Willingham: What’s missing from Common Core standards plan (Part 2)
My guest today is cognitive scientist Daniel Willingham, a psychology professor at the University of Virginia and author of “Why Don’t Students Like School?” He wrote the first part of this two-parter on July 26 .
By Daniel Willingham
Last week I suggested that the adoption of the Common Core standards was unlikely to improve student outcomes. Many interacting factors determine student performance, and even if state standards improve, it’s all too likely that problems elsewhere in the system will mask the improvement.
When that happens, there is a tendency to get discouraged, and tinker with some other part of the system.
It brings to mind my misguided attempts at debugging my computer program when in college. I remember whining to a more experienced student: “How can it still not run? I changed everything!”
If the educational system has so many interacting parts, how can we ever know whether a change we make to part of the system is useful? I see two ways.
First, we might encourage the development of more detailed models of the system as a whole. Suppose that I, as a researcher, really put my money where my mouth is and propose a model of all the factors that contribute to educational outcomes, and how they interact.
My model would be detailed enough that I could make not just qualitative predictions but quantitative predictions.
Qualitative predictions include statements like “this type of standard is better than that type of standard, and if a state adopts the better one, scores will go up some unspecified amount.”
Quantitative predictions include statements lsuch as, “This standard rates 7.3 in my model, whereas that standard rates 8.2, so I predict that if New York adopts the latter, students scores on this particular test will improve by 4.4%.”
Note that these models have metrics by which standards (and other features of the system) are evaluated, and these metrics are not based on student test scores. So the model goes out on a limb as to what are “good standards.”
Those metrics usually come from nothing more reliable than the modeler’s forested mind, but the model is kept honest because it makes very precise predictions about student outcomes, and thus is easy to prove wrong.
The likelihood that I or any other researcher would happen on the right model is remote.
Rather, we expect that the enterprise as a whole will move forward as a broad variety of models is proposed. Some will be better than others, and patterns will emerge, for example, that models with a particular feature never fare well.
The second way we might address this problem is to make greater use of epidemiological methods.
Epidemiology is the study of disease in populations, especially in identifying risk factors and protective factors.
Education researchers have made good use of these methods in identifying characteristics of individuals, families, and neighborhoods as risk factors for dropping out of school, for example, or for going to college.
These methods have less often been applied to classroom practices, administrative practices, and other features of the educational landscape that might be under our control.
This is true, no doubt, because they are much harder to measure than personal characteristics such as whether or not a child’s parents are divorced. To measure them, we would have to accept untested theoretical assumptions, gambling that, in time, which assumptions are sound and which are hollow would become clear.
Both roads I have suggested here have enough foreseeable problems to be discouraging, and in my opinion if you undertake a complex project assuming twice as many unforeseen problems as foreseen ones, you’re an optimist.
But given the lack of progress in our understanding of the educational system as a whole, and given its importance, I argue that we have to at least try.
Next week I’ll describe what I take to be the most important political requirement in this enterprise.
Follow my blog all day, every day by bookmarking washingtonpost.com/answersheet. And for admissions advice, college news and links to campus papers, please check out our Higher Education page at washingtonpost.com/higher-ed Bookmark it!
| August 2, 2010; 11:00 AM ET
Categories: Daniel Willingham, Guest Bloggers, National Standards | Tags: Daniel Willingham, common core standards, national standards
Save & Share: Previous: Ed policies ignore science on how/when kids learn
Next: Proficiency should mean college ready—and an acceptance letter
Posted by: bsallamack | August 2, 2010 1:02 PM | Report abuse
Posted by: mattmel | August 2, 2010 3:07 PM | Report abuse
Posted by: bsallamack | August 2, 2010 3:54 PM | Report abuse
Posted by: bsallamack | August 2, 2010 4:11 PM | Report abuse
Posted by: natturner | August 3, 2010 3:03 AM | Report abuse
Posted by: dccitizen1 | August 3, 2010 4:28 AM | Report abuse
Posted by: sideswiththekids | August 3, 2010 2:46 PM | Report abuse
Posted by: DanielTWillingham | August 4, 2010 10:52 AM | Report abuse
The comments to this entry are closed.