literacy skills more so than overt teaching
(Cabell et al., 2010; Justice et al., 2006). There are added
indirect positive effects for print concepts provided by lit-
eracy experiences associated with grammatical language
intervention. As the significance of the effect was observed
only for continued growth in print concepts, more time may
be needed for this effect to be demonstrated. It has been
suggested that in contrast to social skills development, the
abstraction of knowledge for print concepts might not occur
as easily, as literacy is an artificial construct that does not
naturally occur by virtue of exposure to others (Washington,
2011). Instead, as noted by other researchers, these exposures
122 American Journal of Speech-Language Pathology • Vol. 22 • 113–125 • February 2013
must be based in literacy experiences (Cabell et al., 2009,
2010; Justice et al., 2006).
Continued opportunities for exposure to literacy experi-
ences likely occurred for preschoolers from the end of the
postintervention to 3 months postintervention. Thus, pre-
schoolers may have required more time for the integration,
application, and manifestation of the incidental knowledge
gained for print concepts during the intervention period. The
lack of a significant relationship between continued growth
in expressive grammar gains and continued growth in print
concepts changes could also be explained using this rationale.
Therefore, for the effects of continued growth in grammar
to be manifested, the evaluation of continued growth in print
concepts would need to extend past the postintervention to
3 months postintervention time frame used in this study.
Limitations and Future Directions
The main limitation of this study was that it was a post
hoc question. The incidental effects of grammatical language
intervention on social skills and print concepts were not
planned as a primary research question for the original study.
Consequently, there was no control group to compare the
effects of incidental teaching of social skills and print con-
cepts to a context in which these incidental features were not
included. In future investigations, to determine the differ-
ential effects of intervention on print concepts, researchers
should explore using a control group that receives the gram-
mar intervention without literacy experiences.
Another limitation was that direct examination of pre-
schoolers’ social skills was not completed; instead, the cur-
rent study relied on parental report of these skills using a
validated measure. These reports may have been affected by a
treatment participation bias. That said, recent literature indi-
cates that parents can make accurate and reliable judgments
about children’s social skills, specifically regarding partici-
pation (Thomas-Stonell, Oddson, Robertson, & Rosenbaum,
2009). Direct evaluation and observation of preschoolers’
social skills in their everyday settings at different time points
would provide more empirical insights into how grammatical
language intervention affects the spontaneous use of social
skills in different settings.
The lack of a true randomization procedure in the larger
Washington et al. (2011) study is acknowledged as another
limitation. It is possible that because parents in the larger
study self-selected their preschool-age children for enroll-
ment in the study, parents and their preschoolers were more
likely to complete the study than were others who would
have been randomly selected.
The sample of preschoolers included in the current study
may not be representative of all preschoolers with SLI. This
fact is acknowledged as a significant limitation. The current
preschoolers had normal receptive language skills and were
already actively engaging in print concepts tasks such as
shared book-reading activities with their parents. Furthermore,
performance within the average range for both social skills
and print concepts was also observed. Thus, the deficits in
language skills for grammatical production did not appear to
affect social skills and print concepts development for these
preschoolers. It is possible that the participants’ age as well as
the measures used to evaluate the social skills and print
concepts might account for this discrepancy. As a result of
these limitations, application of the study findings to other
clinical populations of preschoolers with language impair-
ment may be limited. Children with more severe language
impairments, or those with poor social skills and print con-
cepts, might not be able to benefit from the incidental learning
opportunities provided by the grammatical language therapy
in this investigation.
Further research using a larger and more representative
sample of preschoolers with SLI could strengthen the con-
clusions that can be drawn about the impact of implemented
grammatical language intervention on social skills and print
concepts. There is considerable evidence that most children
with a history of developmental language impairments are at
risk for failure to develop literacy skills and establish suc-
cessful and meaningful social interactions (Baker & Cantwell,
1987; Cabell et al., 2010; Catts et al., 2002; Fujiki et al., 2004).
Thus, growth in both the social and literacy domains would
be a natural consequence of language intervention for this group
of children.
Print concepts, which are successful in predicting future
reading success, remain under-researched in the preschool
population and should be the focus of continued investiga-
tions. Researchers also should examine other areas of social
skills, overall emergent literacy skills, and other predictors of
performance to provide more information about the potential
impact of intervention on the development of these skills.
Use of other measures of social and print concepts functioning
