Shanahan, Emma2023-09-192023-09-192023https://hdl.handle.net/11299/257064University of Minnesota Ph.D. dissertation. 2023. Major: Educational Psychology. Advisor: Kristen McMaster. 1 computer file (PDF); xii, 154 pages.Difficulties in writing can emerge as early as preschool, and often coincide with developing difficulties in reading (Berninger et al., 1997; Graham & Santangelo, 2014; Graham et al., 2020), as reading and writing are fundamentally connected skills (e.g., Fitzgerald & Shanahan, 2000). Writing instruction in general has had positive effects on reading outcomes of students across grade levels (Graham & Hebert, 2011). However, more research is needed to examine whether writing instruction can support the reading skills of elementary students with writing difficulties (Graham, 2020). Data-based instruction (DBI) in writing, which includes research-based writing instruction activities, frequent progress monitoring using curriculum-based measures (CBM) in writing, and data-based decision-making (DBDM) to individualize instruction, has been found to have promising effects on the writing outcomes of students with significant writing difficulties (McMaster et al., 2020), and may similarly support reading. Thus, the purpose of this dissertation was to examine whether DBI in writing can play a positive role in the foundational reading development of elementary students who benefit from intensive writing support. The current study used data from one cohort of participants in a multi-cohort randomized control trial evaluating the effects of DBI professional development on teachers’ use of DBI in writing and students’ writing outcomes. Participants in this study included 42 teachers (19 treatment, 23 control) and their 105 students with significant writing difficulties in Grades 1 to 5 (46 treatment, 59 control). Treatment teachers implemented writing instruction, collected CBM-writing data, and engaged in DBDM with fidelity while receiving ongoing, collaborative support via learning modules and twice-monthly coaching. Treatment students received an average of 37.3 hours of DBI in writing across 20 weeks of study participation. Depending on teachers’ assessment of students’ needs, teachers typically targeted spelling, but also taught handwriting and/or text generation. Pretest correlations indicated that the two reading outcomes, letter sound knowledge and decodable word reading, as measured by FastBridge Letter Sounds correct letter sounds per minute (LS CLSPM) and Decodable Words correct words per minute (DW CWPM), were associated with writing skills spanning from spelling to written expression. Hierarchical linear models controlling for the effect of teacher intercept indicated that DBI in writing did not have a positive effect on LS CLSPM. DBI did, however, have a significant positive effect on log-transformed DW CWPM after controlling for log-transformed pretest scores, meaning that DBI in writing had the strongest effect on the reading of students with higher initial decoding skills. Future research should investigate the effects of more specific letter sound writing interventions on letter sound knowledge and examine whether and why Matthew effects (Stanovich, 1986) may occur in DBI. Implications for teachers’ integration of reading and writing interventions as well as next steps for system-level writing assessment are discussed.endata-based instructionelementaryliteracyreadingspecial educationwritingThesis or Dissertation