New America Education Policy Brief THE CASE AGAINST EXIT EXAMS Anne Hyslop #exitexams edcentr.al/exitexams About the Author Anne Hyslop is a Policy Analyst in New America’s Education Policy Program. She can be reached at hyslop@newamerica.org. Acknowledgments My appreciation to the researchers, analysts, and experts who graciously shared their insights and knowledge with me and to those who took the time to review an earlier draft of this report, especially Bob Rothman, Diane StarkRentner, and Kedda Williams. Thanks also to Kevin Carey and Laura Bornfreund for their guidance in writing and editing, Ross van der Linde for his design and production support, and to Honey Ghods and Cary Lou for their research assistance. This report was funded by the Bill & Melinda Gates Foundation. New America thanks the foundation for its support. The views expressed in the paper are those of the author alone. About New America New America is a nonprofit, nonpartisan public policy institute that invests in new thinkers and new ideas to address the next generation of challenges facing the United States. The New America Education Policy Program’s work is made possible through generous grants from the Alliance for Early Success; the Annie E. Casey Foundation; the Bill & Melinda Gates Foundation; the Evelyn and Walter Haas, Jr. Fund; the Grable Foundation; the Foundation for Child Development; the Joyce Foundation; the Kresge Foundation; Lumina Foundation; the Pritzker Children’s Initiative; the William and Flora Hewlett Foundation; and the W. Clement and Jessie V. Stone Foundation. © 2014 New America This report carries a Creative Commons license, which permits non-commercial re-use of New America content when proper attribution is provided. This means you are free to copy, display and distribute New America’s work, or include our content in derivative works, under the following conditions: Attribution. You must clearly attribute the work to New America, and provide a link back to www.newamerica.org. Noncommercial. You may not use this work for commercial purposes without explicit prior permission from New America. Share Alike. If you alter, transform, or build upon this work, you may distribute the resulting work only under a license identical to this one. THE CASE AGAINST EXIT EXAMS CONTENTS The Call: More College- and CareerReady Students 2 The Context: History of High School Exit Exams 5 The Collision: Higher Standards vs. Higher Attainment 9 The Challenge: Holding Students Accountable for College and Career Readiness 12 The Changes: States’ High School Assessment Plans 13 The Changes: States’ Student Accountability Policies 21 The Case: Moving Away from High School Exit Exams 31 2 THE CASE AGAINST EXIT EXAMS THE CALL: MORE COLLEGEAND CAREER-READY STUDENTS High school graduation is the final act in the American teenage rite of passage. Each spring, “Pomp and Circumstance” plays, speeches are delivered, and cameras flash as graduates cross the stage to receive their diplomas. But finishing high school provides students with something far more valuable than photos and memories. It is a prerequisite for life-long economic stability. Without a high school degree, college—let alone the federal financial aid to pay for it—is off the table. And good luck trying to get a well-paying job, or any job, as a high school dropout. But a high school diploma is no guarantee for longterm success either. Forty years ago, seven of ten jobs only required a high school education, or less. Today, the opposite is true: By 2020, 65 percent of all available jobs will demand some postsecondary training, often a two- or four-year degree (Figure 1).1 Unfortunately, many students never get that far. In 2012, two-thirds of high school completers nationally enrolled in higher education within a year of earning their diplomas, but many of them will not complete their degrees, even years later.2 According to the National Student Clearinghouse Research Center, just over half of students graduate within six years of starting a two- or four-year degree.3 Newly released federal data are equally grim. Among over 13,000 high school sophomores in 2002, nearly half had no postsecondary degree ten years later. These students were more likely to be unemployed than their peers who had completed a college program, whether undergraduate certificate, Associate’s degree, or Bachelor’s degree (Figure 2).4 Given the clear economic benefits, why do so many high school graduates drop out of college? Poor preparation is partly to blame. A high school diploma is not synonymous with postsecondary readiness—and students pay the price for this mismatch between high school and higher education in time, in tuition, and in their chances for earning a degree.5 Lacking the required skills and knowledge, many high school grads are placed in noncredit-bearing, remedial coursework in college to master basic content they should have learned in high school. Over half of students entering a two-year college require remediation, and the same is true for 20 percent of students at four-year colleges. The numbers are even worse for low-income and minority students at these institutions.6 Too often, remediation is a college completion death sentence. Students are diverted from college-level work, requiring more time to earn their credentials and spending more money in the process, even though research shows many could likely do well in more difficult courses.7 And once students enrolled in remedial classes actually finish them, many never go on to complete related, college-level work, and they are less likely to graduate than students who avoided remediation. According to Complete College America, remediation is a “Bridge to Nowhere.” Only one of ten community college students placed in remediation finishes a two-year degree within three years.8 These challenges create an uncomfortable dilemma for those trying to improve educational attainment. On the one hand, a high school education has never been more important. Without it, a student’s future is bleak, and educators know it. High school graduation is more than just a rite of passage—it is practically a “right” of passage, and the decision to withhold this right is fraught from a legal perspective and from an educator one.9 Teachers care deeply about students and want to see them succeed. Further, high schools are held accountable for their graduation rates, creating additional incentives to ensure as many students graduate as possible. On the other hand, many high school diplomas are not rigorous enough. Students complete their requirements and don their caps and gowns only to walk straight from their graduation ceremony into remediation. These students may attain a high school degree, but they are less likely to attain the postsecondary credential they also need. Increasing numbers of states have responded to this dilemma with a particular solution: high-stakes exit exams. Up from 18 states in 2002, 24 states now require high school students to pass a state assessment in various subjects in order to graduate. In theory, this motivates them to meet higher standards and prevents unprepared students from earning a high school diploma.10 But this testing is controversial. And forced to choose between holding students accountable for meeting high expectations or denying students a diploma, states prioritize the latter and set the passing bar low enough so that the majority of test-takers can pass on their first try (often in ninth or tenth grade). Many states also offer students multiple re-takes of the exam, special waivers, or alternate testing options.11 THE CASE AGAINST EXIT EXAMS Figure 1. Proportion of Jobs Requiring a Postsecondary Education (1973-2020) Proportion of jobs requiring less than postsecondary education Proportion of jobs requiring postsecondary education or more 44% 41% 35% 56% 59% 65% 1973 1992 2010 2020 72% 28% Source: Anthony P. Carnevale, Nicole Smith, and Jeff Strohl, “Recovery: Job Growth and Education Requirements through 2020,” Georgetown University Center on Education and the Workforce, June 2013. Figure 2. The Link Between Higher Educational Attainment and Employment The students in the ELS:2002 sample (13,000 high school sophomores followed for ten years) who had earned Bachelor’s degrees saw the lowest rates of unemployment, while students who lacked postsecondary credentials had the highest unemployment rates. The small proportion of the sample that dropped out of high school was the worst-off: over a quarter was unemployed. 3+13+3210933 Educational Attainment in 2012 of 13,000 High School Sophomores in 2002 HS Dropout (3%) HS Dropout (26%) HS Diploma (13%) BA Degree (33%) Some College (32%) AA Degree (9%) Unemployment Rate in 2012 of 13,000 High School Sophomores in 2002 BA Degree (5%) AA Degree (10%) Certificate (12%) HS Diploma (15%) Some College (14%) Certificate (10%) Source: E. Lauff and S.J. Ingels, Education Longitudinal Study of 2002 (ELS:2002): A First Look at 2002 High School Sophomores 10 Years Later (NCES 2014-363) (U.S. Department of Education. Washington, DC: National Center for Education Statistics, 2013). 