Are Hispanic students the driving force behind the rise in urban test scores?

When the National Center for Education Statistics reported on Dec. 18th that fourth and eighth graders in the country’s largest cities had shown marked improvement in test scores over the past decade, a question kept popping into my head. To what extent is white gentrification of cities driving this test score increase? Is it possible that the achievement gap between low-income inner-city minorities and the rest of America isn’t really closing by 30-40 percent, but a different population of youngsters with Ivy League-educated and artistic parents is now taking these tests?

Answering this question is proving tough. But here’s what I do see in the data tables.

1) High income kids of all races posted larger gains than low income kids in large cities between 2003 and 2013.

Data source: NCES TUDA data 2013 for fourth grade math scores in large city public schools. Poor are students who are eligible for free or reduced priced lunch. “Not poor” are students who are not eligible for the lunch program. Data found using custom data tables here.

2) But to my surprise, the student population of large cities hasn’t shifted to become richer and whiter. Indeed, there are proportionally fewer whites in cities today and more poor kids than there were 10 years ago. The biggest changes are the decline in black populations and the rise of Hispanic populations.

Data source: NCES TUDA data 2013 for fourth grade math scores in large city public schools. Poor are students who are eligible for free or reduced priced lunch. Not poor are students who are not eligible. Data retrieved using custom data tables here.

(When you look city by city, however, you sometimes see a different picture. In New York City, for example, the percentage of white fourth graders has increased by 2 percentage points over the last decade to 17 percent. And the percentage of poor kids has declined 9 percentage points from 88 percent to 79 percent. So New York is richer and whiter. But the city had a modest 10 point increase in fourth grade math test scores.

(Washington DC, which had an impressive 24 point gain in fourth grade math test scores, had a much larger shift in its white population —  up 9 percentage points to 13 percent. But student poverty also increased in the District by 5 percentage points to 76 percent.)

3) Next I tried to break down the demographic shifts by income. I was able to find NAEP TUDA data that broke down lunch program eligibility by race/ethnicity. I multiplied that figure by the share of students who were eligible (or not eligible). For example, 10% of all the city kids eligible for the lunch program in 2013 were white, and 73% of city kids were eligible for the lunch program. So .10 x .73 = 7% in the chart below. In other words, 7 percent of all urban fourth graders were white and poor.

Data source: NCES TUDA data 2013 for fourth grade math scores in large city public schools. Poor are students who are eligible for free or reduced priced lunch. Not poor are students who are not eligible. Data retrieved using custom data tables here.

Conclusion:

From this analysis, white gentrification doesn’t seem to be a driving force in the higher city scores. The percentage of whites who aren’t poor didn’t increase in the past decade. The bigger demographic shifts are the decline in the black population — both middle class and poor — and an increase in the Hispanic population.

I wish I had a finer way to measure income. There could be a churning of population within demographic categories. For example, middle class whites could have been replaced with wealthy whites. But since neither category qualifies for free lunch, this data wouldn’t capture that type of demographic shift.

Related stories:

Top performing cities on NAEP test have the least poverty — but some poor cities do surprisingly well

Low-income inner-city achievement gap starts to close, test scores of urban school districts improve faster than nation over past 10 years, Washington D.C. stands out