Heinz School Helps Local Principals Make the Most of Standardized Tests

The standardized tests required of public schools under the federal No Child Left Behind Act promise to produce reams of student data. To help schools use this information to improve instruction, Carnegie Mellon's H. John Heinz III School of Public Policy and Management will offer an intensive, month-long course in educational statistics this summer to 16 principals in the Pittsburgh Public Schools.

The course, Statistics and Educational Assessment, mirrors a class taught last fall as part of the Heinz School's Master of Science in Education Technology Management program, which trains education professionals to make better use of instructional technology. The ultimate goal of the summer program is to improve the quality of instruction in city schools, said Robert P. Strauss, a professor of economics and public policy in the Heinz School. The course began July 6.

"It was foreseeable that the federal legislation would require school districts around the country to begin collecting lots of data. The problem is that many school managers have had very little statistics and also probably have not taken advantage of the personal computer revolution," Strauss said.

The course will be team-taught by Strauss and Norma Chang, a doctoral student in Carnegie Mellon's Department of Psychology, who brings to the course her doctoral research into how people learn to use statistics and data analysis, and her earlier experience as a high school science and mathematics teacher. During the course, each principal will work with data from his or her own school and use the information to pinpoint students' specific weaknesses.

"Our goal is for these principals to learn how to analyze student data to track long-term progress and evaluate the success of instructional interventions. This is an especially important skill to develop because principals need to be able to draw sound inferences about large groups of students, whereas people naturally find it easier to reason about individual cases," Chang said.

The principals will learn how to use up-to-date statistical software packages to analyze and interpret test score data, and how to design, implement and evaluate educational interventions, as well as tailor those interventions to match the needs of groups of students. At the end of the course, the principals will devise and administer a pretest on a portion of the curriculum that is being taught during the summer school session in their schools. Then, they will identify from research literature different ways to improve student achievement in that area. Finally, the principals will administer a post-test and, with the statistical skills developed in the course, ascertain whether or not the intervention actually raised achievement more than traditional teaching methods.

"We hope to build the capacity of our principals to become better instructional leaders by using data on student achievement to focus instruction on what students need to learn," said Philip Parr, chief of staff for the Pittsburgh Public Schools.

This program is one of several that Carnegie Mellon has offered to schools in Pittsburgh and elsewhere to help them improve their students' performance on standardized tests.

Jonathan Potts