3 4 THE CASE AGAINST EXIT EXAMS These tensions between higher standards and higher educational attainment will only amplify as states embark on the latest effort to increase rigor in K–12 education: the Common Core State Standards, a set of academic standards adopted by over 40 states that define what students need to know by the end of high school to be ready for college and careers. Students will likely be caught in the middle of this shift. Transition to the new standards creates a ripple effect across the entire K–12 system, from testing to textbooks to accountability systems, as policies must be updated to reflect the new college- and career-ready expectations. But how quickly should students be expected to master the more challenging standards? The Common Core-aligned tests being developed by two state consortia are expected to be harder than ones states administer now, and, particularly at first, fewer students will pass them. Given that so many states require students to demonstrate mastery, via proficiency on a state exam, in order to graduate from high school, should these policies remain in place with the new standards? What could happen if they do? To answer these questions, this report considers evidence from past high school exit exams, while taking stock of states’ current high school assessment policies and any planned changes to them as they transition to college- and career-ready standards.12 We find that states, in general, have suffered many of the negative consequences of high-stakes exit exams, like higher dropout rates, but reaped few of the promised rewards, including better college and workforce outcomes. Further, many states are in danger of repeating history as they implement the Common Core or similar college- and career-ready standards. As many as 21 states plan to continue their exit exams in English Language Arts and math during the shift to higher standards. Further, 10 states may use Common Core-aligned tests designed by one of two state consortia as exit exams once old assessments are phased out. Of these states, six consortia members are choosing a transition strategy that is particularly challenging—and complex—in its execution, as they attempt to maintain continuity in their exit exam policies and hold students accountable, while phasing in new assessments and higher standards (Figure 3). States’ new assessments are being designed to include more critical thinking and complex items than previous standardized tests, with performance tasks, computeradaptive features, and open-response questions. While better assessments are certainly welcome, a better assessment used as an exit exam is still an exit exam. States also need better policies for how their assessments are used. Exit exams have forced states to choose between two worthy goals: enforcing higher academic standards and making higher education an option for as many students as possible. But states do not have to make that choice. Unlike school accountability, which is shaped in part by federal policy, the stakes attached to tests for students are the result of state and local policy choices alone. More important, states have other options, beyond exit exams, that allow them to pursue both goals, without pitting them against one another. Figure 3. As States Begin Using College- and CareerReady Standards and Tests in 2015: 21 states plan to continue high school exit exam policies in ELA and math 10 of these 21 states plan to use PARCC or Smarter Balanced tests as exit exams 6 of these 10 states face a particularly complicated transition for their high school exit exam policies Source: New America analysis. THE CASE AGAINST EXIT EXAMS THE CONTEXT: HISTORY OF HIGH SCHOOL EXIT EXAMS The idea to raise graduation standards first gained traction during the economic uncertainty and stagflation of the mid-1970s, as critics argued that a high school diploma was no longer a meaningful credential for employers, due, in part, to weak standards and social promotion. In response, states began to require high school students to take standardized tests of basic skills, and by the early 1980s, 19 states not only administered such minimum competency exams (MCEs), but also required students to pass them to graduate.13 These early high school exit exams (HSEEs) were seen as one way to elevate and signal the value of a diploma to students, employers, and the public, but the early versions were typically pegged to basic, often middle school-level, skills. It took the landmark 1983 publication of A Nation at Risk by Secretary of Education Terrel Bell’s National Commission on Excellence in Education, and its claim that a “rising tide of mediocrity” was directly linked to these minimum expectations, for states to reconsider the design of their exit exams.14 As the modern standardsbased reform movement took root from the 1980s through the 2001 enactment of federal standards-based Figure 4. The Goals Behind High School Exit Exams Clear Standards for High School Learning Greater Student Motivation to Earn a Diploma Clear Signal to Higher Ed of Students’ Knowledge Clear Signal to Employers of Students’ Skills Increase High School Achievement Increase Postsecondary Success 5 6 THE CASE AGAINST EXIT EXAMS accountability in No Child Left Behind (NCLB), states steadily updated their exit exam policies and increased the rigor of the tests over time. But regardless of the difficulty of states’ exit exams over the years, they all shared common attributes. Each iteration was premised on two basic goals (Figure 4). First, an exit exam would increase student achievement overall by setting a clear standard for high school learning and by motivating students to earn the more meaningful, and valuable, diploma. Second, an exit exam would improve students’ postsecondary success by providing a clear signal to employers or colleges that graduates possessed valuable skills. The reaction to exit exam policies has also been consistent. Linking high school graduation to a test score has always been met with trepidation by those who worried that the new requirements would exacerbate high school dropout and lower educational attainment, particularly for disadvantaged students. Many also worried exit exams would produce unproductive responses from schools and educators seeking to increase passing rates at all costs, even if that meant narrowing high school curricula or diluting the standards. Research on decades of exit exam policies have produced very few conclusive findings on whether these twin goals were accomplished, but in many studies, the fears were confirmed. In short, typical students do not appear to be any better off after the exit exam policy, and those that were already vulnerable, including low-income and minority students, often became more so. In one of the broadest findings, a blue ribbon commission formed by the National Research Council, the Committee on Incentives and Test-Based Accountability, found that high school exit exams nationwide had not increased student achievement, but rather decreased graduation rates by two percentage points, on average.15 Although the experience in some states, like Massachusetts, has been more positive, even in this best-case scenario, there have been negative effects, particularly for at-risk students, as a result of exit exams. (See sidebar, “Massachusetts: A Model High School Exit Exam Policy?”) Massachusetts: A Model High School Exit Exam Policy? Massachusetts is often considered the archetype for effective high school exit exam implementation, and for good reason. The state is known for its high-quality, rigorous standards and tests, and its impressive student outcomes.16 The state first used its high school assessment, the MCAS, as an exit exam for the class of 2003—and, critically, this policy was accompanied with added funding and resources for schools to help students meet the new standard. The passing score was also set at a “Goldilocks” level initially, according to former State Commissioner David Driscoll, and increased gradually over time so that students and schools could adjust to the new requirements.17 And many in the Bay State believe that the exit exam increased students’ motivation to learn more rigorous material, and schools’ motivation to help them. As Paul Reveille, a former state Board member and Secretary of Education, put it, “People underestimated the effort of teachers and students once they focused on a clear set of goals.”18 exam policies have exacted a cost on a particular subgroup of vulnerable students. For