Abstract
Justice, L., Logan, J., Kaderavek, J., Schmitt, M. B., Tompkins, V., & Bartlett, C. (2015). Empirically based profiles of the early literacy skills of children with language impairment in early childhood special education. Journal of Learning Disabilities, 48, 482–494.
Recently I was in an Individualized Educational Plan (IEP) meeting for a kindergarten child with a significant phonological impairment. At that time, her IEP specified only speech therapy. In the district, all diagnostics are done centrally. School-based clinicians are not to do any testing. If a child has been qualified for speech services only, technically the speech-language pathologist (SLP) is not to assess or treat language. This is problematic because a large percentage of children with speech–sound disorders also exhibit language impairment. Considering this, children with significant speech–sound disorders should regularly be assessed for language impairment. At the IEP meeting, the SLP mentioned deficits the child was showing in phonological awareness and print awareness skills. (In her graduate program she had appropriately been trained to incorporate phonological awareness/print awareness activities in phonological speech training). The educator in charge of the IEP meeting quickly stopped her, saying that it was not within the SLP’s scope of practice to address phonological awareness—that belonged to the reading teacher. If the SLP was concerned about the child’s phonological awareness skills, she should refer the child to the Student Assistance Team (SAT) for a response to intervention (RTI) process.
The response of this educator represents a lack of understanding of the relationships between speech–sound disorders, language impairment, and reading disability. Speech, language, and reading disabilities do not exist in isolation. It is unreasonable to attempt to intervene in only one area, ignoring related impairments. This article provides strong justification for attending to emergent literacy skills when working with children with language impairments.
Considerable research shows there are high levels of diagnostic comorbidity between language impairment (LI) and reading disability (Catts, Fey, Tomblin, & Zhang, 2002; Newbury et al., 2011; Pennington & Bishop, 2009). Children with LI are highly susceptible to reading problems at all stages of development, from the preschool years (Boudreau & Hedberg, 1999; Cabell, Justice, Zucker, & McGinty, 2009) and into primary and secondary schooling (Catts et al., 2002; Morgan, Farkas, & Maczuga, 2011). In general, one half of children identified with LI at kindergarten will have significant problems with reading (Catts et al., 2002). Given the likelihood of preschool children with LI later exhibiting reading difficulties, researchers suspect that susceptibility for reading disability might be apparent prior to the primary grades and the advent of formal reading instruction. This expectation is based on prior work identifying specific early literacy skills that are most strongly associated with future reading achievement; conversely, when children show lags in developing these literacy skills, their risk for future reading disability is heightened (Catts et al., 2002). There is convincing evidence linking children’s early literacy skills as measured at 3 and 4 years of age to their future reading achievement (e.g., Storch & Whitehurst, 2002) as well as risks for future reading disability (e.g., Catts, Fey, Zhang, & Tomblin, 2003). Considerable evidence points to early “code-based skills” as being particularly important to predicting future reading outcomes in word recognition/decoding (National Early Literacy Panel, 2008), these include (but are not limited to) print awareness, name-writing ability, phonological awareness, and alphabet knowledge.
The purpose of this study was to examine the early literacy skills among preschool-aged children with LI and to consider whether there are specific profiles of children who appear most susceptible for future reading disability. The sample includes children with both SLI (specific language impairment) and secondary LI, as a potentially important individual difference among children with LI. This study used a unique research design. Much of the work investigating early literacy skills among young children has involved variable-centered approaches. However, several recent studies have used person-centered methods to examine whether children can be profiled with respect to their skills and whether these profiles show longitudinal stability (Cabell, Justice, Konold, & McGinty, 2011; Cabell, Justice, Logan, & Konold, 2013). The goal of the person-centered approach is to determine whether there may be specific profiles of children that signal risk for future reading problems. This type of research can identify whether there are optimal groupings that are reliably differentiated within a given population (i.e., profiles). Person-centered research is distinguished from variable-centered research that often aims to predict longitudinal outcomes or document relations among variables. In this study, the authors build on recent findings showing that young children exhibit reliable profiles of early literacy skill, that these profiles show longitudinal stability (particularly children in relatively “poor” profiles), and that profiles can predict future reading achievement (Cabell et al., 2011; Cabell et al., 2013).
