Skip Navigation
Funding Opportunities | Search Funded Research Grants and Contracts

IES Grant

Title: Doubling Up? Understanding the Long-Term Effects of Ninth-grade Algebra Reform on College Persistence and Graduation
Center: NCER Year: 2017
Principal Investigator: Nomi, Takako Awardee: Saint Louis University
Program: Postsecondary and Adult Education      [Program Details]
Award Period: 2 years (07/01/2017-06/30/2019) Award Amount: $559,485
Goal: Efficacy and Replication Award Number: R305A170602

Co-Principal Investigator: Stephen Raudenbush (University of Chicago)

Purpose: In this project, the research team will assess whether the "double-dose algebra" policy implemented by Chicago Public Schools (CPS) between 2003 and 2004 led to increases in college attainment and degree completion for students assigned to extra math instruction as a result of the policy. This project builds on three previous IES-funded projects, all of which found positive effects associated with the double dose policy. An initial Efficacy and Replication project found that 9th-grade algebra grades, algebra pass rates, and standardized test scores were higher for students who received the extra algebra instruction than those who did not. A second Efficacy and Replication project found that students who received the extra algebra instruction during 9th grade had higher math ACT scores in 11th grade, were more likely to graduate from high school, and were more likely to enroll in college. A subsequent Exploration project found significant variation in policy impacts by the extent of skill-based sorting in 9th-grade math classes, with students in mixed-skill classrooms performing better on 9th-grade standardized math tests than students segregated into low-skill classrooms.

The current project extends the previous projects by assessing whether the double-dose algebra policy improves college persistence and degree attainment, whether policy impacts are mediated by skill-based sorting in math classrooms and the selectivity of colleges that students attend, and whether positive impacts extend to students who entered high school with test scores significantly below the national median. This project addresses the need for research evidence on whether successful remediation efforts in early high school can set in motion a positive sequence of events leading to improved postsecondary outcomes. Project findings will inform policymakers and education agencies across the country who are considering the long-term impacts of high school course programming.

Project Activities: This follow-up study extends the previous projects by observing students for six years after high school graduation. During the first year of the project, the team will merge six years of data from the National Student Clearinghouse (NSC) to an extant database of measures constructed from data provided by CPS, code measures and assess data quality. During the second year of the project, the team will conduct data analysis, run robustness checks on their results, and prepare three manuscripts for submission to peer-reviewed journals.

Products: Researchers will produce evidence of the efficacy of the double-dose policy for improving the postsecondary outcomes of students who enter high school with below-average math skills. The researchers will produce three peer-reviewed papers, one focusing on average postsecondary impacts of the double-dose policy, a second focusing on variation in the policy impacts across schools and mediating mechanisms through skill-based sorting, and a third focusing on policy impacts for students who enter high school with substantially below-average math skills.

Structured Abstract

Setting: Participating schools' are neighborhood high schools located throughout the Chicago Public Schools district.

Sample: The sample includes four cohorts totaling approximately 61,000 students who entered one of 62 Chicago neighborhood high schools between fall 2001 and fall 2004. The students were largely from low-income and minority backgrounds, with approximately 85 percent eligible for free or reduced-price lunch programs and from either African American or Latino families.

Intervention: The double-dose policy combines an additional algebra support course with the regular ninth-grade algebra course, resulting in students taking two periods of math. For all first-time ninth-grade students who scored below the national median on the eighth-grade Iowa Test of Basic Skills math exam, the district required double-dose algebra during the policy implementation period and increased math instructional time for low-achieving students. However, high schools varied substantially in their approaches to policy implementation, with some segregating low-achieving students into the same algebra courses while others preserved their pre-policy practices of distributing low-achieving students across course sections.

Research Design and Methods: To assess whether double dose algebra improved postsecondary outcomes for students close to the national median for pre-high school algebra test scores, researchers will use a regression discontinuity design to estimate the impact of double-dose algebra as compared to regular algebra during the policy implementation years. To assess whether double dose algebra improved outcomes for students who entered high school with lower math skill levels, researchers will use an interrupted time series design to assess whether these students' scores exceeded the general trend for similar students at their schools during the pre-policy years.

Control Condition: The control condition is regular (single-period) ninth-grade algebra with no additional math instruction. This condition differed somewhat between pre-policy and post-policy periods because the double-dose algebra policy induced skill-based sorting in a significant portion of schools. For students with incoming skills just above the eligibility cut-point, classroom peer skill levels of regular algebra classes were, on average, higher post-policy than pre-policy.

Key Measures: Researchers will rely on previously-collected administrative data from Chicago Public Schools that includes demographic characteristics, end-of-year test scores, and high school course transcripts for all students in the four study cohorts. They will construct college persistence and degree completion (within 6 years) measures including earning a certificate, an Associate's degree, a Bachelor's degree, and earning a certificate or degree in a STEM field, from administrative data obtained from NSC. The researchers will also use the NSC data on college sector to explore whether students' postsecondary outcomes are mediated by enrollment in a two- or four-year institution.

Data Analytic Strategy: Researchers will implement the main regression discontinuity model, using a random effects framework. This modeling strategy compares outcomes for students just above and below the test score cutoff for assignment to double-dose algebra, using a two-level model that allows for variation in policy impact across schools. Schools' compliance with the cut-score-based course assignment rule as well as the extent of skill-based sorting within their classrooms are also allowed to vary at the second (school) level of the model. Researchers will employ a second two-level model to implement the interrupted time-series design in order to estimate the policy impact for students who entered high school with test scores significantly below the cutoff. Using random parameters at the school level, researchers will measure a school-specific "jump" in double-dose enrollment for this group of students at the time of policy implementation and use this to estimate the policy impact on the outcome (i.e., the school-specific "jump" in the outcome).

Related IES Projects:
Comprehensive Evaluation of the Effects of District-Wide High School Curriculum Reform on Academic Achievement and Attainment in Chicago (R305R060059)
Doubling Up? The Impact of Remedial Algebra on Students' Long-Run Outcomes (R305A120466)
Curricular Reform and Classroom Peer Ability: School-Specific and Citywide Effects (R305A130703).