most students in the Bay State, barely failing the MCAS in 10th grade has no effect on their high school graduation prospects. But 2010 research by John Papay, Richard Murnane, and John Willett found that low-income, urban students who barely fail the math portion of the MCAS have graduation rates that are eight percentage points lower than similar students that barely pass. These students are also four percentage points more likely to drop out of high school the year following the test.19 While Massachusetts allows retakes of the exam, low-income urban students are not as successful in their subsequent attempts at the MCAS as similarly skilled, but more affluent, suburban students. All of this may be true, but even in the best case implementation scenario—the Massachusetts example—exit Now that the Bay State is implementing new collegeand career-ready expectations in the Common Core State Standards, the question is whether there are policy alternatives that can still encourage all students to reach for the higher standards, but do so in a way that does not limit opportunity for the state’s most disadvantaged students. Along similar lines, a 2010 meta-analysis on the effects of high school exit exams, including minimum competency versions and newer, more rigorous tests, found that, in general, the “evidence indicates that exit tests have produced few of the expected benefits for students overall and nearly all of the expected costs for disadvantaged and at-risk students.”20 Across 46 studies, the authors examined student outcomes related to exit exam policies and high school achievement, high school graduation, postsecondary education, and workforce participation, as well as the effects of failing an exit exam on students in terms of achievement, dropout, and postsecondary outcomes. Across all outcomes studied, exit exam policies often resulted in mixed or inconclusive findings, and were rarely associated with positive outcomes for students (Figure 5). THE CASE AGAINST EXIT EXAMS Figure 5. All Costs, Few Benefits: How Exit Exam Policies and Student Performance on Them Affects High School Attainment and Postsecondary Success How to read this table: MCE refers to older minimum competency exams. HSEE refers to newer standards-based exit exams. Indicates a decrease in the performance outcome (e.g., lower K–12 student achievement, lower dropout rates and GED attainment, lower rates of college success, and lower rates of workforce success) Indicates an increase in the performance outcome (e.g., higher K–12 student achievement, higher dropout rates and GED attainment, higher rates of college, and higher rates of workforce success) ? Indicates no general effect on the performance outcome (e.g. no change in K–12 student achievement, dropout rates and GED attainment, college success, and workforce success) Indicates inconsistent or inconclusive effects on the performance outcome Effect of Exit Exam Policy Expected Effect? Actual Effect? (MCEs) ? K–12 Student Achievement Dropout and GED Attainment For Which Students? There is no impact on overall student achievement, or for lowperforming students. MCEs did not exacerbate typical student dropout, but may increase dropout for low-performing students and for black males. MCEs are not conclusively linked to greater GED attainment or delays in graduation. More difficult HSEEs, are associated with higher dropout rates and GED attainment rates, and delays in high school graduation, especially for minority students and students in high-poverty areas. ? ? College Success ? For Which Students? It is inconclusive whether students, especially lowperforming ones—those the policy most sought to motivate— improve. College Readiness Workforce Success Actual Effect? (HSEEs) Research has found a negative association between HSEE policies and state ACT and SAT scores, but the studies’ methodologies prohibit drawing definitive conclusions. There is no definitive evidence that college enrollment rates increase with MCEs. Some studies found positive effects, especially for particular student groups, but others found no association. There is no impact on college enrollment rates for typical students. The effects on employment or wages for students overall are mixed (both positive and negative). Some studies of only high school graduates found positive effects on wages for specific subgroups and cohorts. HSEEs are likely not associated with higher employment or earnings for students overall. However, heterogeneous effects (both positive and negative) were observed for particular subgroups of students. 7 8 THE CASE AGAINST EXIT EXAMS Effect of Exit Exam Performance Expected Effect? Actual Effect? (HSEEs) For Which Students? K–12 Student Achievement Students barely missing the HSEE passing score on the first try see no effects on their subsequent achievement, and do not appear to be more motivated or more discouraged. Dropout Students that score below the HSEE passing score on “last chance” exams have an increased probability of high school dropout, and the effect seems to be stronger for low-income, minority, and lowachieving students. College Success Students that fall just below the HSEE cut score are less likely to attend college, and those that do, earn fewer credits compared to students just above the cut score. Workforce Success Failing a “last chance” HSEE is associated with reduced earnings just after high school, but disparities decreased over time, relative to students that just barely passed the last chance exam. Source: Jennifer Jellison Holme, Meredith P. Richards, Jo Beth Jimerson, and Rebecca W. Cohen, “Assessing the Effects of High School Exit Examinations,” Review of Educational Research, December 2010 (80): 476–526. doi: 10.3102/0034654310383147 Since the 2010 meta-analysis, new evidence has reinforced the conclusion that exit exams disproportionately affect a subset of students, without producing positive outcomes for most. A 2013 study from Olesya Baker and Kevin Lang found that more rigorous exit tests, not MCEs, were associated with lower graduation rates, particularly in states that had not previously had a MCE policy in place. Further, the lower graduation rates were not fully offset by increased GED attainment. As with other studies, Baker and Lang found that there was no relationship between exit exam policies and labor market outcomes. They also examined incarceration rates as another long-term cost of exit exam policies. They found that both MCEs and more difficult HSEEs increased the likelihood of incarceration, but the findings were only significant for the more rigorous tests. In fact, these kinds of exit exams were associated with a 12.5 percent increase in incarceration rates.21 “ Exit exams have tended to add little value for most students but have imposed costs on already at-risk ones Another 2013 study, by Steven Hemelt and Dave Marcotte, examined the relationship between exit exams and high school completion, with a particular focus on dropout rates by grade and the impact of alternate diploma pathways. They found that, across all grades combined, exit exams had no effect on dropout, but did increase dropout by 11 percent when only examining the twelfth grade cohort. Further, in states with no alternate pathways for students that failed the exit exam, dropout rates were 23 percent higher than in states that provided other routes to graduation. And yet again, there were greater negative effects for particular subgroups. Exit exams increased dropout for Hispanic and black students, particularly when there were no alternate routes to graduation.22 There is also new research that clarifies how just failing an exit exam affects future high school course-taking patterns and achievement. In a 2014 study by Thomas Ahn, students in North Carolina that barely failed their Algebra I exit exams in ninth grade were five percentage points less likely to take a more rigorous, college preparatory math sequence than students that just passed the exam. These decisions could have longterm implications for students’ postsecondary success, including higher education admissions and placement.23 In other words, the research is not conclusive, but it is fairly consistent: exit exams have tended to add little value in terms of increasing achievement or better preparing most students for life after high school, but have imposed costs on already at-risk students in the form of higher dropout rates and GED attainment—and lower chances for college and career success. THE CASE AGAINST EXIT EXAMS THE COLLISION: HIGHER STANDARDS vs. HIGHER ATTAINMENT Given the less than encouraging results of decades of research on exit exams, why do states continue to administer them? A 2012 survey by the Center on Education Policy (CEP) at George Washington University asked this question of 25 states with exit