The aims of this research were as follows:
To determine (and describe) the number of profiles that best characterize the early literacy skills of young children with LI across indices of print awareness, name-writing ability, phonological awareness, and alphabet knowledge.
To examine the extent to which these profiles are associated with children’s oral language ability.
To examine the extent to which there are differences in the early literacy profiles for children with SLI and secondary LI.
Method
Participants
Participants were 218 children (73% male, 27% female) with IEPs specifying receipt of speech-language services; the children were enrolled in 83 publicly supported early childhood special education (ECSE) classrooms across a single state. The majority of the classrooms enrolled children with disabilities with their typical peers. The teachers were enrolled in a randomized controlled trial (RCT) involving implementation of a book-reading program in their classrooms.
The intent was to identify up to four children with LI per classroom for ongoing in-depth assessments (three assessments over a 2-year period). Teachers completed a screener for each child consisting of eight questions designed to identify children with language disabilities in their classrooms. Research staff examined all of the screeners available for a given classroom and identified those who appeared, on the basis of teacher report, to have a primary impairment of language (SLI), to have secondary LI (including children with language-related etiologies, such as Down syndrome and autism), or to have significant problems with language based on teacher report. The top-ranked children were assessed. The final sample of children (n = 218) was approximately 4 years of age (M = 4.2 years); 27%, 62%, and 11% of children were 3, 4, and 5 years of age, respectively. Ninety percent of children had IEPs designating provision of speech-language services in their schools, and an additional 5% of children were receiving speech-language services outside of school. The remaining 5% of children was considered by their teachers to have language skills lagging behind their peers but was not currently receiving speech-language services. About one fifth of the children had diagnosed developmental disabilities, including autism spectrum disorder (n = 12), cerebral palsy (n = 8), Down syndrome (n = 6), Stickler syndrome (n = 4), Tourette syndrome (n = 2), and apraxia (n = 2).
Procedures
Children were individually assessed in the fall of the year by trained research staff. The following measures from the larger battery administered to children were used to represent the constructs of print awareness, emergent writing ability, phonological awareness, alphabet knowledge, and oral language ability:
Preschool Word and Print Awareness (PWPA; Justice, Bowles, & Skibbe, 2006): The PWPA, a criterion-referenced task that was used to examine children’s knowledge of 14 print concepts. Children’s knowledge of these concepts is assessed in a storybook reading task.
The Name Writing Subtest of the Phonological Awareness Literacy Screening–PreKindergarten (PALS; Invernizzi, Sullivan, Meier, & Swank, 2004): Children are asked to draw a portrait of themselves and sign their name to it. The name production is scored on a 7-point scale.
The Phonological Awareness Subtest of the Test of Preschool Early Literacy (TOPEL; Lonigan, Wagner, Torgesen, & Rashotte, 2007), a norm-referenced standardized tool, was used to examine children’s phonological awareness. This subtest includes four sets of items to examine skills in both blending and segmenting across a range of segment sizes (e.g., syllables, rimes, phonemes).
The Alphabet Knowledge Subtest of PALS (Invernizzi et al., 2004), comprising both the upper-case tasks and lower-case tasks, was used to examine children’s knowledge of the alphabet. This criterion-referenced subtest includes a total of 52 items (all of the upper- and lower-case letters). Children are presented the individual letters in an array and are asked to produce the name of each letter.
Six core subtests of the Clinical Evaluation of Language Fundamentals–Preschool (CELF-P; Wiig, Secord, & Semel, 2004) were administered to children to represent their expressive and receptive language skills: Sentence Structure, Word Structure, Expressive Vocabulary, Concepts and Following Directions, Recalling Sentences, and Basic Concepts. An overall composite (Core Language Composite) was derived from the Sentence Structure, Word Structure, and Expressive Vocabulary subtests. A Receptive Language Composite was derived from the Sentence Structure, Concepts and Following Directions, and Basic Concepts subtests, and an Expressive Language Composite is derived from the Word Structure, Expressive Vocabulary, and Recalling Sentences subtests.
