Poster #72 - Prediction of Kindergarten Reading Skills: Unique Contributions of Preschool Writing and Emergent Literacy Skills

• In Event: 1-005 - Poster Session 01
  In Poster Session: PS 01 Section - Education, Schooling

Integrative Statement

Emergent literacy skills are predictive of later reading and can help identify children at-risk for reading difficulties (Lonigan, Schatschneider & Westberg, 2008). Oral language, phonological awareness (PA), and print knowledge (PK) have been identified as core emergent literacy skills (Lonigan, Burgess & Anthony, 2000; Whitehurst & Lonigan, 1998). Although reading and writing are highly correlated throughout early childhood and elementary school (Abbott et al., 2010; Berninger et al., 2006), few studies have examined if writing skills can aid in the early identification of reading difficulties. This study was designed to examine if preschool writing skills accounted for variance in kindergarten reading ability beyond variance accounted for by emergent literacy skills.
Preschool children (N = 252) were recruited as part of a larger longitudinal study concerning the development of literacy skills. Data were collected when children were in preschool and when they were in kindergarten. In preschool, children completed a battery of measures including Clinical Evaluation of Language Fundamentals (CELF), Test of Preschool Early Literacy (TOPEL), and Test of Early Reading (TERA) as well as three early writing tasks: Name writing, letter writing, and invented spelling. In kindergarten, children completed the reading subtests of the Woodcock-Johnson Test of Achievement (WJ) and the Test of Word Reading Efficiency (TOWRE).
Multi-level regression models--accounting for nesting at preschool and kindergarten--were conducted to examine the unique contribution of each preschool variable in predicting kindergarten reading outcomes. Age and IQ were included in each model as covariates. Together preschool writing and emergent literacy skills accounted for 45%, 29%, and 34% of the variance in the kindergarten outcomes of the Letter-Word Identification, Word Attack, and Passage Comprehension subtests of the WJ, respectively. For the TOWRE, 29% and 39% of the variance in Phonological Decoding and Sight Word Efficiency subtests, respectively, was accounted for. As seen in Tables 1 and 2, emergent literacy variables were consistently and uniquely associated with kindergarten reading outcomes. In the multi-level regressions, PK added between 3-7% of unique variance to all kindergarten reading outcomes. Invented spelling added between 3-6% unique variance to four of the five reading outcomes, but it contributed no unique variance to Phonological Decoding on the TOWRE. Neither name-writing nor letter-writing measures were uniquely predictive of any kindergarten reading outcome. Name writing acted as a suppressor variable for one outcome (see Table 1).
These results indicate that preschool invented spelling offers a small but unique amount of predictive variance for kindergarten reading outcomes beyond the ability of emergent literacy skills to predict kindergarten reading outcomes. It is not clear why invented spelling contributed unique variance whereas other preschool writing skills did not. It may be that invented spelling requires the integration of multiple literacy-related skills (e.g. PA, letter writing, print knowledge). Results highlight the potential utility of including invented spelling as part of a screening for reading difficulties. Although our results indicate that only one preschool writing measure was uniquely predictive of kindergarten reading outcomes, future research should examine if preschool writing is predictive of later writing outcomes.

Authors

• Eric Hand, Florida State University
  Presenting Author

• Brittany M Morris, Florida State University
  Non-Presenting Author

• Christopher Lonigan, Florida State University
  Non-Presenting Author