This past Sunday, the L.A. Times kicked off a series of articles investigating the effectiveness of public elementary schools and individual teachers in the Los Angeles Unified School District by presenting a statistical analysis of recent standardized test scores of more than 600,000 students.
The Times, using money from a Hechinger Report grant, hired Richard Buddin, a senior economist and education researcher at Rand Corp., to analyze the data through a value-added model.
Buddin discovered that conventional wisdom about teacher effectiveness seemed to clash with his findings—the best teachers weren’t concentrated in affluent neighborhoods, and the worst teachers didn’t come from poorer areas. He also noted that “many ‘important’ teacher qualifications have little to no impact on student outcome,” and that most student background factors (parents’ educational level, socio-economic status) affect value-added scores less than expected. (He uses his findings to justify a merit pay system, for what it’s worth.)
To the Times’ credit, they did acknowledge the limitations of their study:
Value-added ratings reflect a teacher's effectiveness at raising standardized test scores. As such, they capture only one aspect of a teacher's work, and, like any statistical analysis, they are subject to inherent error.
Buddin’s full report can be found here.