exams, and the answer was straightforward: rigor matters. Fifteen states maintained their exit exams in order to “attribute greater meaning or significance to a high school diploma,” while 12 also sought to “ensure students who receive a high school diploma are ready for college and career.”24 And state policymakers are not alone in caring about rigor. It is a popular argument as well. In her 2013 New York Times bestseller, The Smartest Kids in the World, Amanda Ripley, a former Emerson Fellow at the New America Foundation, identified high standards and rigor as key distinctions between the United States and high-performing nations like Finland, Poland, and South Korea—all countries that administer exit exams with far higher stakes than any test given in the United States and that see better student results on international achievement tests. Ripley explains: Finland had required a graduation test for 160 years; it was a way to motivate kids and teachers toward a clear, common goal, and it made a high school diploma mean something. Korea rerouted air traffic for their [sic] graduation test. Polish kids studied for their tests on nights and weekends, and they arrived for the exam wearing suits, ties, and dresses. In America, however, there were still many people who believed in a different standard, one that explained a great deal about the country’s enduring mediocrity in education: According to this logic, students who had passed the required classes and come to school the required number of days should receive their diploma, regardless of what they had learned or what would happen to them when they tried to get a job . . . Those kids deserved a chance to fail later, not now. It was a perverse sort of compassion designed for a different century.25 Ripley is right. Sending high school graduates to college or work unprepared is often just sending them to fail somewhere else. And it is the norm in most states across the country. The same 2012 CEP survey found that only eight of the 25 states with exit exams in 2012 gave an exam that was aligned to college- and career-ready standards, and 22 provided alternate paths to general education students who could not meet the testing requirements.26 When it comes to high school course requirements, more states have established college- and careerready expectations. Achieve’s Closing the Expectations Gap 2013 found that 19 states and Washington, D.C. had adopted college- and career-ready high school graduation requirements, but only seven states and Washington, D.C. made these requirements mandatory. In other words, most states with college- and career-ready course requirements also offered less rigorous pathways for students that chose not to take the college- and career-ready curriculum. Further, seven of the 19 states allowed students to opt out of individual courses in the recommended sequence, especially advanced math, but still awarded these students the same diploma as those that had completed the full college- and career-ready curriculum.27 This kind of misalignment between secondary and postsecondary expectations is one reason why states, with federal encouragement, have adopted new, collegeand career-ready standards in reading and math for K–12 schools. The new standards—either the Common Core developed with backing from the National Governors Association (NGA) and the Council of Chief State School 9 10 THE CASE AGAINST EXIT EXAMS Officers (CCSSO), or standards unique to individual states, like the updated Virginia Standards of Learning— are perceived as more rigorous and challenging than previous academic standards. By bridging the gap between high school and higher education, these efforts explicitly aim to ensure students mastering the K–12 standards will be ready to enter credit-bearing courses in college, the military, or job training programs. As teachers shift their instruction to reflect the new standards, states are also developing new standardized assessments in English Language Arts and math to measure student learning against them. Some states are choosing to do so collaboratively via two consortia that are designing Common Core-aligned tests with support from the federal government and philanthropy: the Partnership for Assessment of Readiness for College and Careers (PARCC) and the Smarter Balanced Assessment Consortium (Smarter Balanced). Other states plan to use assessment systems developed by ACT. And still others continue to develop their own tests with assistance from vendors, like Pearson or AIR (see “The Changes: States’ High School Assessment Plans”). Regardless of the assessment chosen, however, most states will administer their new tests next year, although some have already implemented updated assessments. Given the large number of students that currently require remediation in college, as well as the latest results on the National Assessment of Educational Progress (NAEP) and other tests, it seems certain that many students will fail to meet the new standards if cut scores on the new tests are set to accurately reflect postsecondary readiness. Drawing on ten years of research, in 2013, the National Assessment Governing Board produced data on twelfth graders’ academic preparedness for college, based on their NAEP scores, for the first time. Only 39 percent of the nation’s high school seniors were prepared for college-level math, and only 38 percent were prepared in reading.28 Moreover, many students are far from that standard: 35 percent scored below basic in math, and 25 percent did so in reading.29 And among ACT test takers nationally, only a quarter met all four college- and career- ready benchmark scores in English, reading, math, and science in 2013.30 The experiences of early adopters of college- and career-ready tests are also instructive. While changes in proficiency rates from one year to next are crude growth measures, as different student cohorts are included in the year-to-year comparison, Kentucky and New York saw steep, double-digit declines in reading and math proficiency rates when they adopted college- and career-ready tests in 2011 and 2012.31 “ Two key elements of state policy are about to collide as states launch their new standards and assessment systems Thus, two key elements of state policy are about to collide as states launch their new standards and assessment systems. If college- and career-ready standards and tests are simply fitted into states’ existing infrastructure of high-stakes exit exams and graduation requirements, the pipeline of students from high school to college and the workforce could suddenly, catastrophically, clench shut. But if history is any indication, that is unlikely to happen. Instead, the impulse to avoid this outcome would be predictable: the dilution of the college- and career-ready standards and/ or lower cut scores on the new assessments so that more students can pass and graduate. (See sidebar, “Responses to Accountability”.) THE CASE AGAINST EXIT EXAMS Responses to Accountability As standards-based accountability has been implemented widely, schools’ and educators’ responses have sometimes sacrificed rigor to avoid punitive consequences. So-called “gaming the system” has been especially problematic in places where administrators and educators lack the capacity, resources, or skills to adjust their practice to meet more challenging standards. For example, under NCLB, some states tinkered with their school improvement goals or their proficiency cut scores rather than engage in NCLB’s regimen of improvement activities for greater numbers of schools.32 And many blame highstakes school and educator accountability for high profile cheating scandals in Atlanta, Washington, D.C., and other districts.33 In this regard, updating accountability systems over the next several years to adjust to states’ college- and career-ready standards will be especially fraught. Because educational accountability relies, in part, on standardized tests to differentiate between levels of performance and quality, states’ new college- and career-ready assessments will play a critical role in how students, educators, and schools are held accountable for meeting the more rigorous expectations. And by all accounts, most stakeholders expect these tests to be much harder and demanding. As the new tests are implemented, school and educator accountability systems aren’t the only kinds of accountability susceptible to gaming or tinkering. In the past, high-stakes accountability for students has also put pressure on districts and states to reconsider the rigor of assessments or graduation requirements, or find easier pathways for those that can’t meet them. For example, Los Angeles Unified School District adopted a plan in 2005 to require all students to pass the state’s college prep curriculum, known as A-G requirements, Neither outcome