Trapdoors

High school exit tests are the trapdoors of the education world. These are the tests that tie scores to high school diplomas and push students who miss the mark out of school into the streets, the unemployment lines, and the prisons.

A national uprising has highlighted the many ways the misuse and overuse of standardized testing hurts students. Now the effort to end high school exit testing may be its next step.

In the last few years, 10 states have repealed or delayed high school exit exams. California, Georgia, South Carolina, and Arizona even decided to issue diplomas retroactively to thousands of students denied them due to scores on discontinued tests. Although 13 states still use exit testing for diplomas and policies are in flux in several others, the number is down from a high of 27 states during the testing craze promoted by No Child Left Behind (NCLB).

There are several reasons for this retreat, including the research on exit testing, which clearly shows that exit tests don’t help the students who pass and hurt the students who don’t. They increase dropout rates and incarceration rates without improving college participation, college completion levels, or economic prospects for graduates in states that have them.

A 2014 report by the Gates-funded New America institute, The Case Against Exit Exams, declared, “New evidence has reinforced the conclusion that exit exams disproportionately affect a subset of students, without producing positive outcomes for most.” It found that “rigorous” exit testing was associated with lower graduation rates, had no positive effects on labor market outcomes, and, most alarmingly, produced a 12.5 percent increase in incarceration rates. Exit exams, the study concluded, “have tended to add little value for most students but have imposed costs on already at-risk ones.”

The introduction of the new, harder Common Core (CCSS) tests and the contradictions behind “college for all” rhetoric have been another factor. Even supporters of these new tests have acknowledged that using them as exit exams would produce dramatically negative effects. A report from the Carnegie Corporation predicted that, without an unprecedented increase
in educational supports for high school youth, the implementation of “college and career ready” tests as diploma requirements would cause the national high school graduation rate to plummet from more than 75 percent to the low 50 percent range. “As a political matter, it will be difficult for supporters of CCSS, already under attack from various corners, to sustain a system in which large numbers of students are denied diplomas,” said researcher Richard Kahlenberg.

Ready for What?

Exit testing relies on several related, flawed premises. One is that standardized testing can serve as a kind of “quality control” for high school graduates, guaranteeing that graduates are “college or career ready.” Another is that they have “predictive” value for future success in academic or workplace situations, and serve a useful gatekeeping function for institutions that ration access to opportunity.

But there is little evidence for these contentions. The tests don’t reliably measure what they pretend to measure—intelligence, academic ability, college readiness—and they don’t measure at all qualities that high schools should nurture in all young people, like responsibility, resilience, critical insight, and empathy. Although the passing or “cut” scores on standardized exit tests can be manipulated to produce varied outcomes, their main impact is to narrow access to opportunity for some, not to produce better preparation for all.

For example, both federally funded Common Core test consortia, PARCC (the Partnership for Assessment of Readiness for College and Careers) and SBAC (Smarter Balanced Assessment Consortium), claim that students who earn a “college ready” designation on their exams will be better prepared for college work. But there is little real evidence connecting scores on these tests with future college success. As Bill Mathis of the National Education Policy Center noted, “Both the PARCC and SBAC tests were administered to the nation’s schoolchildren without a single empirical study demonstrating the tests actually had the predictive capability they claimed.”

Last year, PARCC boosters began to cite a study by the research firm Mathematica that compared first-year college grades with scores on PARCC and on the Massachusetts Comprehensive Assessment System (MCAS), the state’s traditional high school exit test. PARCC CEO Laura Slover boasted that the study proved that PARCC “does what it was designed to do: measure college readiness.” But Mathis showed the actual “predictive validity” for both PARCC and MCAS was extremely low, accounting for less than 20 percent of the variation in college grades. “Standardized tests can predict scores on other standardized tests,” Mathis concluded, but they “cannot validly predict college readiness at any meaningful level.”

Like the SAT and ACT before them, scores on the new Common Core tests closely mirror existing patterns of inequality and privilege. Expanding their use would reinforce those patterns rather than disrupt them. Although sorting and labeling students with test scores occurs throughout elementary and middle grades, high school exit testing is one of the most blatant examples of how high-stakes testing hurts the most vulnerable students.

Numerous studies have shown that high school grade-point averages are a better predictor of academic success in college than standardized tests. This makes sense for several reasons. Even though grading is notoriously subjective and variable across subjects and schools, a student’s grade-point average reflects nuanced judgments from many teachers over multiple years. It also reflects more subtle abilities that contribute to academic success, including the ability to read and respond to institutional expectations, find appropriate support, develop persistence and resiliency, and solve problems. Grading, of course, can also reflect institutional and individual bias. But as problematic as that can be, the substitution of commercially produced standardized exams for teacher judgment further undermines education for students and contributes to the deskilling of the teaching profession.

