Publication Date: 2002-04-03
This paper, delivered at the annual AERA Conference in 2002, outlines issues related to the reliability of accountability systems, describes methods used to calculate reliability, proposes a new method, and shows that the method produces accurate findings by comparing results from generated data to actual statewide results. It further demonstrates that school scores within year vary as if they were a random draw from the population in the school's catchment area.
AERA 2002_RH03.pdf (Size: 160 KB)
Center for Assessment