Dayton Daily News education reporter Scott Elliott writes about schools, kids, teaching and learning.
Home > Blogs > Get on the Bus > Archives > 2008 > October > 01 > Entry
Understanding Botkins vs. Jackson Center
I’ve gotten a lot of interesting feedback about my column last week, which argued the state should consider how school districts perform relative to the wealth in the community when judging their test performance.
Online I listed the Miami Valley’s top overachieving and underachieving school districts.
There is one pair of districts that has puzzled me ever since I first began making this comparison two years ago — Botkins and Jackson Center.
These districts border each other in Shelby County and the median income for the two is nearly identical. Yet, Jackson Center is the area’s second worst underachiever while Botkins is the second biggest overachiever.
At one point, I went up and visited both districts to try to discover why. Had there been a dramatic change in Jackson Center, like a plant closing? Not that I could find. Were Jackson Center schools just less efficient at instruction? Perhaps, but I didn’t find evidence of a dramatic difference between the two districts when it came to instruction.
Could it be somehow cultural? Could the two communities place a different value on education?
That didn’t seem likely for two districts so close together. But then I noticed something interesting while looking at this fabulous map of school district test performance the DDN’s graphics department created using report card data.
As I clicked through the interactive map, I noticed something odd. There appeared to be a dividing line that roughly tracked the eastern border of Miami and Shelby counties. On the west side of the line, district had high performance on the state report card. On the east side, performance was not as good.
Then I pulled the income data for the districts along the line and sure enough, there was a divide. Here are the districts on the west side of the line and how they scored on my measure of test performance compared to median income:
Miami East +13
Sidney +231
Anna +73
Botkins +191
Wapakoneta +68
And here are the districts on the east side of the line:
Waynesfield-Goshen -260
Jackson Center -218
Indian Lake -119
Fairlawn -161
Graham -236
Northwestern -119
Tecumseh -52
Bethel -67
To the west, districts tended to score better than their incomes would predict. To the east, they score worse than predicted.
This, again, led me back to the question of culture and the value of education in those communities. Could it be that the districts to the west are somehow different culturally than those to the east?
If you know anything about these communities and can offer insight, please share your thoughts in the comments.
Permalink | Comments (3) | Post your comment | Categories: Testing
Comments
By Mary
October 6, 2008 2:21 PM | Link to this
I agree with Rick. Also, wealthier people tend to benefit from the status quo, are more motivated to maintain, defend and excel in the existing systems - whether education or work. They have the leg up and the power in the existing system. They are not necessarily smarter.By Rick
October 4, 2008 4:34 PM | Link to this
Scott, there is correlation between education and wealth because wealthy people tend to be achievers, goal oriented people and they pass that on to their kids. There is also correlation between the education levels of parents and the achievement levels of their children. In addition, there have been studies over the years comparing schools with similar demographics but different achievement levels. So while wealth may have a correlation, it is not determinative.By School Supporter
October 3, 2008 10:45 AM | Link to this
About ten years ago the educational benefits of west central Ohio culture appeard in a paper from Richard Vedder with a couple of his Ohio U students. As to variations within that region, you might be on to something, but, frankly, a 100 point spread probably isn’t significant, and that might be true of a 500 point spread as well. Among the problems: using rank rather than normal curve equivalents and using demographic data for entire communities rather than families with children (please correct me if I’m wrong on this!). If poverty in a town comes from high concentrations of the elderly (e.g. nursing homes), that’s different from a district that has a migrant worker trailer park with schoolchildren who are English language learners. It might be valuable to interview county superintendents, but skip the munged data and provide them scatter plots of the raw data.