Rich schools get richer

The neediest students are being shortchanged, concludes Funding Gaps, 2006, an Education Trust report. Federal Title 1 funds — $13 billion a year to fund extra help for low-income students — go disproportionately to wealthy states.

For example, Maryland has fewer poor children than Arkansas but receives 51 percent more Title I aid per poor child, even though Arkansas dedicates more of its taxable resources to education than wealthier Maryland.

In about half the states, high-poverty and high-minority districts receive less funding than low-poverty and low-minority districts.

On average, states and localities spend $908 less per student in districts educating the most students of color, and $825 less per student in districts educating the most low-income students as compared to what is spent in the wealthiest and whitest districts.

Within school districts, “substantially less money is spent in high-poverty and high-minority schools.” Experienced, top-scale teachers cluster in schools with more affluent students. Austin spends an extra $383,700 per year at an advantaged school with 100 teachers.

In addition, districts balance the federal funds for poor students by spending most of their extra state and local funding at lower-poverty schools.

It’s no surprise Title I hasn’t closed the achievement gap, the Washington Post points out.

Experts say children raised in poverty need more instructional time and specially trained teachers to help overcome their disadvantages — resources that require more spending.

More money won’t solve all the high-poverty schools’ problems, but they deserve at least as much as the kids in middle-class schools. In particular, it would make a real difference if needy schools could offer more money to lure experienced and master teachers.

About Joanne


  1. This analysis is a good starting point for discussion, but it has some serious holes from a design standpoint. They treat a dollar in Arkansas equal to a dollar in Maryland or Massachusetts, when the differences in purchasing power among the three are different because the cost of goods and services are dramatically different. Using the CNN/Money cost of living calculator, doing a quick comparison between Baltimore and Little Rock, you find that food is 12.5% higher, housing is 43.9% higher, transportation is 5.2% higher, and healthcare costs are 11.3% higher in Baltimore. This means that Baltimore schools will have a much higher cost of employment per teacher than Little Rock schools will. As the report is currently presented, the intra-district comparisons are more meaningful than the interstate comparisons because of the failure to account for differences in the costs of goods and services in different geographic locations.

    These average revenue per student calculations can be misleading. The authors acknowledge that they did not attempt evaluate how money is spent, so systems that are growing and have high capital expenditures, along with systems that have high overhead costs, will reflect a higher per student revenue, but that additional revenue does not necessarily translate into better services for the students.

  2. It’s not just cost of living that’s not figured into the report. Expectations/attainments aren’t considered either.

    Without knowing what the expectations are of the schools or districts, or how well they perform, it’s impossible to determine value, either absolutely or comparatively.

    It’s not just the buying power of the dollar that has to be considered, the value received for the dollar is also important. The more so since there’s no correlation between spending and education, a relationship which this report, perhaps unintentionally, assumes.

  3. Person who read the report says:


    The numbers in the report are adjusted for differences in the costs of goods and services in different geographic locations.

    Goodwin Liu’s section of the paper uses the Cost of Education Index developed for the U.S. Department of Education by Jay Chambers. Ross Wiener and Eli Pristoop’s section of the paper uses the Comparable Wage Index, which was developed by and is available through the U.S. Department of Education.

    This is mentioned in the body of the report on pages 2 and 6, and in the note to table 2 on page 4 and in the note to tables 3 and 4 on page 7. Additionally, there is a lengthy discussion of both the rationale for and method used to make these adjustments on pages 2, 3, and 4 of the technical appendix.

    It is kind of surprising to me that you would go to the trouble of doing an analysis to prove to readers how problematic it is that the authors neglected to make geographic cost adjustments before going to the trouble of reading the paper closely enough to see whether or not they actually made them.

  4. To Person who read the report –

    After reading your comment, I went back and did a better read of both. I acknowledge my error in that it is clear that they did adjust for geographic differences in the analysis. But as I read their work, I took note that they included a number of tables intended to reinforce their work that did not. Some of these were published well before the recent NCES CWI report. I still stand behind the second point, and there are a number of other problems with the analysis as well, not the least of which is that the author assumes inequities without accounting for how any money, including federal, is actually spent. He also uses total number of poor children per state, knowing full well that each state will have a number of poor children who do not attend Title 1 Schools, which will decrease statewide per student Title 1 average funds without accurately representing the true state of affairs.