is ideal. No state wants to deny large numbers of students the opportunity to graduate from high school, especially when these students have not even been taught based on the new standards for the majority of their time in the K–12 system. Moreover, there are legal implications if states choose to withhold diplomas based on new college- and careerready assessments: students must first be provided with adequate notice of the test requirement, and a fair opportunity to learn the material which they are required to master, which would likely be more difficult to demonstrate with standards that have only been in place for a few years.37 But if states weaken the intent of college- and career-ready standards or the assessments in order to ensure students can graduate from high school, it will counteract their efforts to increase rigor and student achievement, build stronger curricula, starting with the class of 2016. But as the deadline crept nearer, the district adjusted the new rules during initial implementation, allowing students—temporarily—to pass their A-G classes with a ‘D’ despite the fact that the state university system sets a ‘C’ as the passing mark.34 More recently, Texas eliminated its longstanding requirements for high school students to pass Algebra II to graduate, even as the state is simultaneously implementing college- and career-ready standards.35 And this spring, New York lowered the cut scores on its college- and career-ready version of the Regents exams, worried about the effects of lower passing rates. The new scores are so low that students can answer two-thirds of the Algebra I questions incorrectly, and still earn a passing mark.36 While these kinds of changes are often reasonable responses to ensure students are not subjected to unrealistic requirements or denied higher education opportunities, they do illustrate the tensions inherent in high-stakes, student accountability policies. As more and more states embed college- and careerready standards and assessments within their accountability measures, policymakers must be especially careful to design these systems in ways that avoid, to the greatest extent possible, detrimental responses from educators and local officials. These designs could include a gradual phase-in of accountability, or the creation of two different performance standards—one for graduation and another for college and career readiness. States, the testing consortia, and the federal government should also establish mechanisms to monitor the rigor of the standards and assessments as implemented, and should consider the kinds of training, resources, and skills local educators need to adjust their practice based on the information and data that accountability systems produce. authentically evaluate students’ postsecondary readiness, create buy-in from higher education institutions, and use the assessments as one way to place students in collegelevel coursework. The danger of this collision happening is real: at the time of the 2012 CEP survey, over 70 percent of Common Core-adopting states with exit exams planned to replace them wholesale with a consortia-designed assessment in English Language Arts and math. Given the obvious tensions between holding students accountable for higher standards and promoting higher educational attainment, the key question for parents, educators, policymakers, business and civic leaders, and advocates remains: how can we best increase the rigor of a high school diploma and the number of students obtaining one simultaneously? 11 12 THE CASE AGAINST EXIT EXAMS THE CHALLENGE: HOLDING STUDENTS ACCOUNTABLE FOR COLLEGE AND CAREER READINESS From 1975 well into the new millennium, national high school graduation rates were essentially stagnant, fluctuating between 71 and 75 percent.38 But in 2009, America’s high schools broke through the 75 percent barrier, and by 2012, the nation’s average four-year high school graduation rate reached 80 percent for the first time, despite any negative association between exit exams and high school graduation rates.39 While there are still sizable graduation rate gaps between minority and white students, students of color have made some of the biggest gains in the last decade. In its annual Grad Nation report, America’s Promise Alliance cited better data and school accountability as two possible reasons for the recent progress.40 But as states develop school, educator, and student accountability systems aligned to their new standards, getting students to graduate is no longer the only goal. The goal is to graduate them college- and careerready. Balancing these two goals will be challenging, particularly as the standards and assessments are first implemented and students will not have experienced instruction aligned to these expectations for most of their time in the K–12 system. Too much emphasis on college and career readiness within accountability systems, and schools could be rewarded for neglecting, or even pushing out, students that are far below that standard. But with too great an emphasis on graduation rates, efforts to improve students’ postsecondary preparedness and implement the new standards could be ignored or undermined. As states transition to their new standards and assessments, the following sections analyze when and how they will administer college- and career-ready assessments in English Language Arts and math for high schools and how these assessments will be used for high-stakes accountability decisions for students. Given the disappointing research on the effectiveness of high school exit exams and past state and district responses to greater accountability, we are particularly concerned for states that plan to continue their exit exam policies and potentially use a college- and career-ready score as the passing mark. There are simply too few proven benefits for individual students as a result of exit exams, and some of the potential benefits of the new standards and assessments, including using them as one measure of student readiness for college-level courses, could be undermined if these policies continue. This is because it is very likely that states will manipulate or alter the college- and career-ready benchmark to allow more students to graduate if that benchmark is incorporated into exit exam policies. Fortunately, many states are avoiding this result—either by eliminating the exit exam requirement, phasing in the higher passing scores slowly, or setting two distinct scores, one for graduation and another for college and career readiness. “ Getting students to graduate is no longer the only goal: they must also be college- and career-ready THE CASE AGAINST EXIT EXAMS THE CHANGES: STATES’ HIGH SCHOOL ASSESSMENT PLANS Profiling States’ High School Assessment Choices Today, there are 44 states (including Washington, D.C.) that are implementing the Common Core standards, and 34 states committed to either PARCC or Smarter Balanced as governing members. But in the past year, states’ assessment choices have become increasingly complicated and controversial. (See sidebar, “Just How Common is the Common Core?”) And the high school assessment picture is even more so. For example, six states that are, ostensibly, governing members of PARCC or Smarter Balanced are not fully committed to using these assessments in high schools, even as they plan implement them in grades 3–8. Another seven states in the consortia face so much political uncertainty over their assessment policies that it is difficult to predict which tests will be administered to high schools, or to all schools, in the coming years. Taking this information on states’ policies and politics into account, we categorize states into five different college- and career-ready assessment profiles: the Honor Roll, the Exchange Students, the Loners, the Varsity Athletes, and the Drama Club (Figure 6). Just How Common is the Common Core? Five years ago, nearly every state used a different standardized assessment in its K–12 schools. Of course they did—states did not share academic standards, so each needed to develop a test that would be aligned to its particular expectations for students. That began to change in 2010, when the final draft of the Common Core State Standards was released by NGA and CCSSO on behalf of 48 states and Washington, D.C., the culmination of years of work by governors, state chiefs, nonprofits, educators, and national and state content experts.41 That year, the U.S. Department of Education also began accepting state applications for its first Race to the Top competition as part of the federal stimulus program, where states that committed to certain education reforms, including common college- and career-ready standards and assessments, could win a share of over $4 billion—an unprecedented figure for a federal competitive grant at a time