State Data Analysis

Skip to content Skip to main navigation
Teachers College, Columbia University
Printer-friendly Version
Teachers College, Columbia University Logo

State Data Analysis

NCPR is conducting studies using state administrative data on dual enrollment, postsecondary remediation, and financial aid. These analyses have focused on data from two main states: Florida and Tennessee.  Bridget Terry Long of Harvard University is leading this work.

Florida Remediation Study

For this project, we use an administrative dataset obtained from the Florida Department of Education to estimate the impact of developmental/remedial courses on community colleges student outcomes, such as passing college-level courses, retention, degree attainment, transfer to a four-year university, and credits earned, and on overall institutional performance. The dataset focuses on the universe of first-time students who enrolled at any of the 28 Florida community colleges from fall 1997 to fall 2000 and sought at least an associate (two-year) degree. Students’ enrollment is tracked term-by-term for a total of six years for each cohort.

The evaluation design take advantage of Florida's statewide placement policy, implemented exclusively at community colleges, to estimate the impact of this program using a quasi-experimental regression discontinuity design. In addition, we use different statistical techniques to deal with two methodological threats to the regression discontinuity design: noncompliance and endogenous sorting around the cutoff.

Tennessee Remediation Study

NCPR is conducting analysis of the effects of remediation in the state of Tennessee using data from the Tennessee Higher Education Commission (THEC) and the Tennessee Board of Regents (TBR). The sample covers all students in the public higher education system and includes data from admissions, transcripts, and remedial/developmental placement tests. Because of the state’s use of one clear policy to determine placement in remedial, developmental, or college-level courses (students could be assigned to one of fours levels in mathematics and one of three levels in reading and writing), researchers are able to use a regression discontinuity research design to estimate the causal impact of such courses on student outcomes. Outcomes examined include persistence, degree completion, the number of total and college-level credits completed, and college GPA.

Florida Dual Enrollment Study

NCPR is currently undergoing an analysis of high school students who have participated in dual enrollment in Florida. In the past year, NCPR has examined summary statistics of the data and looked at how dual enrollment participation varies across school districts. Researchers have compared participation to high school GPA to explore whether schools are using the 3.0 and 2.0 GPA eligibility requirements. NCPR is now expanding the original dataset to include more information on high school grades, test scores, and details on the location of students’ dual enrollment courses. Using the new data, NCPR researchers will estimate the strength of the causal relationship between dual enrollment and those outcomes.


Associated Publications:

Does Remediation Work for All Students? How the Effects of Postsecondary Remedial and Developmental Courses Vary by Level of Academic Preparation (An NCPR Working Paper). By: Angela Boatman and Bridget Terry Long (September 2010).

Using longitudinal data from Tennessee, this study estimates the effects of placement into varying levels of mathematics, reading, and writing courses for students attending public four- and two-year colleges and universities. This is possible due to the state’s multi-tiered system, in which students could be assigned into one of four levels of mathematics and one of three levels of reading and writing courses. Using regression discontinuity (RD) techniques, the authors provide causal estimates of the effects of placement on a number of student outcomes, including persistence, degree completion, the number of total and college-level credits completed, and college GPA. The results suggest that remedial and developmental courses do differ in their impact by the level of student preparation. Similar to other research, the authors find negative effects for students on the margin of needing any remediation. However, at the other end of the academic ability spectrum, the negative effects of remediation were much smaller, and occasionally the effects were positive. These results suggest that remedial and developmental courses help or hinder students differently depending on their level of academic preparedness. Download the PDF | Download the Brief Version

The Impact of Postsecondary Remediation Using a Regression Discontinuity Approach: Addressing Endogenous Sorting and Noncompliance (An NCPR Working Paper)By: Juan Carlos Calcagno and Bridget Terry Long (April 2008). [This paper has also been released as NBER Working Paper 14194 (July 2008).]

This paper reports findings from a study that uses a detailed dataset and a regression discontinuity design to identify the causal effect of remediation on the educational outcomes of nearly 100,000 college students in Florida. The paper also discusses concerns about endogenous sorting around the policy cutoff, which poses a threat to the assumptions of the regression discontinuity model in multiple research contexts. Download the PDF | Download the Brief Version