A Data‐Driven Approach to Text Adaptation in Teaching Material Preparation: Design, Implementation, and Teacher Professional Development

Syntactic Complexity in Adapted Teaching Materials: Differences Among Grade Levels and Implications for Benchmarking

“…Differences with the greatest effect sizes were found in the lexical indices used, particularly those pertaining to lexical diversity, word frequency, and word familiarity. This finding provided empirical evidence for the observation that lexical modifications tend to be the major focus in text adaption practices (Jin & Lu, 2018). They further noted that some lexical substitutions in fact made the texts more difficult to comprehend, echoing Crossley et al's (2007) finding that variable and sometimes unexpected linguistic modifications are not uncommon among individual textbook authors and language teachers.…”

“…We believe that future research on benchmarks of syntactic complexity (or of linguistic complexity in general) will benefit from a broader view on the text adaptation enterprise that includes not only source texts, but also teachers’ adaptation practices and the pedagogical effectiveness of adapted texts in the research agenda. Meanwhile, as Jin and Lu () pointed out, second and foreign language teachers who engage in text adaptation will benefit from professional development projects designed to help improve their data literacy, including both knowledge of how language data works (e.g., how syntactic complexity benchmarks are established) and skills for utilizing such data (Mandinach & Gummer, ; Zareva, ), and to foster a stronger awareness of the close connection between research and practice (Farley–Ripple et al., ).…”

“…A plethora of vocabulary lists derived from large text corpora, including corpora of teaching materials and textbooks, are now available, which can serve as benchmarks for evaluating the lexical complexity of texts and for providing feedback to teachers regarding what lexical items to consider removing or replacing when adapting a text (e.g., Brezina & Gablasova, ; Gardner & Davies, ; Jin & Lu, ; Schmitt & Schmitt, ). The importance of including syntactic complexity as an explicit and distinct component in text complexity models has been well argued for (Frantz et al., ), and tools for evaluating text complexity do indeed all incorporate measures for this construct (Graesser et al., ; Sheehan et al., ).…”

“…First, both researchers and teachers have recognized that quantitative assessment of text readability and complexity would be more informative to teachers when standard points of reference or benchmarks of relevant dimensions of text complexity are available (e.g., Allen, ; Jin & Lu, ). In the absence of such benchmarks, it is difficult for teachers to make full sense of the quantitative assessment provided.…”

“…Relatedly, the role of teachers’ access to feedback based on benchmarked measures in successful text adaptation has also been recognized (Allen, ), but to date research on this role has been largely limited to feedback pertaining to vocabulary (Jin & Lu, ; Nunan, ) and measures of text readability (Greenfield, ). In a recent study, Jin and Lu () reported that teachers responded positively to an online text adaptation system that provided feedback in the form of a ranked list of low‐frequency or off‐list words to be removed or replaced based on carefully selected vocabulary lists for benchmarking and real‐time reassessment of the lexical difficulty of the text being adapted, after each change. Most computational systems for automatic text readability or complexity analysis, however, provide little feedback based on benchmarks of the measures they integrate to guide teachers through the text adaptation process (Crossley, Greenfield, & McNamara, ).…”