when state education budgets sorely needed the funding.42 In 2010, Race to the Top funds were also awarded to help form two consortia of states, PARCC and Smarter Balanced, to develop shared assessments aligned to the Common Core. States did not need to adopt Common Core or participate in either consortium to apply for Race to the Top, but they were awarded extra points in their applications for doing so, and all the eventual state winners did both. While state surveys have shown that the rigor of the standards was the predominant reason for adopting the Common Core, Race to the Top funding was also a factor.43 In total, by the fall of 2011—just 18 months after the first Race to the Top awards—45 states and Washington, D.C. had adopted the common standards and were participating in at least one of the groups developing common assessments, although the number of governing states in the testing consortia was lower: 20 states in Smarter Balanced and 15 in PARCC. Governing states commit to not only participating in test development, but also to piloting the new tests in 2013–14 and fully implementing them in 2014–15. Over time, the ranks of governing states grew as more and more states made their Common Core assessment choices. From March 2012 to July 2013, there were 41 governing states in either PARCC or Smarter Balanced, and nearly all the others were at least considering using one of the new shared college- and career-ready assessments.44 This level of interstate coordination in the name of common academic standards and assessments 13 14 THE CASE AGAINST EXIT EXAMS was astounding and unparalleled. Even by a more conservative count inclusive only of governing members, in less than three years, states committed to a plan where the U.S. education system would transition from using 51 different K–12 academic standards to as few as six, and from using 48 different statewide testing systems to as few as 12. This level of cohesion, however, would not last. Over the last year, the robust coalition of states supporting the Common Core and the assessment consortia has started to fray. Now, concerns about the Common Core and the consortia have grown to dominate state education policy and politics, especially as the 2014–15 school year—the first year of full implementation for the tests—approaches. The criticisms run the gamut from the specific content of the standards (No calculus? Too much emphasis on informational texts?), to higher costs and technological requirements for the online testing systems (Do schools have sufficient bandwidth? Can we afford to spend $10 more/student?), and even outright conspiracy theories (Left-wing indoctrination? A Muslim Brotherhood plot?).45 Despite the outrageous nature of some of these critiques, the pushback against common standards and tests has gained momentum. Since January 2013, 12 states, including those like Florida that were once among the strongest proponents of common assessments, have instead chosen to continue using tests particular to their individual states.46 And in the spring of 2014, in the midst of implementation, Indiana and Oklahoma became the first states to un-adopt the Common Core standards. Further, recently enacted legislation in South Carolina requires the state to adopt new academic standards before the 2015–16 school year, with similar bills under consideration in Missouri and North Carolina.47 Common Standards and Assessments Over Time From 51 sets of state academic standards and 48 different state tests before the Common Core officially launched, to nine sets of standards and 19 different state tests in development today, states have generally converged around the idea of shared academic standards and assessments over the last four years. However, common assessments have become less appealing than the common standards as the 2014–15 school year approaches, with 10 states choosing to exit the consortia and build their own assessment system in the last 12 months. Further, as of June 2014, three states have left the common standards effort as well: Indiana, Oklahoma, and South Carolina. Number of State Academic Standards 50 40 30 20 10 0 2010 2011 2012 Number of State Assessments (governing and advisory states included in PARCC/Smarter Balanced) 2013 2014 Number of State Assessments (only governing states included in PARCC/Smarter Balanced) 50 40 30 20 10 0 2010 Source: New America analysis. 2011 2012 2013 2014 THE CASE AGAINST EXIT EXAMS Figure 6. Who’s Who: A Profile of State Assessment Choices The Honor Roll States that have adopted Common Core and plan to use PARCC or Smarter Balanced for grades 3–8 and high schools.1 PARCC: Arkansas, Colorado, Illinois, Maryland, Mississippi, New Jersey, New Mexico, Ohio, Rhode Island, Washington, D.C. 23 Smarter Balanced: Connecticut, Delaware, Hawaii, Idaho, Maine, Montana, New Hampshire, North Dakota, Oregon, South Dakota, Vermont, Washington, West Virginia The Exchange Students States that were planning to give common tests, but ultimately, went back to their own countries.2 10 Alabama, Florida, Georgia, Indiana, Kansas, Kentucky, Oklahoma, Pennsylvania, South Carolina, Utah The Varsity Athletes States that are part of consortia for grades 3–8 but may make an exception and give special treatment (like using other tests) to high schools.3 6 California, Massachusetts, Missouri, Nevada, New York, Wisconsin The Loners States that stayed to themselves and avoided the common standards and tests altogether.4 5 Alaska, Minnesota, Nebraska, Texas, Virginia The Drama Club States that may be involved with consortia but face legislative and executive conflicts around the new tests, for all grades or for high schools only.5 7 Arizona, Iowa, Louisiana, Michigan, North Carolina, Tennessee, Wyoming 1. Rhode Island will wait to use PARCC in its high schools for an extra year (in 2016). 2. Alabama, Kentucky, and Wisconsin plan to administer tests developed by ACT as their statewide assessment in high schools. Alabama is using the ACT Aspire system for grades 3–8 and high school, while Kentucky and Wisconsin only use ACT-developed exams in high schools (this includes ACT-developed end-ofcourse exams in Kentucky and the ACT Aspire system in Wisconsin). 3. All of these states are governing members of one of the consortia. California will be administering Smarter Balanced in high schools, but has not taken steps to also eliminate the California High School Exit Exam. Some of these states (e.g., Massachusetts, New York) could eventually use Smarter Balanced or PARCC assessments within their high school assessment systems, but it is unclear when or if this will occur.4. Alaska was briefly an advisory member of Smarter Balanced; Minnesota has adopted the Common Core in English Language Arts only. 5. All of these states, except Arizona and Tennessee, are governing members of one of the consortia. Arizona and Tennessee will both issue competitive bids to vendors, which could include the consortia, for their college- and career-ready assessments. Louisiana Governor Bobby Jindal has announced he plans to withdraw the state from PARCC, but Louisiana’s Superintendent of Education, John White, is a supporter of PARCC and does not want the state to leave the consortium. However, White has acknowledged that there will be at least a two-year delay of the PARCC tests for high schools and is receptive to selecting a different high school assessment. The North Carolina Board of Education has postponed adopting any new assessment until 2016–17, and there is legislation pending to reconsider the state’s standards altogether. Iowa, Michigan, and Wyoming also face legislative and statutory barriers in implementing new college- and career-ready assessments. For example, Michigan Governor Rick Snyder recently signed a budget that prevents Smarter Balanced from being implemented in the 2014–15 school year, although the state may be able to purchase some test items from the consortia to use in its state-developed assessment. 