|Title:||Beyond Triage: A Randomized Experiment in Sustained Pre-College Advising|
|Principal Investigator:||Bettinger, Eric||Awardee:||Stanford University|
|Program:||Evaluation of State and Local Education Programs and Policies [Program Details]|
|Award Period:||4 years (7/01/2013-6/30/2017)||Award Amount:||$1,705,221|
|Goal:||Efficacy and Replication||Award Number:||R305E130009|
Co-Principal Investigator: Anthony Antonio
Purpose: Well-qualified students are currently discouraged from pursuing higher education because of avoidable barriers such as a lack of information about college admissions and financial aid. College advising is one of the key mechanisms used to aid students as they navigate the college access process. There is a large network of college access programs that provides assistance to underserved students; however, the quality of college access programs can vary within the same district. This project will evaluate a state-sponsored college access program called Advise Texas (Advise TX) run by the Texas Higher Education Coordinating Board (THECB). The primary goal of Advise TX is provide a uniform and readily available college access program that will raise the rates of college enrollment and completion among low-income, first-generation-college, and underrepresented high school students in Texas.
Project Activities: The project will follow three cohorts of graduating seniors from 112 Texas high schools to evaluate the impact of Advise TX on student college-going. Thirty-six high schools were randomly assigned to receive Advise TX and 76 were assigned to the control group. The project will collect and compare intermediate outcomes towards college-going (e.g., completing an application to attend college, applying for financial aid), college-going (enrollment and persistence), and changes in the high-schools' college-going culture.
Products: The products of this project will be evidence of the efficacy of Advise TX for increasing high school students' college-going. The evidence will be directly provided to the THECB through the participation of department personnel on the project and to the research community, education practitioners, and policymakers through presentations and peer-reviewed publications.
Setting: This project will take place in approximately 100 Texas high schools.
Sample: In the Spring of 2010, the THECB identified 418 high schools that met their initial criteria to receive Advise TX (the criteria were based on percent free/reduced lunch participation, percent graduating students attending college within 1 year, and percent of students experiencing a college-prep curriculum). Two hundred thirty-seven of these schools applied to take part in Advice TX. A selection committee examined these schools, excluded 50, and accepted 84. The remaining 112 schools were divided into 23 regions and 36 schools were randomly selected among these 23 regions to receive Advise TX. The remaining 76 schools were assigned to the control group. The sample includes high school students at the 36 treatment schools and the 76 control schools for the 2011–12, 2012–13, and 2013–14 school years.
Intervention: The Texas Higher Education Coordinating Board (THECB) adopted Advise TX in the 2011–2012 school year. Through Advise TX, an adviser is placed in a high school to provide one-on-one student mentoring and group counseling to students and parents on applying to college and applying for financial aid. The adviser, a recent graduate from one of five Texas universities, receives ongoing training to be an adviser from his/her university. In addition, the adviser works with the high school staff to coordinate college-going activities in the schools, and seeks to create a college-going culture. Advisers serve all students at the school but focus on low-income and potential first-generation college students who often lack information on how to apply and misperceive college costs and aid availability. Advise TX's central features include: (1) a near-peer (no more than 6 years after high school graduation) counselor for high school students who has graduated from a Texas university, (2) the adviser remains linked to their university and receives ongoing training from the university, (3) the adviser serves the whole high school and attempts to promote a college-going culture, (4) the adviser focuses on finding the best matched college for students.
Research Design and Methods: A randomized experiment is used based on the random assignment of 36 schools to Advise TX and 76 schools to the control condition. School-level outcomes will be compared for the graduating classes of 2012, 2013, and 2014. In addition, at 8 larger treatment schools, 50 students in each school who are on the margin of college-going (based on their GPA) are randomly chosen to meet with the adviser on a monthly basis. Student-level outcomes for these students will be compared to those of the other students at the 8 schools who may receive the treatment but not at such an intensive level.
Control Condition: The comparison group includes the 76 Texas high schools that applied to implement Advise TX and met the requirements to receive Advise TX but were not randomly selected to receive the program.
Key Measures: Three outcomes will be addressed: college enrollment and college-going, intermediate outcomes to college-going, and creating a high school culture of college-going. College enrollments and completions will be measured using initial enrollment in college, full-time versus part-time enrollment, 2-year or 4-year institution, students' persistence through their first year of college, students' transfer rates from 2- to 4-year colleges, and college completion for the first cohort of high school graduates. Intermediate outcomes to college-going will be measured include course selection, grades, completion of college applications, filing of financial aid forms, and participation in college entrance exams. College-going culture will be measured using the proportion of students taking rigorous college preparation courses in both treatment and control schools and responses to items on a student survey regarding postsecondary aspirations as well as engagement in college-prep activities. Fidelity of implementation measures will be drawn from three sources: an administrator survey, a student survey, and monthly reports from advisers. Data on comparison group practice will be drawn from the administrator survey and the student survey.
Data Analytic Strategy: For the school-level analysis, multivariate regression will be used to further control for region, time, and school-level covariates (e.g., school size, the school's selection score, the percent of students attempting various levels of curriculum strength, schools' college enrollments for prior cohorts, the percent of the school on free/reduced lunch, the percent of the school from various minority groups, and the availability of other college-access programs in the school). Subgroup analysis based on these school-level covariates will also be done. For the student-level analysis, multivariate analysis will be used to control for the school, student characteristics.