Cross-Linguistic Validity of Developmental Written Language

• In Event: Written Language Studies Across Culture

Abstract

The goal of the current study is to test the cross-age (from the lower graders to the higher graders), cross-linguistic (i.e., from English to Chinese) and cross-passage (i.e, two testing occasions) validity of Wagner et al. (2011) five-factor written language assessment model, in which it includes (a) macro-organization of idea, (b) number and diversity of words used, (c) mean length of T-unit and syntactic density, (d) spelling and punctuation, (e) handwriting fluency.
To achieve this goal, we collected writing data among 400 Chinese Grade 4, and Grade 7 students who are writing in Chinese as their first language. Two 10-minute writing samples were coded following the rubrics used in Wagner et al (2011) along with students’ handwriting fluency (including stroke-copying, and character-copying tasks). In contrast to the model established among the English population, we examined 1) the partial measurement invariance between passage A and passage B for Grade 3 and Grade 7 respectively; 2) the measurement invariance between grades 3 and 7 on passage A and passage B respectively using multi-group CFA.
The results from the current study support our theoretical framework that the latent factors were defined by the corresponding variables: Macro-organization was defined by the Order and Topic + Number of Key Elements; Complexity was defined by Mean length of T-unit and Density; Productivity was defined by the Total Number of Words (NTW) and Number of Different Words (NDW); Spelling and Punctuation (i.e., mechanical errors) was defined by the Number of Phonological Errors, the Number of Orthographic Error, and the Number of Period Errors; Handwriting fluency was defined by the Stroke Printing Fluency and Sentence Copying Fluency. This five factor model has been established in Wagner et al’s (2011) English population, and has been replicated in the Chinese population with a monior adaptation.
In addition, the results from this current study confirm that the handwriting fluency plays a more salient role in kids’ written production and macro-organization for lower grade (e.g., Grade 4), than that for higher grade (e.g., Grade 7). Because multiple measures and factors were used in the current study, we were able to point to the specific comparison between measure correlations across grades and between two written samples (i.e., two testing occasions).
This study is among the first to examine the generalizable feature of cross-language, cross-age, and cross-testing occasions of written-language model. Results of this study suggested that we were measuring the same latent variables across passages, across different grades (lower and higher grades), and across two testing occasions. We have found the common features of written language assessment model, which provide us with a good rationale to 1) focus on writing-skill-related training intervention, 2) establish a good and reasonable assessment model to measure the writing-related skills according to our five-factor model. Perhaps, in the future this study would shed light on what language educators should do to optimize students’ written-language skills.

Authors

• Qun Guan, East China Normal Uniiversity

• Rick Wagner, Florida State University

• Feifei Ye, University of Pittsburgh

• Wanjin Meng, China National Institute for Educational Research