15 16 THE CASE AGAINST EXIT EXAMS The stakes attached to high school tests for students, like those applied by exit exams, may be one reason for the additional hesitance and hand-wringing over high school testing. While the tests administered in grades 3–8 today usually matter for school accountability and, in a few places, educator accountability, they are not often linked to consequences for individual students in the elementary and middle grades. In other words, there may be greater anxiety about the consequences of switching tests in high schools, particularly when any costs (or benefits) directly affect a student’s ability to progress from high school to college. A state knows how many students typically pass its state exit exam, but it has no idea how they will fare on the new Smarter Balanced or PARCC tests. And the majority of the 13 states that fit either the Drama Club or Varsity Athletes profile currently administer assessments that students must pass to graduate from high school, or that count for a portion of students’ final course grades in certain subjects. “ But this hesitance to administer the new college- and career-ready tests in high schools is, in some ways, counterintuitive given that the high school-level PARCC and Smarter Balanced tests are the ones that should give states the best gauge of whether students have actually met the new, more rigorous college- and career-ready standards in English Language Arts and math. What kind of message does it send when states adopt college- and career-ready standards, but do not use the accompanying assessments as the final measure of whether students are actually prepared? Yet when policies—like exit exams—potentially pit readiness for higher education against access to it, it is little wonder that some states are especially cautious about using a new test, with different content and different features, if it could affect high school graduation rates or limit students’ postsecondary opportunities. When policies - like exit exams pit readiness for higher education against access to it, it is little wonder that some states are especially cautious THE CASE AGAINST EXIT EXAMS Figure 7. Testing Colleges and Career Readiness in High School: A Mix of Timelines and Developers The Honor Roll States that have adopted Common Core and plan to use PARCC or Smarter Balanced for grades 3–8 and high schools.1 2015: Arkansas, Colorado, Illinois, Maryland, Mississippi, New Jersey, New Mexico, Ohio, R. Island, Washington, D.C. (all PARCC), Connecticut, Delaware, Hawaii, Idaho, Maine, Montana, New Hampshire, North Dakota, Oregon, South Dakota, Vermont, Washington, West Virginia (all Smarter Balanced) 2016: Rhode Island (PARCC) The Exchange Students States that were planning to give common tests, but ultimately, went back to their own countries.2 2012: Kentucky (ACT) 2013: Pennsylvania (Data Recognition Corp.) 2014: Alabama (ACT), Utah (AIR) 2015: Florida (AIR), Georgia (CTB/McGraw-Hill), Kansas (KU CETE), Oklahoma (vendor unknown), South Carolina (vendor unknown) 2016: Indiana (vendor unknown) The Varsity Athletes States that are part of consortia for grades 3–8 but may make an exception and give special treatment (like using other tests) to high schools.3 2014: New York (updated Regents exams) 2015: California (Smarter Balanced+ETS), Missouri (CTB/McGraw-Hill), Nevada (vendor unknown), Wisconsin (ACT) ?: Massachusetts (PARCC), New York (PARCC) The Loners States that stayed to themselves and avoided the common standards and tests altogether.4 2010: Nebraska (Data Recognition Corp.) 2012: Virginia (Pearson) 2014: Texas (Pearson) 2015: Alaska (KU CETE), Minnesota (Pearson) The Drama Club States that may be involved with consortia but face legislative and executive conflicts around the new tests, for all grades or for high schools only.5 2015: Arizona (vendor unknown) 2016: Tennessee (vendor unknown) ?: Louisiana (PARCC), Iowa, Michigan, North Carolina, Wyoming (all Smarter Balanced) Notes: These timelines may be particular to the high school assessment only. For example, New York updated its 3–8 exams before the 2013–14 school year, and Massachusetts plans to decide if it will implement the PARCC exam in grades 3–8 after the 2014–15 school year, but will take longer to make a decision for its high school tests. 17 18 THE CASE AGAINST EXIT EXAMS Transition TImelines and Diverse Developers States’ high school assessment implementation timelines also vary—even within states of similar profiles, as practical and political concerns have come to light (Figure 7). Thirty-four states are adopting new high school tests in the spring of 2015, but eight have already made changes to their high school assessment programs. While the transition timeline is a source of confusion, an even greater one is that many states have not yet selected vendors for a new high school assessment, or their entire K–12 testing system. This is especially problematic in states that, until recently, were governing members of PARCC or Smarter Balanced, including Arizona, Indiana, South Carolina, and Tennessee. Other consortia members have not yet decided to implement a new high school assessment—or state politicians will not let them. Massachusetts and New York are both in the undecided camp. Massachusetts’ school districts can decide whether to use PARCC in 2014–15 or continue with the current testing system, the MCAS. The state board of education will not make a decision on which test to use in grades 3–8 until fall 2015, and will likely wait longer to make a decision for its high schools.48 In New York, PARCC should be implemented in grades 3–8 in 2015–16 (although further delays are not out of the question), but there is no timeline for a similar high school transition, and the state is concurrently updating its long-standing Regents exams to be more aligned with the Common Core.49 Politics are also delaying the transition in Louisiana and North Carolina. In the course of a few months, Governor Bobby Jindal has become one of the most vocal conservative opponents of the Common Core and PARCC, proposing to pull Louisiana from both efforts against the wishes of the state board of education and Superintendent of Education John White.50 White would like to maintain Louisiana’s PARCC transition plan, in large part to provide teachers greater stability. He argues policymakers must give teachers “time to settle in and lead the way” on the Common Core.51 But even if Jindal’s effort is unsuccessful and Louisiana sticks with White’s plan, it is uncertain whether the state will ever use PARCC in its high schools. It will be 2016–17, at least, before Louisiana could use the high school PARCC tests, and White would rather “wait and see how the marketplace resolves itself” before making a decision on the high school exam.52 In North Carolina, the state board of education has already decided to delay its transition to new tests until the 2016–17 school year, with a task force making recommendations in the fall of 2014 about which test to use.53 But now that both chambers of the state legislature have voted to direct the board to develop new standards that could replace the Common Core, the fate of Smarter Balanced in the Tar Heel state is less certain than ever.54 “ Other consortia members have not yet decided to implement a new high school assessment—or state politicians will not let them State legislation is also a potential roadblock for new testing systems in Iowa, Wyoming, and Michigan. Wyoming needs to amend current law to retire its existing testing system, and a task force in Iowa is currently meeting to make recommendations to its state board and legislature on whether any new test should be implemented.55 Meanwhile, Michigan’s legislature has denied funding for Smarter Balanced to the state education agency for 2014–15, despite the fact that the state had suspended development of its own test to plan for the Common Core assessments and has recommended using Smarter Balanced after studying all the possible testing options.56 Now, the state education agency is exploring purchasing some of the Smarter Balanced questions to enhance its existing state test.57 “If we don’t have Smarter Balanced, we won’t have a test,” according to State Superintendent Mike Flanagan.58 End-of-Course or Comprehensive College- and Career-Ready Exams The choices states are facing as they decide whether to give Smarter Balanced or PARCC and when also have implications for the overall design of high school assessment systems. Our analysis shows that the trend toward high school end-of-course exams (EOC), rather than a comprehensive assessment given at a certain grade, will likely continue in the Common Core era, even though the two consortia typify both models (Figure 8).59 Smarter Balanced is planning to deliver a comprehensive exam in grade 11, with the option for states to administer similar exams in grades 9 and 10, while PARCC is developing exams that can more easily adapt to an end-of-course model, including Algebra I, Algebra II, Geometry, English I, English II, and English III. Many states use—and will use—a combination of both kinds of tests, particularly when including exams in other subjects, like science, or college entrance exams the state administers for free to all students, like the ACT or SAT. THE CASE AGAINST EXIT EXAMS Figure 8. End-of-Course Exams vs. Comprehensive Collegeand Career-Ready Exams NOW 25 states administer only comprehensive assessments in high schools, including college entrance exams like the ACT or SAT, where the state administers them to all students. Alaska, Arizona, California, Colorado, Connecticut, Idaho, Illinois, Iowa, Kansas, Maine, Michigan, Minnesota, Montana, Nebraska, Nevada, New