The Definitional Vocabulary subtest of the TOPEL (Lonigan et al., 2007) was used to represent expressive vocabulary skills. It requires children to provide the name of individually presented pictures (e.g., door), and then to provide additional information about that item (e.g., What is it for?). It captures the depth of children’s lexicon and thus provides additional information beyond that available in the vocabulary measure of the CELF.
Results
Profile Results
Profile 1: High risk (prevalence = 55%). The first group, comprising slightly more than one half of the sample, was labeled as high risk. This profile had mean scores on all four of the early literacy measures that were about 0.5 SD below the mean of the overall sample. As a group, these children appeared to know very few concepts about print, could not represent any letters in their names, performed very few phonological awareness tasks with accuracy, and knew about three letters of the 52 upper- and lower-case alphabet letters. Given the overall poor performance of this profile across multiple indices, these children can be characterized as being at significant risk for future reading difficulties, as all four indices are significant predictors of future reading achievement in both decoding and comprehension.
Profile 2: Moderate risk (prevalence = 14%). The second, comprising 14% of the sample, showed average performance on print awareness (relative to the group), low performance on the indices of name-writing ability and phonological awareness (about −0.5 SD of the mean), but quite strong performance on alphabet knowledge (accurately identifying 37 of the 52 upper- and lower-case letters). Although the children’s performance on two of the four indices suggests some susceptibility for future reading problems, the higher-than-average alphabetic knowledge may provide an important source of resilience.
Profile 3 (prevalence = 11%) and Profile 4 (prevalence = 19%): Low risk–high alphabet and low risk–average alphabet, respectively. The third and fourth, collectively representing about one third of children with LI in these early childhood special education settings, appear to show little susceptibility for future reading disability. Profile 3 (low risk–high alphabet) had relatively strong performance across all four early literacy indices, with scores in the +1 to +2 SD range. Profile 4 (low risk–average alphabet) had scores on three of the four indices (print awareness, name-writing ability, and phonological awareness) in the +.05 SD range. The primary difference between the two groups was on alphabet knowledge, with children in Profile 4 performing much lower than those in Profile, 3 but still in the average range in relation to the entire group, knowing about 12 of the 52 upper- and lower-case letters. Although the two groups have some distinctive differences in their profiles, the researchers view them as similar in having low susceptibility for future reading disability.
Profile Prediction From Language Impairment
Children in Profile 1 (high risk) had significantly poorer performance on all three language measures compared with children in Profiles 3 and 4 (the low-risk profiles).
Children in Profile 2 (moderate risk) had significantly poorer performance on all three language measures compared with children in Profile 4 (low risk).
Children in the high-risk and moderate-risk groups tended to exhibit standard scores of about 1 SD below those of children in the low-risk groups, thus indicating that children at greater susceptibility for future reading problems tend to exhibit poorer language skills compared with children who are least susceptible to future reading problems.
Profiles of Primary and Secondary LI
The majority of the secondary LI sample (56%) was in the high-risk Profile 1; only a few were in the other groups (14% Profile 2, 18% in Profile 3, and 12% in Profile 4).
The main difference with students with primary LI and secondary LI was that a higher percentage of children with secondary LI were classified in the high-risk group (56%) compared with 44% of the primary LI group.
Discussion
The results of the study show that four profiles of early literacy skills can be reliability identified. Half of the preschoolers with LI were at high risk for reading problems. They exhibited consistently poor performance across all four literacy measures compared with their peers. They knew only about three of 14 concepts about print, could not write their own names, and knew only three letters of the alphabet. Furthermore, children in the high-risk and moderate-risk groups had significantly poorer language skills than the two low-risk groups. The research provides justification for SLPs to include emergent literacy skills in their interventions.