|Title:||The Effects of Leveled Literacy Intervention on Secondary Students in the Oakland Unified School District|
|Principal Investigator:||Wing, Jean||Awardee:||Oakland Unified School District|
|Program:||Low-Cost, Short-Duration Evaluation of Education Interventions [Program Details]|
|Award Period:||2 years (7/1/2016-6/30/2018)||Award Amount:||$250,000|
|Type:||Other Goal||Award Number:||R305L160008|
Co-Principal Investigator: Eric Isenberg (Mathematica Policy Research)
Partner Institutions: Oakland Unified School District, Mathematica Policy Research
Purpose: Struggling readers in secondary schools may end up graduating without the literacy skills needed for postsecondary education and the labor market. These students may be several years below the reading level for their grade due to lags that began in primary school or because they are English learners. The Leveled Literacy Intervention (LLI) is a short-term, intensive intervention system designed to help teachers provide daily, small-group instruction for students who are not achieving grade-level expectations in reading. The Oakland Unified School District (OUSD) is already using LLI in primary grades, and will partner with Mathematica Policy Research to study the effectiveness of LLI with older, struggling readers and long-term EL students in grades 6–9.
Project Activities: Struggling readers at 11 OUSD schools will be randomly assigned to receive the regular English Language Arts course or the regular course plus LLI. The partnership team will estimate the impacts of LLI on reading comprehension by comparing the Scholastic Reading Inventory scores of students assigned to these two groups.
Products: OUSD and Mathematica will disseminate the results of the project through a briefing, policy brief, conference presentations, and journal articles.
Setting: The project will take place in the Oakland Unified School District California.
Population/Sample: The partnership team will identify approximately 1,200 struggling readers in grades 6–9 at 11 schools located in low-income neighborhoods. The majority of these students are expected to be eligible for free or reduced-price lunch, and to be either African American or Latino. About half of these students are expected to be classified as English learners.
Intervention: The Fountas & Pinnell (F&P) Leveled Literacy Intervention (LLI) was originally developed for the primary grades and was adapted for secondary grade students who are several years behind in reading. Under LLI, teachers provide daily small-group instruction using texts in a variety of genres with lessons focused on intentional vocabulary development, academic language development, and student discussion. Oakland Unified School District currently uses LLI in its primary grades and in the 2015–2016 school year started introducing LLI in grades 6–9. Lessons are provided as a separate class or during a pull-out from a non-English class. Teachers select age-appropriate texts of progressing difficulty and match the texts to each student's reading level. The partnership will assess students using the F&P Benchmark Assessment System that matches student reading level to the LLI text-level scale. Students scoring below grade label are eligible for the intervention. Three-to-five eligible students with similar F&P scores receive a daily lesson of 30 to 45 minutes from a teacher until they reach preset growth goals. Students normally remain in the program for 20–24 weeks. Teachers will receive initial 2-day training in LLI from the publisher with additional professional development available during the year, as well as access to reference and lesson materials. Oakland also hired two literacy specialists to support the LLI teachers. Teachers will monitor students' reading level, select appropriate texts, and support independent reading and comprehension.
Research Design and Methods: The partnership team will conduct a randomized controlled trial. Approximately 1,200 students in 11 schools will be identified as eligible for LLI. From this group, the team will randomly assign about 160 students to receive LLI and 890 to receive the regular English Language Arts class without LLI. Another 150 students will be randomly assigned to a non-research group that can receive LLI after other treatment-group students complete or leave the program. (The non-research group will not be included in the analysis but allows schools to provide the maximum number of students with LLI. ) The probability of assignment to LLI will vary by school due to differences in the number of eligible students and LLI teachers. Researchers will stratify student assignment by school and grade to allow for weighting of results. They will also stratify student assignment by English learner status to examine impacts on EL students. The Scholastic Reading Inventory (SRI) test is given to students three times a year and will be used in an intent-to-treat analysis to check baseline equivalence in reading as well as determine impacts at mid-year (January) and at the end of year (May). The team will complete a treatment-on-the-treated analysis if group crossover is found.
Control Condition: Students will receive the regular English instruction provided by their school and may receive any other supports provided for students reading below grade level.
Key Measures: The partnership will assess students for eligibility in the intervention using the F&P Benchmark Assessment System that matches student reading level to the LLI text level scale. The team will use the SRI test which measures student literacy using both literary and expository texts in such areas as drawing details from passages, drawing conclusions, and making comparisons and generalizations. A computer algorithm adapts test items to the reading level of the student.
Data Analytic Strategy: The partnership team will estimate the overall impact of LLI by using a regression-based approach. The team will adjust for any differences in observable baseline characteristics. The team will also check overall and differential attrition and correct any bias by creating a set of nonresponse weights if needed.