Building On-Track Indicators for High School Graduation and College Readiness: Evidence from New York City

Issue/Topic: Data-Driven Improvement; High School--College Readiness
Author(s): Kemple, James; Stephenson, Nickisha; Segeritz, Michael
Organization(s): New York University
Publication: Journal of Education for Students Placed at Risk (JESPAR)
Published On: 2013

Students' engagement and performance in their first year of high school offer strong signals about their prospects of earning a diploma four years later. Districts can use performance measures to construct on-track indicators to identify needs of specific students who are at risk of dropping out. This study analyzes ten years of longitudinal data to test the reliability and validity of New York City's graduation and college readiness indicators.

To analyze several on-track indicators that predict the likelihood of graduating with a New York State Regents Diploma; to track changes in graduation rates and in the rates at which students are on track for graduation at the end of their ninth-grade year; and, to identify student, school, and system-level characteristics associated with students' readiness for college or work.


Why Are On-Track Indicators Important?

  • Being on track at the end of the ninth grade can supersede prior performance as a crucial signal about students' prospects for graduating with a Regents diploma.
  • While students on track at the end of ninth grade are more likely to graduate on time, gender and racial-ethnic gaps are still pervasive. Despite these gaps, however, the on-track indicator remains a powerful mechanism for identifying students who are most likely to need help meeting graduation requirements. 

Which On-Track Model Works Best?

  • The current New York City Department of Education indicator (i.e., earning 10 or more credits in the ninth-grade year) is a reasonably accurate predictor of the likelihood that a student will graduate with a Regents diploma within four years of entering high school: 73.9%.
  • The model that provides the best balance between the correct on- and off-track predictions (i.e., 81.8% and 75.7%, respectively) measures attendance rates, the number of credits earned, and the number of Regents exams passed. This model would require teachers and administrators to make subjective evaluations about students' progress.
  • Earning 10 or more credits and passing at least one Regents exam has the best fit because it does not permit subjective evaluations or variable value combinations. Using this model would increase the on-track prediction rate from 73.9% to 85.2%.

How Predictive Are On-Track Indicators?

  • Between 2001 and 2011, the model correctly predicted graduation rates between 77 and 82% of the time.
  • During the same time period, the off-track prediction rate decreased substantially, meaning more students closed their ninth-grade credit and exam completion gaps than the model estimated. The researchers speculate that targeted efforts to identify and assist students who fell behind has improved graduation rates for the off-track subgroup.
  • The on-track indicators are helpful because they signal the percentage of students who are at-risk of dropping out. For the 350+ high schools in the longitudinal sample, about one quarter had 70% or more students off-track for graduation and another quarter had 70% or more students on-track. The majority of high schools fell somewhere in between.

Policy Implications/Recommendations:
  • Models that require teachers and administrators to make subjective judgments on whether students are on track for graduation can complicate understanding of the factors that predict student success.
  • Even when students are on track at the end of ninth grade, differences in ultimate graduation rates persist among gender and racial subgroups. The greater use of on-track models can identify when schools narrow the graduation rate gaps. The relationship between data and effective practice could lead to scaling of instructional models to other schools.
  • The on-track indicators are a descriptive signal; they are not valid substitutes for effective student identification and intervention. While the data signals act as early warning systems for at-risk students, they should provoke a larger discussion on how to transform instruction to disrupt the dropout cycle.

Research Design:
Study uses the New York City Department of Education's individual student administrative records (n=576,424) to measure the predictive value of 14 potential on-track indicators on high school graduation.

First-time ninth graders who were enrolled in New York City schools between 2001-2002 and 2010-2011. The total longitudinal sample size is 576,424. The sample draws from more than 350 high schools.

Year data is from:


Data Collection and Analysis:
The data include detailed demographic, enrollment status, attendance information, as well as the number of courses taken, number of credits accumulated, and scores on the New York State Regents exam.


Reference in this Web site to any specific commercial products, processes or services, or the use of any trade, firm or corporation name is for the information and convenience of the public, and does not constitute endorsement or recommendation by the Education Commission of the States. Please contact Kathy Christie at 303-299-3613 or kchristie@ecs.org for further information regarding the information posting standards and user policies of the Education Commission of the States.