should be considered for determining growth in these areas.
Conclusions and Implications for Clinical Practice
Statistically significant and clinically meaningful out-
comes in social skills and print concepts outcomes are asso-
ciated with grammatical language intervention for preschoolers
with SLI. Gains in expressive grammar are a significant
predictor for both of these outcomes. Therefore, gains in
social skills and print concepts should coincide with improved
expressive grammar skills.
The grammatical language intervention was associated
with improvements in social skills and print concepts out-
comes that were achieved immediately postintervention and
were maintained for 3 months afterward. NI participants
failed to experience these outcomes. This is an impressive
finding that attests to the effectiveness of the intervention
provided. Specific intervention features provided added
learning benefits that offered enrichment rather than reme-
diation for the development of social skills and print concepts
in the sample of preschoolers. Ultimately, the mechanisms of
EGI spurred changes that were not apparent for the NI group
with the same characteristics and enrolled in the same edu-
cational settings. Therefore, positive, broad-based effects of
grammatical language intervention provided by SLPs have
been supported.
Because the control group did not experience a similar
performance pattern, the outcomes evident in specific social
skills and print concepts were beyond those attributed to
maturation alone. Attention in therapy to syntactic knowledge
that contains literacy experiences as well as therapy in which
Washington: Grammar Intervention and Social and Literacy Outcomes
opportunities for guided socializing can occur are important
factors that can lead to broader based outcomes for pre-
schoolers with SLI. Because the observed social skills and
print concepts outcomes are best predicted by gains in gram-
matical skills, programs that enhance language, social, and
literacy development are encouraged. A cascade of positive
experiences in these co-occurring areas of development
after grammatical language intervention is possible for
preschoolers with SLI.
Acknowledgments
This research project was completed as part of the author’s
doctoral thesis at the University of Western Ontario in London,
Ontario, Canada. The research was financially supported by the
Canadian Language and Literacy Research Network, the Ontario
Graduate Scholarship, the Ontario Graduate Scholarship in Science
and Technology, the Research Alliance for Children with Special
Needs, and a University of Western Ontario Thesis Award. The
author acknowledges the unwavering support and contributions of
Genese Warr-Leeper. The author also acknowledges the families,
children, speech-language pathologists, graduate students, and
research assistants whose participation made possible the timely
completion of this study. Finally, the author would like to thank
Julia Colangeli and Nancy Creaghead for their editorial contribu-
tions as well as her family for their unconditional support.
References
Bain, B., & Dollaghan, C. (1991). Treatment efficacy: The notion
of clinically significant change. Language, Speech, and Hearing
Services in Schools, 22, 264–270.
Baker, L., & Cantwell, D. (1987). A prospective psychiatric
follow-up of children with speech/language disorders. Journal
of the American Academy of Child and Adolescent Psychiatry,
26, 546–553.
Bishop, D. (1997). Uncommon understanding: Development and
disorders of language comprehension in children. Hove, East
Sussex, United Kingdom: Psychology Press.
Cabell, S., Justice, L., Zucker, T., & McGinty, A. (2009).
Emergent name-writing abilities of preschool-age children with
language impairment. Language, Speech, and Hearing Services
in Schools, 40, 53–66.
Cabell, S., Lomax, R., Justice, L., Breit-Smith, A., Skibbe, L.,
& McGinty, A. (2010). Emergent literacy profiles of preschool-
age children with specific language impairment. International
Journal of Speech-Language Pathology, 12, 472–482.
Catts, H. (1993). The relationship between speech-language
impairments and reading disabilities. Journal of Speech,
Language, and Hearing Research, 36, 948–959.
Catts, H., Bridges, M., Little, T., & Tomblin, J. (2008).
Reading achievement growth in children with language
impairments. Journal of Speech, Language, and Hearing
Research, 51, 1569–1579.
Catts, H., Fey, M., Tomblin, J., & Zhang, X. (2002). A longi-
tudinal investigation of reading outcomes in children with
language impairments. Journal of Speech, Language, and
Hearing Research, 45, 1142–1157.
Clay, M. (1979). The early detection of reading difficulties: A
diagnostic survey with recovery procedures. Portsmouth, NH:
Heinemann.
Dunn, L., & Dunn, L. (1997). Peabody Picture Vocabulary Test—III.
Circle Pines, MN: AGS.
123
Durkin, K., & Conti-Ramsden, G. (2007). Language, social
behavior, and the quality of friendships in adolescents with and
without a history of specific language impairment. Child
Development, 78, 1441–1457.
Fey, M., Long, S., & Finestack, L. (2003). Ten principles of
grammar facilitation for children with specific language impair-
ment. American Journal of Speech-Language Pathology, 12,
3–15.
Frome Loeb, D., Stoke, C., & Fey, M. (2001). Language
changes associated with FastForward: Evidence from case
studies. American Journal of Speech-Language Pathology, 10,
216–230.
Fujiki, M., Spackman, M., Brinton, B., & Hall, A. (2004).
The relationship of language and emotion regulation skills to
reticence in children with specific language impairment. Journal
of Speech and Hearing Research, 47, 637–646.
Green, S., & Salkind, N. (2011). Using SPSS for Windows and
Macintosh: Analyzing and understanding data (6th ed.). Upper
