Immigrants and Poor Kids Aren’t to Blame for Poor PISA Performance

Recently I wrote about the latest PISA results for American 15-year-olds in reading, math, and science. (See PISA Results Show We’re No. 1…in Spending, Not Performance.) One commenter stated that she saw “very little control for demographics” with international assessments such as PISA.

In particular, the commenter objected that while other countries may surpass the United States in terms of the quantity of immigrants and still perform better, what matters is the quality of those immigrants in terms of their educational backgrounds. (For countries’ immigration rates, see Figure II.3.6, p. 76. See here for a performance snapshot of participating countries.)

This is a valid point—one that the Organization for Economic Cooperation and Development (OECD), which administers PISA, analyzes in great detail.

The impact of immigration status on PISA student performance varies, depending in part on a country’s immigration policy. Australia, Canada, and New Zealand, for example, have very selective policies that favor better educated and more well-to-do immigrants. The United States has less selective immigration policies that tend to favor keeping families together, meaning we’re more open to poor and less-educated immigrants (See here p. 72, and here p. 11).

Yet it would be wrong to conclude that less educated or disadvantaged immigrants explain the United States’ PISA performance. What follows is a summary of some of the leading findings from the OECD on the effects of students’ immigrant status.

On the 2003 PISA assessment, 9 percent of students across OECD countries had an immigrant background. By 2012, the share of immigrant students across OECD  countries with comparable data increased to 11 percent yet the mathematics performance gap between immigrant and non-immigrant students shrunk by about 10 score points (See here p. 74). So greater immigration doesn’t equal lower performance.

Countries with selective immigration policies tend to draw immigrant students with well educated parents. However, the OECD finds that “adjusting for parental educational seems to explain at best 25 percent of the difference between the outcomes of immigrant and non-immigrant students. Differences in parental education also fail to explain why, in some cases, second-generation students do not perform as well as first-generation students” (See here p. 13).

Not surprisingly, immigrant students tend to be socioeconomically disadvantaged in comparison to non-immigrant students (See here p. 14). Yet as the OECD analysts explain:

The wide performance differences between students of similar socioeconomic status and a common country of origin suggest that schools and education policy in the host countries influence these students’ performance. While immigration policies, similarities between the immigrants’ and the host culture, and other social policies also explain some of these differences in performance, some education systems appear to facilitate the integration of immigrant students better than others. … The fact that immigrant students from the same country of origin, cultural backgrounds and socioeconomic status perform so differently across host countries indicates that education and social policy can have an impact not only on these students’ performance but also on how prepared they are to make the most of available opportunities in their host countries (See here p. 84).

A leading policy recommended by the OECD is school choice because by decoupling schooling from housing, school choice promotes better socioeconomic integration and increases the likelihood that immigrant and disadvantaged students can access schools with better teachers (See here pp. 13-14).

Experts could control for student demographics every which way to Sunday, and the Unites States would still not shoot to the top. On the contrary, once scientific controls are applied we sink lower in the rankings, from 36 to 38 out of 63 OECD countries and economies (See Figure II.2.5, p. 42).

This means that based on our demographics we should be doing better—not worse—than our global competitors.

However tempting it may be to blame the United States’ academic performance on poor children and immigrants, the reality is we have a poorly performing schooling system that needs a healthy dose of competition—not more excuses—to improve.