For forty years, the federal government has sought to close the "achievement gap" by increasing federal involvement in and spending on public education. But the achievement gap persists; students from higher-socioeconomic backgrounds and white students continue to outperform children from lower socio-economic or minority backgrounds.

Expanding government efforts have failed to solve the achievement gap, but religious schools have succeeded where the public schools have failed. If eliminating the achievement gap is a main goal of education policy, new evidence on religious schools suggests it is time to rethink the current strategy.

According to the National Assessment of Educational Progress (NAEP), the achievement gap is virtually unchanged since the mid-1980s and remains alarmingly wide. In 2005, only 15 percent of students eligible for a free or reduced price lunch scored at or above proficient on the 4th grade reading exam, compared to 42 percent of students not eligible for federal assistance. Broken down by race and ethnicity, 47 percent of white students are at or above proficient, compared to only 13 percent of black students and 19 percent of Hispanic students. The numbers for the mathematics exam are even more dismal.

The typical response to results like these has been to spend large amounts of money on schools with a high concentration of impoverished students. This was the purpose of the original Elementary and Secondary Education Act of 1965 signed by Lyndon B. Johnson. Between 1970 and 2005, federal spending on education increased by 146 percent.

Over that time, NAEP scores have been flat.

In 2001, President Bush signed the No Child Left Behind Act, the latest version of the Elementary and Secondary Education Act, which focused on the achievement gap with renewed vigor. NCLB increased elementary and secondary education funding by 34 percent, or $6 billion.

Understanding that merely throwing more money at education has never worked, President Bush pursued other policies, such as mandatory testing, separating testing data by student sub-groups, and sanctions for schools that fail to make adequate yearly progress. But again, more federal mandates haven't led to meaningful improvement in NAEP scores.

But all hope isn't lost. A recent study by researcher William H. Jeynes of Baylor University found that private religious schools have been very successful in reducing the achievement gap.

Evidence suggests that private religious schools have a positive academic effect on all students. Moreover, these schools provide greater benefits to students from the poorest families and minority students.

The result is that private religious schools have a much narrower achievement gap than their public school counterparts.

According to Jeynes, the research shows "that religious schools reduce socioeconomic and racial achievement gaps." It is still unclear, however, "what features of religious schools help to explain the alleviating of these achievement gaps." Social scientists have put forward three possibilities: the nature of religious devotion, a difference in school culture, and increased parental involvement.

Religious schools outpace public schools in all of these areas, and Jeynes' research found that all three have a significant impact on academic achievement.

Religious schools' success could have a greater impact narrowing the achievement gap through parental choice in education, particularly for students from low-income families. Not only would these students benefit dramatically by attending the private school of their choice, but competition would cause public schools to mimic private schools' best practices.

Private school choice programs have historically had this effect.

According to the U.S. Department of Education's National Center for Education Statistics, parents whose children took advantage of school choice were more satisfied with their children's schools and more involved in their children's education. Harvard University economist Caroline Hoxby has found that public schools improve in response to school choice competition.

If policymakers are serious about narrowing the achievement gap, they should pay attention to the Baylor University study. While top-down government approaches have failed for forty years, private schools have succeeded in reducing the achievement gap. Allowing disadvantaged students to choose schools like these is a promising solution to this persistent problem.

Evan Feinberg is a research assistant in domestic policy studies at the Heritage Foundation, www.Heritage.org.