Saddle River, NJ: Prentice Hall.
Hadley, P., Olsen, J., & Earle, C. (2005, November). Clinical
decision-making with late-talking children. Miniseminar pre-
sentation to the American Speech-Language-Hearing Associa-
tion Annual Convention, San Diego, CA.
Hayes, S. (1990). Nine ducks nine. Cambridge, MA: Candlewick
Press.
Justice, L., Bowles, R., & Skibbe, L. (2006). Measuring pre-
school attainment of print-concept knowledge: A study of
typical and at-risk 3- to 5-year-old children using item response
theory. Language, Speech, and Hearing Services in Schools, 37,
224–235.
Justice, L., & Ezell, H. (2001). Word and print awareness in
4-year-old children. Child Language Teaching and Therapy,
17, 207–226.
Justice, L., Kaderavek, J., Fan, X., Sofka, A., & Hunt, A.
(2009). Accelerating preschoolers’ early literacy development
throught classroom-based teacher-child storybook reading and
explicit print referencing. Language, Speech, and Hearing
Services in Schools, 40, 67–85.
Kaufman, A., & Kaufman, N. (2004). Kaufman Brief Intelli-
gence Test—2 (KBIT–2). Circle Pines, MN: AGS.
Lee, L. (1974). Developmental sentence analysis: A grammatical
assessment procedure for speech and language clinicians.
Evanston, IL: Northwestern University Press.
Leonard, L. (1998). Children with specific language impairment.
Cambridge, MA: MIT Press.
Leonard, L., Camarata, S., Pawlowska, M., Brown, B., &
Camarata, M. (2006). Tense and agreement morphemes in the
speech of children with specific language impairment during
intervention: Phase 2. Journal of Speech, Language, and
Hearing Research, 49, 749–770.
Leonard, L., Camarata, S., Pawlowska, M., Brown, B., &
Camarata, M. (2008). The acquisition of tense and agreement
morphemes by children with specific language impairment
during intervention: Phase 3. Journal of Speech, Language, and
Hearing Research, 51, 120–125.
Liiva, C., & Cleave, P. (2005). Roles of initiation and respon-
siveness in access and partcipation for children with specific
language impairment. Journal of Speech, Language, and
Hearing Research, 48, 868–883.
McCabe, P. (2005). Social and behavioral correlates of preschoolers
with specific language impairment. Psychology in Schools, 42,
373–387.
Miller, J. (1996). Progress in assessing, describing, and defining
child langauge disorder. In K. Cole, P. Dale, & D. Thal (Eds.),
Assessment of communication and language (pp. 309–324).
Baltimore, MD: Brookes.
124 American Journal of Speech-Language Pathology • Vol. 22 • 113–125 • February 2013
Plante, E. (1998). Criteria for SLI: The Stark and Tallal legacy and
beyond. Journal of Speech, Language, and Hearing Research,
41, 951–957.
Portney, L., & Watkins, M. (2009). Foundations of clinical
research: Applications to practice (3rd ed.). Upper Saddle
River, NJ: Prentice Hall Health.
Scarborough, H. (2001). Connecting early langauge and literacy
to later reading (dis)abilities: Evidence, theory, and practice.
In S. B. Neuman & D. K. Dickson (Eds.), Handbook of early
literacy research (pp. 97–110). New York: Guilford Press.
Sealey, L., & Gilmore, S. (2008). Effects of sampling context on
the finite verb production of children with and without delayed
language development. Journal of Communication Disorders,
41, 223–258.
Sparrow, S., Balla, O., & Cicchetti, D. (1984). Vineland
Adaptive Behavior Scales (VABS): Interview edition. Circle
Pines, MN: AGS.
Stark, R. E., & Tallal, P. (1981). Selection of children with specific
language deficits. Journal of Speech and Hearing Disorders, 46,
114–122.
Thomas-Stonell, N., Oddson, B., Roberston, B., & Rosenbaum, P.
(2009). Predicted and observed outcomes in preschool children
following speech and language treatment: Parent and clini-
cian perspectives. Journal of Communication Disorders, 42,
29–42.
Threats, T. (2003). The conceptual framework of ASHA’s new
scope of practice for speech-language pathology. Journal of
Medical Speech-Language Pathology, 10, xvii–xxiv.
Tomblin, J., Records, N., Buckwalter, P., Zhang, X., Smith, E.,
& O’Brien, M. (1997). Prevalence of specific language im-
pairment in kindergarten children. Journal of Speech, Language,
and Hearing Research, 40, 1245–1260.
Tyler, A., Kerry, E., Haskill, A., & Tolbert, L. (2002). Efficacy
and cross-domain effects of a morphosyntax and phonology
intervention. Language, Speech, and Hearing Services in
Schools, 33, 52–66.
Washington, K. (2007). Using the ICF within speech-language
pathology: Application to developmental language impair-
ment. International Journal of Speech-Language Pathology, 9,
242–255.
Washington, K. (2011, November). Social and emergent literacy
outcomes following expressive grammar intervention in pre-
school SLI. Paper presented at the Annual Convention of the
American Speech-Language-Hearing Association, San Diego,
CA.
Washington, K., & Warr-Leeper, G. (2006). A collaborative
approach to computer-assisted treatment of preschool children
with specific language impairment. OSLA Connection Journal,
2(2), 10–11, 16.
Washington, K., Warr-Leeper, G., & Thomas-Stonell, N.
(2011). Exploring the outcomes of a novel computer-assisted
treatment program targeting expressive-grammar deficits in
preschoolers with SLI. Journal of Communication Disorders,
44, 315–330.
Werner, O., & Kresheck, J. (1983). The Structured Photographic
Expressive Language test—Preschool (SPELT–P). Dekalb, IL:
Janelle Publications.
Whitehurst, G., & Lonigan, C. (2001). Emergent literacy:
Development from prereaders to readers. In S. B. Neuman & D. K.
Dickson (Eds.), Handbook of early literacy research (pp. 11–29).
New York, NY: Guilford Press.
Wiig, E., Secord, W., & Semel, E. (1992). Clinical Evaluation of
Language Fundamentals—Preschool. San Diego, CA: Harcourt
Brace & Company.
Appendix
Sample SLP–Preschooler Intervention Dyad

