The National Assessment of Educational Progress (NAEP) “was never intended to measure the quality of education and was never linked to any curriculum,” writes research psychologist, Gerald Bracey, in Phi Delta Kappan. Since states do not adhere to a standard curriculum that is aligned with the NAEP tests, results can not show which states are doing the best job in educating their children.
In fact, the differences in state scores are due, in large part, to differences in state spending on education and student populations. Recently, United States Department of Education officials have been discussing the idea of “adjusting” scores to determine how a state would perform if it had a “representative” sample of students.
Adjusted scores would then be used to compare the quality of the education provided by each state. However, because most states do not have “representative” student populations, Bracey believes that adjusting test scores for demographic factors simply covers up the real problems states face.
Besides, he says, 89 percent of the variance in NAEP scores can be attributed to four factors that are beyond the control of schools: the number of parents in the home, the level of parental education, the type of community, and state poverty rates.
Bracey recommends that instead of adjusting scores, existing state-level NAEP results could be used as an indicator of the difficulty of each state’s educational task and could serve as a means to determine more equitable ways to fund education.
Bracey concludes that because districts vary so greatly demographically, parity in education will never be obtained using property taxes as the principal funding source. He cites the National Science Foundation which reports that in some countries poor school districts receive two or three times the funding of wealthy ones. In the United States, however, it is the wealthiest districts that have two or three times the funding.
“NAEP and the Quality of Education”, Phi Delta Kappan, Volume 76, Number 1, September 1994, p.84.
Published in ERN, November/December 1994, Volume 7, Number 5.