A reasonable case can be made that test scores should be one part of a student’s high school record that includes credits earned, courses taken, activities, service, attendance, projects, and other indicators of academic accomplishment. But diploma decisions should be based on evaluation of that record by teachers and schools that know the student best, in collaboration with the students and their families. They should not depend on the make-or-break measure of a state “cut score.”

Case in Point: New Jersey

As a high school teacher, I saw New Jersey use an alphabet soup of state exams as exit tests over a 30-year period: the Minimum Basic Skills Test (MBS), the High School Proficiency Test 9 (HSPT9), the High School Proficiency Test 11 (HSPT11), the High School Proficiency Assessment (HSPA), and, most recently, PARCC. Each successive test promised and failed to ensure that all New Jersey high school graduates would be well prepared for college, careers, and citizenship.

But because the earlier tests were of modest difficulty and because the state also mandated an alternative pathway for students who did not pass, New Jersey’s relatively soft exit testing regime and generally strong public schools helped sustain one of the nation’s highest graduation rates. It currently ranks No. 2 in the nation at just over 90 percent. Gaps across districts and student groups continue to reflect inequalities in one of the nation’s wealthiest and most segregated states. But overall New Jersey’s current graduation rate for low-income students is nearly the same as the national graduation rate for all students (82.7 percent vs. 83.2 percent).

Enter the CCSS and PARCC. As one of the original “governing” PARCC states, New Jersey committed early on to replacing its state tests with PARCC exams. Even as the number of states using PARCC eroded from the original 25 to just six today, New Jersey doubled down, implementing six new PARCC high school exams, phasing out its traditional state tests, and narrowing the alternative graduation option. In 2016, the state joined New Mexico, which has the lowest high school graduation rate in the nation, in making PARCC a high school graduation requirement.

The implications came into focus as soon as the first PARCC scores were released. New Jersey’s senior class typically has about 95,000 students statewide. Only about 40,000 seniors passed the PARCC English language exam. Twenty thousand failed and another 30,000-35,000 didn’t take the test at all, either opting out in protest or otherwise avoiding a new and confusing layer of testing that was the target of unrelenting negative media coverage and parent protest.

The use of PARCC exams as a graduation standard after a single administration was educational malpractice. Many students had been told the scores wouldn’t count. The passing levels were set by the N.J. State Board of Education six months after the tests were given and applied retroactively to results from the previous spring.

The new policy was also illegal since it was imposed before the N.J. Department of Education (NJDOE) adopted the regulations needed to change state graduation requirements. In an effort to evade the law and diffuse opposition, NJDOE allowed students to use a maze of “transitional options,” from SATs to expanded use of the ASVAB military placement test to satisfy the exit test requirement. Under legal pressure from the Education Law Center (ELC, where I work as a policy advocate) and the American Civil Liberties Union of New Jersey (ACLU-NJ), the state also kept open a modified portfolio process that allows districts to graduate students through a review of student work samples supposedly aligned with PARCC frameworks.

However, the real purpose of the transitional plan was to grease the skids for the institutionalization of PARCC as the primary state exit test and to clamp down on New Jersey’s parent-led opt-out movement, which was one of the largest in the country. State data showed that “about 135,000 students in New Jersey across grades 3Ð11 did not take the state’s test in the spring of 2015.”

Beginning with this year’s 9th graders, only students who sit for all the PARCC high school exams can use the transitional graduation options, and by 2021 all options except PARCC and the narrowed portfolio review will be eliminated. For now, most seniors are still graduating through the transitional pathways, but the 2021 deadline looms.

The PARCC fiasco has provided another chance to end exit testing in New Jersey. ELC and ACLU-NJ have filed a lawsuit challenging the new graduation rules on behalf of the NAACP of New Jersey, the Latino Action Network, the Latino Coalition, and the Paterson Education Fund. Bills that would end the use of test scores to make graduation decisions for individual students are pending in the state legislature, and the leading Democratic candidates in next year’s governor’s race have publicly pledged to end both PARCC and high school exit exams. Parent groups like Save Our Schools NJ continue to press candidates and legislators on testing issues.

Punish or Prepare?

Contrary to popular impression, there is no federal mandate requiring high school exit testing. Since No Child Left Behind was passed in 2001, federal law has required testing once during grades 9-12 in math, English language arts, and science. The Every Student Succeeds Act retains this mandate. But the decision to tie diplomas to the results of those tests is totally a state decision.

There are real issues of inadequate preparation for many students leaving high school. But they are issues that standardized testing has helped create instead of solve. Test-based reform has undermined good education practice in numerous ways, narrowed curriculum, and wasted scarce resources.

When I went to college many years ago, “college for all” meant demands for open admissions, free tuition, and race, class, and gender studies. Today it refers to bootstrap notions of individual preparation validated by test scores. Putting an end to high school exit testing would be a step toward expanding opportunity for young people and putting the focus back on the resources and supports needed to provide it.