Monday, August 15, 2016

Is our children learning?

Anxiety about the quality of American schools has since the 1980s been driven at least in part by American students' mediocre performance on international tests. In the 2012 Program for International Student Assessment (PISA), American 15 year olds finished 17th among 34 OECD nations in reading, 20th in science and 27th in math, which a frustrated U.S. Education Secretary Arne Duncan called "a picture of educational stagnation" (Simon, cited below). Younger students did somewhat better in the 2011 Trends in International Mathematics and Science Study (TIMSS), with 4th graders finishing 11th among 57 countries in math and 7th in science, and 8th graders finishing 9th in math and 10th in science. In the same year, American 4th graders finished 6th among 53 countries in the Progress in International Reading Literacy Study (PIRLS). (Results from 2015 TIMSS and PIRLS will be available this November.) If maintaining America's favorable position in an increasingly competitive, global economy depends on how well-prepared young people are as they enter the work force, these are not good signs, though some might be taken as at least OK.
Johnson STEAM Academy, Cedar Rapids, IA
Lowest test scores in the metro area
82.2 percent of students eligible for free lunch
(photo by author)
Viewing American students as a homogeneous group with mediocre levels of achievement produces, however, a misleading picture of American education. (There are also serious questions about the validity of the test results as statistical indicators, which reservations I share but which I am for the sake of the present discussion going to ignore.) To take just one dimension of diversity, the socio-economic status of the student body, students from low-poverty schools had average PISA scores similar to the top countries in all three categories, while students from low-poverty schools averaged scores that would have put them near the bottom of the OECD (Simon 2013; see also Shultis 2012). Geography matters, too: Several states score consistently much higher than the U.S. national average across-the-board on National Assessment of Educational Progress (NAEP) scores: Minnesota, Vermont, New Hampshire, Massachusetts, Connecticut and New Jersey. Wisconsin, Indiana and Virginia are close behind them. Whatever's going on there is clearly different in a good way than in the states that score much lower than the national average: Alabama, Mississippi, Louisiana and New Mexico. How is it then useful to treat the "crisis" as uniform nationwide?
Westfield Elementary School, Robins IA
Highest test scores in the metro area
6.0 percent of students eligible for free lunch
(photo from school website)
In the Cedar Rapids metropolitan area test scores at the individual school level vary according to the poverty of the student body. Here are the global ratings from and the percentage of students at each public school whose family income qualifies them for free lunch:

0-10 %
11-20 %
21-30 %
31-40 %
41-50 %
51-60 %
61-70 %
71-80 %
81-90 %

es es es




es es


HS MS MS IS es es es
 HS es


es es
MS es es es


IS es





Es es
es es es


MS es


Of the 11 schools with 20 percent or less of their students qualifying for free lunch, all but two score at least 7. Of the seven schools with more than 60 percent qualifying, all score 1 or 2. (There are also some inter-district effects that are hard to explain without further investigation. And I can no longer find SES data for private schools in Iowa. About ten years ago, I did a similar examination of public and private elementary schools that showed (a) private elementary schools had very low rates of free-and-reduced lunch students, and (b) their tests scores were comparable to, not better than, public schools with very low rates of free-and-reduced lunch students. And now it seems this handy metric is going to be lost even for public school comparisions.)

To be sure, as reporter Stephanie Simon notes, "poverty alone does not explain the lagging results in the U.S. Vietnam is a poor nation, yet it outscored the U.S. significantly in math and science." The problem is the nature of poverty in America, a rich country with areas of concentrated poverty particularly in central cities that are isolated from the mainstream economy, cut off from avenues to economic opportunity (Chetty et al. 2014, 2016). In many places this reality is enforced by zoning restrictions that keep the poor out of and away from economically-successful areas, including the schools (Rothwell 2012). Being poor anywhere is difficult, but being poor in America presents a particular set of challenges that are reflected in educational performance.

It should be clear by now that efforts to improve American educational outcomes need to address the subgroups with distinctively low performance: children from poor families living in areas of concentrated poverty. Most analysts' approaches focus either on individuals or society.

The predominant individual approach is some form of school choice, which at least means the ability to choose among public schools within or across school districts. The ability to form charter schools, as well as government vouchers that can be applied to private school tuition, further increase the range of choices. The core assumption--that school performance is driven largely by the talent and effort of the staff, and that market-style competition would spur higher quality and more innovation--seems to me flawed. If the school staff were the causal variable, we'd see performance vary more randomly across the map rather than being so easily predictable by local SES.

Nevertheless there are some reasons to think seriously about school choice: Giving people choices might increase their feelings of personal efficacy as well as responsibility for outcomes; it reminds school staff, to the extent they need reminding, that they are accountable for student learning; it might be a short-term way to start getting the improved social mobility that Chetty et al. argue would come from physical mobility; and it would be a way to reassure new middle-class residents that they could seek the best education possible for their children (Duany et al. 2010: 172, though n.b. their preferred solution is a consolidated regional school district: "Only if city schools are able to share the resources of those in the wealthier suburbs can large numbers of parents be convinced to locate their families downtown").

Efforts to improve racial integration attack the performance problem at a more societal level. Gary Orfield and colleagues (2016) argue from numerous studies that racial as well as economic segregation continues to be linked to inferior economic opportunity, and question the lack of policy "initiatives to mitigate spreading and deepening segregation in our nation's schools." Other societal approaches include improving teacher recruitment and training, essentializing the curriculum ( like "Common Core") and increasing spending on education. (For a list of plausible education policy questions for presidential candidates, see Hansen 2016.) Each of these societal approaches, though, takes the distribution of resources and opportunities in American society as givens and tries to do the best they can with them.

I realize the potential for a mixed message here, and I would not for a moment suggest that either the staff of high-poverty schools or low-income parents should ever give up striving to be the best they can be. There is a lot that individuals and schools can do to make things better (Gran 2016). But as a community, as a country, we have to confront how much poverty, particularly concentrated poverty, affects academic performance. Nothing, really, short of a frontal assault on inequality of opportunity will do.

EARLIER POST: "Starting a Conversation about Education," 16 August 2015,

Andres Duany, Elizabeth Plater-Zyberk and Jeff Speck, Suburban Nation: The Rise of Sprawl and the Decline of the American Dream (North Point Press, 10th anniversary ed., 2010)
Michael Hansen, "What We Need to Know from Candidates on Education Policy," Brookings, 9 August 2016
Gary Orfield, Jongyeon Ee, Erica Frankenberg and Genevieve Siegel-Hawley, "Brown at 62: School Segregation by Race, Poverty and State," The Civil Rights Project, 16 May 2016
"PISA 2012 Results,"
Jonathan Rothwell, "Housing Costs, Zoning, and Access to High-Scoring Schools," Brookings, 19 April 2012
Steven Shultis, "It's the Schools, Stupid (Part I)," Rational Urbanism, 22 September 2012
Stephanie Simon, "PISA Results: 'Educational Stagnation,'" Politico, 3 December 2013,
"TIMSS 2011 Results,"

SEE ALSO: Elizabeth Kneebone, "The Growth and Spread of Concentrated Poverty," Brookings, 31 July 2014, 

No comments:

Post a Comment

Do bicycle boulevards need a purpose?

I was surprised last weekend to find the place where we were staying was on a bicycle boulevard. A bicycle boulevard is "a street ...