Skip Navigation
Funding Opportunities | Search Funded Research Grants and Contracts

IES Grant

Title: Academic Trajectories and Policies to Narrow Achievement Gaps in San Diego
Center: NCER Year: 2013
Principal Investigator: Betts, Julian Awardee: University of California, San Diego
Program: Researcher-Practitioner Partnerships in Education Research      [Program Details]
Award Period: 2 years (7/1/13-6/30/15) Award Amount: $383,187
Goal: Other Goal Award Number: R305H130059
Description:

Co-Principal Investigators: Peter Bell (SDUSD); Dina Policar (SDUSD); Ronald Rode (SDUSD)

Partner: The San Diego Unified School District (SDUSD); Partnership Name: San Diego Education Research Alliance at UCSD (SanDERA).

Education Issue: Research on early warning and on-track indicators finds that students get "off-track" for succeeding and graduating in high school as early as elementary school. This research indicates that students send strong distress signals for years before they leave school. Yet, most school systems do not have estimation strategies that allow them to make use of the data that they already have to determine which students are "on-track" to succeed in school. Knowing which students are off-track is a crucial first step toward providing these students with targeted support that they need to return to on-track status.

Partnership Significance and Goal: The goal of this project is to enhance the district's access to high-quality, relevant information by producing ideal trajectories—what it means to be "on track" at every stage of in ideal student's career—as well as accurate on-track indicators of individual students' on-track statuses. Through the dissemination work of the partnership, the ideal trajectories will be shared with school-site staff, parents, and the broader community. In addition, the on-track indicators—shared only with school-site staff—will help teachers and principals identify students' needs in something close to "real time" rather than from, for example, a retrospective analysis of the records of those who have already dropped out. Tools for school-site staff will include secure online dashboard displays that gauge individual students' and entire classrooms' progress toward short-term and longer term education milestones. For district staff and school leaders, statistical models will estimate which schools are over- and under-performing in moving students to accomplish specific goals. Ideal trajectories will be developed for English learners and students with disabilities. After the 2-year grant period, the ideal trajectories and on-track indicators will guide the district's plans for interventions that will target students most in need of assistance, at the earliest possible grade levels.

Partners and Partnership Activities: The project activities will further solidify the SanDERA partnership which has been in place since 2010. To enhance research activities, the grant will fund purchasing of software for the San Diego Unified School District (SDUSD) partner, and research partners from the University of California—San Diego (UCSD) will share indicator coding and programming expertise with staff from SDUSD. To enhance collaboration, the partnership will convene meetings at several levels. The Executive Committee—comprising principal investigators from UCSD and SDUSD—will meet on a bi-weekly basis. An Advisory Committee with over 20 members will meet quarterly. The partnership will convene focus groups of school-site staff and district administrators to address problem areas in students' trajectories, as well as to understand practices that appear to be working well. The partnership will share findings and seek input from the broader community at a public forum that will be held during year two. The partnership will also keep the broader community abreast of activities and findings through the SDUSD website and communication with local press coverage.

Setting: This project will take place at SDUSD, which is the second largest district in California. It is a highly diverse district, with large proportions of African-American, Asian/Pacific Islander, Hispanic, and White students.

Population/Sample: The trajectories project will compute on-track indicators for all students with available prior-year data: approximately 85,000 students attending all 170 schools in the district.

Initial Analysis: The partnership will build upon prior research on early warning and on-track indicators, as well as prior analytical work conducted by the research team, to arrive at indicators that will promote engagement in teaching and learning. Drawing on data that the district already collects, the partnership will illustrate ideal trajectories and compute on-track indicators from small sets of performance measures that are highly predictive of later school success. The ideal trajectories will demonstrate the performance of an on-track student at each stage of elementary, middle, and high school. The on-track indicators—computed for students in all grades for which data are available—will use early-grade measures to accurately predict whether or not a student is likely to reach key later-grade milestones such passing ninth grade, completing high school, or applying for college. Estimations from these streamlined models will yield forecasts that the partnership team can easily convey to teachers and parents. For example, by estimating the relation between GPA in prior grades and Algebra I passage in eighth grade, researchers will be able to forecast the probability that a student will pass Algebra I. The partnership team will decide the probability level that is necessary for a student to be considered on-track to pass Algebra I. Because of the simplicity of the estimation strategy, forecasts can be easily conveyed to teachers and parents via tables and graphs.

The partnership places a priority on conveying accurate trajectories and indicators to teachers, principals, and district staff. In addition to providing educators with indicators produced from one and two predictors, researchers will make available indicators produced from multivariate regressions, as well as classroom summaries so that teachers will know the numbers of students in their classes who are forecast to cross specific thresholds. To inform educators about how well entire schools are doing, researchers will estimate models of school performance that account for prior achievement of students within them, as well as neighborhood characteristics.


Back