Pages

Friday, July 15, 2011

A modest proposal for improving NCLB

Simply put, we should try to better compare apples to apples.

Currently, No Child Left Behind requires schools to present their test results "disaggregated" by several subgroups including race/ethnicity (White, Black, Hispanic, Asian, American Indian), Limited English Proficient, Special Education, Migrant, and "Free and Reduced Lunch" recipients. Many children are in multiple categories and count for or against their school's "report card" multiple times, and one of those variables (free and reduced lunch, which indicates household poverty) looms huge as a factor for student success. So the data provided by the current disaggregation is of limited value.

My proposal: keep the categories except for free and reduced, and divide each category out into household income quintiles. Then, you'd be able to compare one district's White students in the lowest quintile with another district's White students in the lowest quintile, instead of only being able to compare their White kids (most of whom are poor) with another district's White kids (most of whom are in the fourth or fifth quintile). It would be a more fair comparison in other categories, too: is the English language learner child of visiting professors succeeding at the same rate as the English language learner child of manual laborers?

This change would also address another problem: although currently "free and reduced lunch" kids are lumped into one group, evidence suggests that there is a significant difference in performance between kids who receive free lunch (because their families are very close to or under the poverty line) and those who receive reduced lunch (which can be given to kids whose families make 185% of the poverty limit, and thus would be in the second income quintile, not the lowest one).

This kind of change would require collecting some more data from students' families, by either pulling the info from students' free and reduced lunch applications or having parents check off a box indicating which range of household income they have and how many people are in the household. But the payoff would be huge: a better comparison of which school divisions are actually succeeding at meeting the needs of at-risk students. Can you imagine the implications?

Image courtesy of dano via flickr.

No comments:

Post a Comment