Warning Signs Identify Children Likely To Fail High School Exit Exam
As Early as Fourth Grade, Predictions of Success or Failure on the High-Stakes Test Are Reliable
SAN FRANCISCO, California, June 10, 2008 –Children who are at risk of failing the California High School Exit Exam can be accurately identified as early as the fourth grade, according to a study released today by the Public Policy Institute of California with funding from The William and Flora Hewlett Foundation. The study suggests that shifting resources to struggling students in early grades will be a more effective way to improve achievement than the state’s current approach of focusing on students in the last year of high school.
Identifying both the characteristics that predict exam performance and the optimal age to give a student remedial help has important implications for parents, teachers, school administrators, and policymakers. The exit exam is the only part of the California accountability system with direct consequences for students, and the failure of many students to pass it, even after multiple attempts, is cause for concern. State-funded efforts to boost students’ skills are concentrated on students in 12th grade who are at risk of failing the test and those who already have left school and were unable to pass it. The study concludes that these efforts alone are unlikely to be effective. Of the students who leave 12th grade without passing the exam, few re-enroll in school or take the test again. Allowing students in the class of 2006 the option to retake the exam the following year raised the passing rate only marginally, from 90.4 percent to 90.7 percent.
“Now that we know we can identify students in the earlier grades who are likely to fail, it makes sense to help them while they’re still in school and while they’re being taught the basic skills that will be tested later on,” says Julian Betts, a PPIC adjunct fellow, who co-authored the study with Andrew Zau, a senior statistician at the University of California, San Diego. “We don’t have to wait until the 11th hour or even worse, after students have failed and left the system.”
Like high school exit exams in many other states, California’s has been the focus of legal and legislative challenges from the start. The test, which became a requirement for graduation in 2006, covers math up to the eighth-grade level and English language arts up to the 10th-grade level.
The PPIC researchers were able to follow individual students over time using detailed data from the San Diego Unified School District that included grades, test scores, and academic environment. SDUSD, the state’s second-largest school district, mirrors the demographics of other large districts and its students’ performance on the exit exam reflect those of the state as a whole: 75 percent pass the test in 10th grade, and students who are English language learners, African American or in special education are significantly less likely to pass by the end of 12th grade.
The study concludes that information available in the early grades is such a strong indicator of a child’s future performance on the exit exam that it is possible to predict who will fail nearly as accurately in fourth grade as in ninth grade. Among the key findings:
- Fourth-grade GPA is an especially strong predictor of success on the exam. For every one-point increase in GPA, students increase their likelihood of passing the test by 11.6 percent. In later grades, GPA is a less significant predictor of success on the exit exam.
- Classroom behavior in the elementary grades is nearly as important. Classroom behavior is more important than math and reading test scores in forecasting test performance. San Diego teachers evaluate students in categories such as “follows directions,” “classroom behavior,” and “self-discipline.” The PPIC study translates these measures into a “behavior GPA.” For every one-point increase in the behavior GPA in fourth grade, students increase their likelihood of passing the exit exam in 10th grade by 3.7 percent and in 12th grade by 5-6 percent.
- Test scores are less powerful predictors, and they differ across grades. Math test scores in grades 4-6 are better indicators of success than English test scores, probably because the exam tests eighth-grade math skills. In grades 7-9, English test scores are better forecasters of success, probably because the English section of the exam tests 10th-grade English skills.
- English learner status matters less in early grades than later on. Students who are classified as English learners in fourth grade are no less likely to pass the exam than their peers who are otherwise similar, but students who are still classified as English language learners in ninth grade are much less likely to pass the test.
- High school teachers’ qualifications play a minor role in test performance. Teachers’ demographic background, education level, years of teaching experience, and credentials have only a small effect on students’ chances of passing the exit exam. This is relevant in light of a lawsuit, Valenzuela vs. O’Connell, filed in 2006 and later settled, which contended in part that the exit exam should be not required because some students attend high schools where the teachers are not highly qualified. The PPIC study finds that even if teacher qualifications were equalized across high schools and among students within high schools, passing rates would change very little.
The study points out that it is not just the students who fail the exit exam who are a source of concern. There are many more who barely pass it. Roughly a quarter of students fail both parts of the test in 10th grade, a troubling result on an exam that tests relatively low-level skills. A student who can barely pass a test of eighth-grade math skills and 10th-grade English is probably not well-prepared for a successful career.
Providing remedial help to students in earlier grades when they are learning the material that will be on the exit exam would have a number of benefits beyond raising passing rates on the exit exam. It could improve results on the California Standards Tests and help schools meet achievement goals required by the federal No Child Left Behind law. Help with reading in early grades would benefit students in all other subjects, a particularly important benefit for English learners.
The study recommends an expansion of tutoring on a limited trial basis in randomly selected schools to identify the grade levels at which remedial help is most effective. Different approaches, from after-school tutoring to professional development for teachers, can be tested to pinpoint the most useful. These tests would provide a rigorous research basis for policymakers to determine when and how to best ensure the success of all students.
The Public Policy Institute of California is a private, nonprofit organization dedicated to improving public policy in California through independent, objective, nonpartisan research on major economic, social, and political issues. The institute was established in 1994 with an endowment from William R. Hewlett.