Exit exams don't work

Jonathan Schleifer

Countries with high school exit exams appear to have higher levels of student achievement, as indicated by PISA and some positive evidence from other countries that have used graduation exams. But have they worked in the United States? A recent Education Next forum failed to ask this essential question.

When fourteen public school teachers came together as part of Educators 4 Excellence-New York Teacher Policy Team on how to improve the use of testing in schools, they were taken aback by the depth of research showing the harmful effects of exit exams, which twenty-six states have adopted in one form or another.

There are two relevant research questions: Do exit exams have beneficial effects on students in terms of achievement or labor-market outcomes? And do exit exams have negative consequences, particularly on historically disadvantaged populations of students? The answers are no and yes, respectively. Here’s a representative, though not comprehensive, review:

Studies that find no benefit

  • In 2008, researchers examined a nationally representative sample of students and found no impact on achievement of high school graduation exams for any subpopulation of students, including low achievers.
  • A 2010 study of California’s exit exams used a regression discontinuity analysis to examine students who had just narrowly failed a tenth-grade exam. The results were disappointing: “The analyses show no evidence of any significant or sizeable effect of failing the exam on high school course-taking, achievement, persistence, or graduation for students with test scores near the exit exam passing score.”
  • A 2011 study of Florida requirements could not detect any positive effects of exit exams on student achievement (though it also could not find harmful effects of the policy).
  • A 2008 analysis used national data and found “no evidence that state [exit exams] positively affect labor force status or earnings or that the connections between state [exit exam] policies and these outcomes vary by students' race/ethnicity or the level of difficulty of state [exit exams].”

Studies that find harm

  • A 2005 study showed that, after controlling for demographic differences, students in exit-exam states performed worse on the SAT and had lower graduation rates than students in states lacking such requirements.
  • Exit exams in Massachusetts—which are sometimes held up as a positive example of graduation tests—appeared to increase dropout rates among low-income students.
  • A 2013 study found that exit exams drive increases in dropout rates, particularly among African American students.
  • A recent analysis in North Carolina found that graduation exams may actually lead to lower standards by driving students who just missed exam cut points in ninth grade to select less rigorous math courses later in high school.
  • A 2006 report found that, while Minnesota’s exit exam policy may have benefited students from low-poverty schools, it also harmed students in high-poverty settings, thus widening the achievement gap.

Studies that find harm and no benefit

  • A 2001 study of national data found no effect of graduation tests on twelfth-grade math or reading achievement, but did find increases in dropout rates among lowest-performing students.
  • A 2013 study found no positive effects of exit exams on employment or wage distribution, but disturbingly reported increases in incarceration and decreases in graduation due to such tests.

To be clear, there is some nuance here. As mentioned previously, there is some positive international evidence, and a few American studies that find no harm, yet I’m not aware of any rigorous studies in the United States that found positive effects of exit exams. As a Center for Education report put it, “[I]f exit exams were intended to raise student achievement, have they actually done so? Very little (if any) evidence exists to suggest they have.”

The evidence was also summed up well in a 2010 review of the research that looked at forty-six studies and found that exit exams “produced few of the expected benefits and have been associated with costs for the most disadvantaged students.”

The good news is that policymakers seem to be paying heed to the accumulated evidence. Recently, for example, Anne Hyslop, then of the New America Foundation, wrote “The Case Against Exit Exams,” a powerful critique of the current policy in many states.

In my home state of New York, students must pass a battery of five exams (with extremely limited alternatives) in order to attain a high school diploma. Discussions are underway to modestly ease such requirements by allowing students some limited choice over which exams they must pass.

But this does not go nearly far enough. Research and experience indicate that the policy of high school graduation exams should be examined and fundamentally restructured to ensure that no one test stands between a student and a high school diploma. Here’s hoping New York and other states follow the research and adjust course.

Jonathan Schleifer is the executive director of Educators 4 Excellence-New York, which works to involve teachers in the creation of education policy. He is a former middle school teacher in the Bronx.