As we prepare for the reauthorization of the federal No Child Left Behind Act, a lot of legislators and advocates think they know the bases of student learning and the relative contribution of each. But what is almost startling is how little macro-level work has been done comparing the importance of each factor. As important as anything in the reauthorization is the need to invest our federal research dollars in looking at the field’s biggest questions. The Institute of Education Sciences and the Fund for the Improvement of Education must play a larger and a different role in this and subsequent versions of the Elementary and Secondary Education Act if we are going to align our educational allocations with some sense of what’s effective.
Deeply felt cases have been made for the importance of all these methods of improving student performance:
• Competence of each teacher in the subject matter he or she is teaching;
• Small class size or small schools, or both;
• A scientific, research-based curriculum;
• Assessments that gauge performance and keep teachers on task;
• Merit-based pay, which keeps teachers on task;
• Integration of educational technology into both the curriculum and data management;
• Better physical condition for schools;
• Strong early reading programs;
• Meaningful graduation requirements;
• Acknowledgment of different learning styles and adjustment of the pedagogical technique for the particular student;
• And, of course, a look beyond the classroom to an increase in parental involvement.
So here’s the set of conclusions Congress came to, as deduced from fiscal year 2007 federal funding: State assessments are worth about 1½ times as much as educational technology ($407 million vs. $272 million), but only 40 percent as much as helping slower readers catch up (Reading First at $1.029 billion). Teacher-quality grants are worth about 10 times as much as educational technology, and almost three times as much as early reading support. Support for early learners having trouble with math is worth nothing, and exchanges with historic whaling and trading partners (which I inadvertently left out of my list) is worth as much as 2 percent of what state assessments deserve.
Who came up with this recipe? There is nothing systematic about it. While there are major proponents for the relationship of each of these expenditures to improved performance, support for expenditures is almost always insular: How does providing this service compare to not providing it, or providing less of it? Rarely does Congress receive comparative analyses—how does providing this support differ from or fit in with providing other forms of support?—or multivariate analyses—if I’m providing so much of A already, then which helps more, B or C?
Federal funding tends to support looking inside each of the ‘boxes’ of learning.
There are two primary federal vehicles for research on educational techniques. The biggest program is in the Institute of Education Sciences. It had a $517 million budget in fiscal 2007.
But paring back shows that the bulk of this funding is for performing specialized functions or doing specialized research—for the National Assessment of Educational Progress, for special education, for statewide data systems, and for regional education labs. The amount spent on general research last year was about $162 million, and the allocation for this year likely will stay around the same.
But even this money is largely spent on intramural research. The current management of the IES has worked to make its evaluations more “science-like” and rigorous. But the funding provided tends primarily to support looking inside each of the “boxes” of learning, with grant competitions this past year for separate research on reading and writing; math and science education; teacher quality in each of these areas; cognition and student learning; and high school reform.
The other potential research vehicle, the Fund for the Improvement of Education, provides little for real research. The vast majority of its funding is for earmarks, which mostly provide more money for a particular locality or program to do more of what nonresearch funding is for: after-school programs, teacher training, technology upgrades, and family literacy.
Of course, there will be problems in finally tackling the Big Questions. There may be a need to insulate the work still further from political emphases and conclusions. Because it is founded in the real world where circumstances vary, such research won’t allow for precise experimentation, and the results may not be applicable in every condition. But some comparison of all these factors will allow resources to move in the right direction.
If, after a few years, it turns out that just working within the boxes produced a better result, we can go back. But who will take that side of the bet?