The introduction of accountability systems leads to higher achievement growth than would have occurred without accountability. But simply reporting results of tests has a minimal impact on performance. The systems are much more effective if poor educational results have adverse consequences for the schools.
When teachers and their schools are held accountable for the educational performance of their pupils and face consequences when the children do not measure up to goals, student grades in reading and mathematics do improve. However, the insistence by many American states in the 1990s on educational standards and testing for primary school students has not narrowed the educational gap between blacks and whites, although it did trim the Hispanic-white achievement gap.
These are the key findings of Does School Accountability Lead to Improved Student Performance? (NBER Working Paper No. 10591) by Eric Hanushek and Margaret Raymond. Their analysis of state achievement growth, as measured by the National Assessment of Educational Progress (some times referred to as the "Nation's report card"), is highly relevant to the drive by the federal government to improve educational performance across the nation. A central campaign theme of George W. Bush in his first bid for the White House was to expand educational accountability to all states. This goal was put into law with the No Child Left Behind Act of 2001 (NCLB).
The majority of states had instituted some sort of accountability system by the time NCLB was passed. Only 12 states had such systems in 1996. By 2000, 39 had these programs. The new federal law expanded accountability by requiring all states to have annual testing of students in grades 3 to 8, by mandating disaggregated reporting of data on student performance for all schools, and by adding new sanctions when student performance falls short. States have used their own systems as the basis for implementing the federal law. Thus, analysis of past state results provides insights into the potential impact of NCLB.
In looking state-by-state, Hanushek and Raymond find that the introduction of accountability systems leads to higher achievement growth than would have occurred without accountability. But simply reporting results of tests has a minimal impact on performance. The systems are much more effective if poor educational results have adverse consequences for the schools, thus supporting the contested provisions of NCLB that impose sanctions on failing schools. Hispanics students gain most from accountability while blacks gain least.
The analysis relies on the National Assessment of Educational Progress testing of fourth and eighth graders in reading and math. The data provide performance information for whites, blacks, and Hispanics. In their analysis, Hanushek and Raymond separate the effects of accountability from the impacts of the racial composition, of family characteristics of students, and of other state policies on achievement. For instance, throughout the 1990s, attendance of white students in large urban school systems has decreased and minority concentration has grown, and the authors find that black educational performance appears to be hurt when they attend less integrated schools.
The disaggregated results of accountability present a policy challenge. Accountability increased the black-white gap a little, because the performance of blacks improved less than that of whites. "Achieving multiple objectives with a single policy instrument is not generally feasible," the authors conclude.
Accountability policy has been controversial. Some assert that the new policy has distorted school decisions in undesirable ways, such as leading to higher drop-out rates, more cheating on tests, and undesirable narrowing of what is taught, although evidence on these effects is currently limited. Another charge is that it has prompted schools to weed out poor achievers by placing more students in special education classes - those for the educationally handicapped - and thereby improve the regular achievement score for the school and its classes, regardless of efforts to upgrade actual teaching. The Hanushek-Raymond study finds no such effects at the state level. Between 1980 and 2001 the proportion of students assigned to special education classes rose from 10 percent to over 13 percent. But this trend, one going on for two decades now, was not altered by the introduction of accountability across states in the 1995-2000 period.
-- David R. Francis