Statement Regarding Exit Examinations
Charles E. Ducommun Professor of Education, Stanford University
Faculty Director, Stanford Center for Opportunity Policy in Education
February 26, 2014
I have been asked to comment on the research that might bear on Rhode Island’s consideration of a policy establishing a high school exit examination, potentially in lieu of the system of graduation through a portfolio of student work that the state previously established. The Stanford Center for Opportunity Policy in Education, which I direct, has conducted extensive research on assessment and testing over the last decade, on which I draw for this statement.
The preponderance of research indicates that test-based requirements for graduation do not generally improve achievement, but do increase dropout rates. In 2011, the National Research Council reviewed the existing research on high school exit exam programs and came to this conclusion:
The evidence we have reviewed suggests that high school exit exam programs, as currently implemented in the United States, decrease the rate of high school graduation without increasing achievement.
Since then, additional studies have confirmed this outcome, finding that exit exams often reduce graduation rates and disproportionately discourage minority students from continuing in high school, without improving achievement for those students or others. Studies have raised concerns about reduced graduation rates, especially for African American and Latino students, English language learners, and students with disabilities; reduced incentives for struggling students to stay in school rather than drop out or pursue a GED; increased incentives for schools to encourage low-achieving students to leave school, especially when test scores are part of the state school accountability system, so as to improve the appearance of average school scores; narrowing of the curriculum and neglect of higher order performance skills where limited measures are used; and invalid judgments about student learning from reliance on a single set of test measures, a practice discouraged by professional testing experts.
Professional standards for educational testing indicate that a single test should not be the sole determinant when making important decisions about students.
- The Association of American Publishers, which represents the companies that publish standardized tests, states, “It is important both legally and technically not to put all the weight on a single test when making important decisions about students and schools. Rather, there must be multiple measures or indicators of performance to support important decisions.”
- The National Research Council Board on Testing and Assessment concluded that “no single test score can be considered a definitive measure of a student’s knowledge,” and that “an educational decision that will have a major impact on a test taker should not be made solely or automatically on the basis of a single test score.”
Rhode Island has already developed a system that uses multiple measures of performance in a portfolio of evidence about student competencies to make local decisions about graduation from high school. Research on such systems suggests that the decisions made on the basis of multiple student work samples, linked to ambitious standards, such as those reflected in the Common Core State Standards, are likely to be more valid and more supportive of the development of higher order thinking and performance skills. Such systems have been found to:
- encourage the teaching and evaluation of a more ambitious range of thinking and performance skills (including students’ abilities to conduct research, engage in extended problem solving, use technologies, and communicate effectively in many ways),
- consider different ways of demonstrating learning, which reduces the likelihood of inappropriate decisions for special needs students and English language learners,
- provide diagnostic information that guides improved instruction,
- reward student investment in school attendance and course performance, and
- maintain student engagement and increase the likelihood of students continuing in school through graduation.
While states that instituted test-only graduation systems have typically experienced decreasing graduation rates, states that introduced multiple measures systems of assessment have tended to maintain higher and steadier rates of graduation. Furthermore, studies have found that, in states where assessment systems have included extended writing and mathematics portfolios and performance tasks, teachers typically begin to assign more ambitious writing and mathematical problem solving, rather than teaching to multiple choice tests, and student achievement performance often improves on these higher order measures. Indeed, while causal claims are not possible, Rhode Island’s student scores on the National Assessment of Educational Progress mathematic and reading assessments have improved steadily over the last seven years, most steeply in secondary school. Writing scores have also improved since 2000.
Finally, researchers have found that assessment systems in which teachers look at student work with other teachers and discuss standards in very explicit ways can help schools develop more robust, shared deﬁnitions of quality. Evaluating work collaboratively rather than grading students in isolation helps teachers make their standards explicit, gain multiple perspectives on learning, and think about how they can teach to produce the kinds of student work they want to see. Where teachers do this, a number of studies have found that changes in teaching and schooling practices tend to occur.
If the goal of graduation policies is to both improve the quality of education and to ascertain what students know and can do in ways that will generalize to real world contexts and reflect higher order competencies, further development of Rhode Island’s current system of performance-based evidence for graduation may be likely to produce stronger outcomes than a shift to a traditional exit examination.
 National Research Council. (2011). Incentives and Test-Based Accountability in Education. Committee on Incentives and Test-Based Accountability in Public Education, M. Hout and S.W. Elliott, Editors. Board on Testing and Assessment, Division of Behavioral and Social Sciences and Education. Washington, DC: The National Academies Press.
 See for example, Sean F. Reardon, Allison Atteberry, Nicole Arshan, Michal Kurlaender (2009). Effects of the California High School Exit Exam on Student Persistence, Achievement, and Graduation. Stanford, CA: Stanford University, Institute for Research on Educational Policy and Practice.
 Linda Darling-Hammond, Elle Rustique-Forrester, & Raymond L. Pecheone (2005). Multiple Measures Approaches to High School Graduation. Stanford ,CA: School Redesign Network at Stanford University.
 American Education Research Association, American Psychological Association, National Council on Measurement in Education. Standards for Educational and Psychological Testing. 1999. Standard 13.7.
 The Association of American Publishers. Standardized Assessment: A Primer. Available on line at http://www.publishers.org/school/pdf/Testing%20Primer%20Revised.pdf.
 National Research Council. Heubert, J. P. and Hauser, R. Eds. High Stakes: Testing for Tracking, Promotion and Graduation. National Academies Press: Washington, DC. 1999. p. 3.
 Darling-Hammond, Rustique-Forrester, & Pecheone (2005).
 Darling-Hammond, L. & Adamson, F. (2014, in press). Beyond the Bubble Test: How Performance Assessments Improve Teaching and Learning. San Francisco: Jossey-Bass.
 Darling-Hammond & Adamson (2014).
Linda Darling-Hammond is Charles E. Ducommun Professor of Education at Stanford University where she is Faculty Director of the Stanford Center for Opportunity Policy in Education. She is a former president of the American Educational Research Association and member of the National Academy of Education. Her research and policy work focus on issues of educational equity, teaching quality, and school reform. In 2008, she served as director of President Obama’s education policy transition team. Her book, The Flat World and Education: How America’s Commitment to Equity will Determine our Future, received the coveted Grawemeyer Award in 2012. Her most recent book is Getting Teacher Evaluation Right: What Really Matters for Effectiveness and Improvement (2013).