Sarah Cordes

CV

Sarah Cordes is a fifth year doctoral student at New York University's Robert F. Wagner Graduate School of Public Service. Her research and teaching interests are in education and urban policy, housing, public finance, and applied statistics and econometrics. Sarah's dissertation explores the spillover effects of NYC charter schools on nearby public school students, the effects of residential and school mobility on student performance, and how changes in school resources influence parents' investments in their children's education. In other ongoing research, Sarah is examining the effects of school mobility on student performance and the effects of housing voucher receipt on student performance. Sarah is a recent recipient of the C. Lowell Harris Dissertation Fellowship awarded by the Lincoln Institute of Land and was selected as a finalist for the National Academy of Education/Spencer Dissertation Fellowship. 

 

Sarah received her MPP from the Sanford School of Public Policy at Duke University in 2010, with a concentration in social policy. Prior to attending Duke, Sarah spent two years teaching middle school math in Washington, DC as part of AmeriCorps.

References  
  • Research Interests
Education Policy, Economics of Education, Urban Policy, Public Finance,

 

  • Teaching
PADM-GP.2902: Regression and Introduction to Econometrics                                                                                                                Multiple regression is the core statistical technique used by policy and finance analysts in their work. In this course, students learn to use and interpret this important statistical technique. Specifically, students learn how to interpret regression coefficients, evaluate whether coefficients are biased,  whether standard errors (and thus t statistics) are valid, and whether regression results presented in policy and finance studies can be interpreted as causal. In addition, using a number of different datasets, students learn to apply the concepts learned in class using Stata. Finally, students work in groups to estimate their own regression models applying the techniques learned in class to project datasets supplied by the instructor.