New Mathematica Working Papers Focus on Using Value-Added Models
Contact: Jennifer de Vallance, (202) 484-4692
WASHINGTON, DC—June 20, 2012—As part of its ongoing working paper series, Mathematica recently released two new papers featuring compelling research on using value-added models of performance measurement for both education and health care. These papers contribute to the existing discussion about using value-added to measure teachers' impacts on student achievement and encourage new discussions on using these models to help improve our health care system.
"Rewarding Physicians for Their Patients' Health Outcomes: What Can Medicare Learn from Education's Value-Added Models?," by Greg Peterson and Eric Schone, examines how value-added models of performance measurement used in education can apply to health care when measuring and rewarding physician performance, particularly in the Medicare system. The paper finds that value-added models used to measure teacher effects on student test scores may have useful applications in Medicare because the models provide opportunities to identify impacts on outcomes while controlling for factors that are beyond physicians' control, such as severity of illness or adherence to a treatment plan. The working paper also identifies limitations and challenges associated with value-added measures and discusses potential approaches to address these challenges.
"Methods for Accounting for Co-Teaching in Value-Added Models," by Heinrich Hock and Eric Isenberg, helps to address the issue of isolating the effect of each teacher on student achievement when the student is taught the same subject by more than one teacher. This paper considers and compares three methods—Partial Credit Method, Teacher Team Method, and Full Roster Method—to estimate teacher effects. Based on the analysis, the authors conclude that the latter two methods provide a more stable approach to estimating teacher effects on student achievement. The Full Roster Method also offers the most promise for robust, practical implementation.
Mathematica's working papers offer timely, accurate, evidence-based research before final peer review and publication. In keeping with Mathematica's high standards of quality, they have undergone a rigorous internal quality review.
About Mathematica: Mathematica Policy Research seeks to improve public well-being by conducting studies and assisting clients with program evaluation and policy research, survey design and data collection, research assessment and interpretation and program performance/data management. Its clients include foundations, federal and state governments and private-sector and international organizations. The employee-owned company, with offices in Princeton, NJ; Ann Arbor, MI; Cambridge, MA; Chicago, IL; Oakland, CA; and Washington, DC; has conducted some of the most important studies of health care, nutrition, education, international, disability, family support, employment, and early childhood policies and programs.