SLP: We are going to talk about boys or girls doing different things. You will have lots of time to practice telling
me what different boys or girls are doing. I will be helping you a lot. Now let’s start.
SLP: Who do you want to play with?
Preschooler: Girl.
SLP (using emphatic stress or pointing to grammatical image): Girl?
Preschooler: The girl.
SLP: What is the girl doing? The girl. . .
Preschooler: Eating.
SLP (using emphatic stress or pointing to grammatical image): Eating?
Preschooler: . . .is eating.
SLP: What is the girl eating? The girl is eating . . .
Preschooler: A hot dog.
SLP: Now put it all together.
Preschooler: The girl is eating a hot dog.

Washington: Grammar Intervention and Social and Literacy Outcomes 125

The Association Between Expressive Grammar Intervention and Social and
Emergent Literacy Outcomes for Preschoolers With SLI

Karla N. Washington
Am J Speech Lang Pathol 2013;22;113-125; originally published online Oct 15,
2012;
DOI: 10.1044/1058-0360(2012/11-0026)

The references for this article include 17 HighWire-hosted articles which you can
access for free at: http://ajslp.asha.org/cgi/content/full/22/1/113#BIBL

This information is current as of May 5, 2013

This article, along with updated information and services, is

located on the World Wide Web at:
http://ajslp.asha.org/cgi/content/full/22/1/113