|
Session Title: Using District and State Administrative Data to Evaluate Education Policies and Practices
|
|
Panel Session 724 to be held in Suwannee 14 on Saturday, Nov 14, 9:15 AM to 10:45 AM
|
|
Sponsored by the Pre-K - 12 Educational Evaluation TIG
|
| Chair(s): |
| Kellie Kim, Academy for Educational Development, kkim@aed.org
|
| Discussant(s):
|
| Jeff Sellers, Florida State Department of Education,
|
| Abstract:
Panelists will discuss their experiences using administrative data to evaluate district and state policies and practices in elementary, secondary, and postsecondary education. Each presenter has used district- or state-maintained student records to investigate critical education issues, such as developing behavioral indicators of at risk status (i.e., "early warning signals"), secondary education reforms, and predictors of STEM degree attainment at the postsecondary level. The presentations will describe the benefits and challenges of using student records to evaluation policy and practice.
|
|
Creating a High School Ready Indicator Using District of Columbia Public School Data
|
| Becky Smerdon, Academy for Educational Development, bsmerdon@aed.org
|
| Aimee Evan, Academy for Educational Development, aevan@aed.org
|
|
The panelist will discuss a 12-month collaborative project with the District of Columbia to use D.C. Public Schools' administrative data to develop a clear understanding of the roots of the dropout problem in the District. Because the drop out problem is most pronounced at the ninth grade level, the project focused on the years just prior to ninth grade, middle school, and the transition from middle to high school. Using extant, student-level longitudinal data, this collaborative effort aimed to provide a definition of "high school readiness" that can be used to identify at-risk middle school students and DCPS middle schools for targeted interventions. Designed to serve as a model for how districts (and research organizations) can build research partnerships and employ data-based decision-making and strategic planning across all levels of the system, results from this project provide important lessons about the strengths and limitations of using administrative data.
|
|
|
Evaluating High School Reform Using Baltimore City Public Schools Data
|
| Jennifer Cohen, New America Foundation, cohenj@newamerica.net
|
|
The researchers used student-level administrative data from the Baltimore City Public School System (BCPSS) to assess the impact of a five year high school reform initiative on student sorting, mobility between schools, achievement and school climate. The study's outcomes suggest that the district's new "innovation" schools outperformed other non-selective schools while attracting a slightly less disadvantaged student population. Data were primarily obtained from BCPSS and manipulated for research purposes. Challenges include navigating school enrollment as schools opened, closed, or broke apart, and identifying school membership and attendance in a highly mobile population. Data used include demographics (race, gender, special education status, English language learner status, age), previous achievement (8th grade scores in mathematics and reading), attendance rate, and achievement on English and Algebra standardized tests. Methods employed include descriptive statistics, OLS regression, and hierarchical linear modeling. The advantages and limitations of using BCPSS data will be highlighted.
| |
|
Career Academies in Florida: How Many, What Type, and Who Enrolls?
|
| Kathryn Borman, University of South Florida, kborman@cas.usf.edu
|
| Bridget Cotner, University of South Florida, bcotner@cas.usf.ed
|
| Corinne Alfeld, Academy for Educational Development, calfeld@aed.org
|
| Robert Bozick, Academy for Educational Development, rbozick@aed.org
|
| Reggie Lee, University of South Florida, rlee@cas.usf.org
|
|
This National Science Foundation funded study begins by mapping the current landscape of career academies in Florida. Specifically, this study describes what type and how many career academies are available to high school students in Florida and the characteristics of students who enroll in career academies. This information will help decision makers in Florida and elsewhere to determine the extent to which career academies focus on the high-need industries (e.g., STEM) and enroll students who are historically underserved. This study will also include a coursetaking analysis to provide insights into the extent to which core subject areas and career-oriented content are integrated and the balance of coursetaking in core subject areas and career-oriented topics in career academies and other school settings. Use of Florida's student-level, longitudinal data to address the evaluation questions will be discussed.
| |
|
Studying the Effects of Early College High Schools Using North Carolina Data
|
| Nina Arshavsky, University of North Carolina at Greensboro, narshavs@serve.org
|
|
The presenter will discuss using NC Extant Data to Determine the Efficacy of the Learn and Earn Early College High School Model. Schools participating in this experimental study use a lottery to select their students. The study then links these students to extant data already collected by the North Carolina Department of Public Instruction. Results are compared students who were randomly selected to attend and those who were randomly selected not to attend on the following outcomes: student attendance, achievement, coursetaking patterns, school leaving and dropout status, and disciplinary outcomes. Original survey data on attitudes and school experiences are also collected and linked to the longitudinal North Carolina data, which are housed at the North Carolina Education Research Data Center.
| |