Hampshire, North Dakota, Ohio, Oregon, Rhode Island, South Dakota, Vermont, West Virginia, Wisconsin, Wyoming 10 states administer only EOC assessments in high schools. Indiana, Maryland, Mississippi, Missouri, New York, Oklahoma, Pennsylvania, Texas, Utah, Virginia 16 states administer both kinds of assessments, EOC and comprehensive exams, in high schools. Alabama, Arkansas, Delaware, Florida, Georgia, Hawaii, Kentucky, Louisiana, Massachusetts, New Jersey, New Mexico, North Carolina, South Carolina, Tennessee, Washington, Washington, D.C. FUTURE 18 states plan to administer comprehensive high school assessments, and no state is transitioning to this kind of system. Most of these states are members of Smarter Balanced. Alaska, California, Connecticut, Iowa, Kansas, Maine, Michigan, Minnesota, Montana, Nebraska, New Hampshire, North Dakota, Oregon, South Dakota, Vermont, West Virginia, Wisconsin, Wyoming 18 states plan to administer only EOC exams as their high school assessments. These states are PARCC members or will use their own tests. Arkansas, Arizona, Georgia, Indiana, Illinois, Maryland, Massachusetts, Mississippi, New Jersey, New Mexico, New York, Ohio, Oklahoma, Pennsylvania, Texas, Utah, Virginia, Washington, D.C. 15 states plan to administer both EOC and comprehensive tests. These states represent both consortia and non-consortia members. Alabama, Colorado, Delaware, Florida, Hawaii, Idaho, Kentucky, Louisiana, Missouri, Nevada, North Carolina, Rhode Island, South Carolina, Tennessee, Washington Notes: Italicized states are those that plan to change the format of their future high school assessments. Assessments surveyed include all subjects, and assessments must be required of all of the state’s high school students. Of the seven states whose assessment plans are particularly uncertain (see Figure 7), the format of their high school exams will likely not change regardless of their choice of assessment, with the exception of Massachusetts. If Massachusetts does not implement PARCC in high schools, the state will continue to use both EOC and comprehensive exams. Source: New America analysis. States switching from comprehensive to EOC exams may face the most significant changes, especially because students are often tested in only one grade in a comprehensive system, whereas EOC tests are administered whenever a student takes the course. Illinois and Rhode Island, for example, only test high school juniors with their current exams, but will be using course-specific exams in more grades as they transition to PARCC. The same is true in Ohio, which currently has a tenth-grade assessment. Missouri is the only state moving away from an EOC-only system as it starts to also offer the ACT to all students. During these transitions, students will experience both systems—taking some, but not all, of the new EOC tests as juniors or seniors, or taking the old comprehensive exam in the tenth grade and a new EOC the following year. States will need to carefully explain these changes to students and their families, particularly if the tests are used as part of course or graduation requirements, or if schools are planning to start offering college entrance exams free of charge for the first time. The design of states’ high school assessments also affects how they can be used for student accountability—a particularly important consideration for states that are weighing the potential trade-offs in moving toward more rigorous standards while maintaining their exit exam policies. A series of EOC exams is often more versatile than a comprehensive exam. For example, students could be required to pass a certain percentage of EOC tests, to earn a certain cumulative score across the entire series, or to pass only the exams in lower-level courses (like Algebra I, but not Algebra II). Further, EOC exams can be used as part of final course grades, rather than as an exit exam requirement. A comprehensive high school assessment does not lend itself as well to these kinds of variations in exit exam design. 19 20 THE CASE AGAINST EXIT EXAMS College-and-Career Ready Benchmarks Obviously, many states are planning significant changes to their high school assessments in the coming years, but the most important question may not be whether a state chooses PARCC, Smarter Balanced, or the ACT, but whether these assessments are—or will be— benchmarked to a college- and career-ready performance level. In other words, will these tests actually measure whether students are prepared for college and the workforce? And if they do, using these assessments as exit exams may create a dilemma for state officials: how to hold students accountable for higher standards without erecting additional barriers to opportunity or limiting access to higher education. Much as the College Board and ACT have conducted validity studies to establish a particular score on their tests that is associated with a high probability of success in higher education, both PARCC and Smarter Balanced have determined that a certain performance level on their assessments will indicate a student is likely to succeed in entry-level, credit-bearing college courses. While the particular scale score at these levels has not been determined, PARCC has set its college- and careerready distinction at level four out of five, and Smarter Balanced at level four out of four (although students at level three on Smarter Balanced may subsequently demonstrate their readiness in twelfth grade). “ The most important question is whether high school assessments will be benchmarked to a college- and career-ready performance level. Both consortia plan to establish cut scores for each level by the summer of 2015, at the latest, and reaching a consensus within the consortia will likely be a delicate process.60 The future stakes on these scores are high, and states are accustomed to making these decisions independently. The cut scores may influence not only which states are perceived as “better” within the consortia, but also which schools are low-performing, which educators are most effective, and of course, which students are ready for college. And for states with exit exams policies there is added pressure to get the score “right,” since the exams determine both who is ready for college, and also who can access it by earning a high school diploma. For these reasons, many states are considering how to decouple the two, setting a different score for graduation requirements than for a college- and career-readiness determination. Because PARCC and Smarter Balanced have only set performance level descriptors, and not actual scores associated with each level, the vast majority of states have not yet incorporated scores on these exams into their official policies, preferring a wait-and-see approach. One exception is Colorado, which has set a score of four on the PARCC exam as one way students will be able to meet new competency-based graduation requirements. Colorado has also identified the necessary scores on other exams, including ACT, SAT, Advanced Placement, International Baccalaureate, and the Armed Services Vocational Aptitude Battery (ASVAB), and will be approving other locally designed options for students to demonstrate mastery of the college- and career-ready competencies.61 These policies have not yet gone into effect, but as they are implemented, state officials will be watching closely to ensure there are a number of options for students that cannot meet the “college- and careerready” level on PARCC, especially vulnerable student groups like English language learners and students with disabilities. More states currently use college- and career-ready benchmarks established by the College Board or ACT in various policies, even if these tests are not administered to all of the state’s students. For example, 17 states use SAT, ACT, Advanced Placement, and/or International Baccalaureate test scores in some way in their school accountability systems.62 Additionally, a few states have created college- and career-ready benchmarks on state assessments. California students can opt to take the Early Assessment Program (EAP) test, developed with the California State University (CSU) system, to measure college readiness in English and math. EAP results are then used by all CSUs and many California community colleges to exempt students from college placement tests, helping them avoid unnecessary remediation.63 The EAP, however, will likely be discontinued with the advent of Smarter Balanced in California.64 And since 2012, Kentucky, which uses the ACT as part of its high school assessment system, has had common indicators of college readiness for public institutions of higher education across the state.65 As cut scores are established on the PARCC and Smarter Balanced exams, more states may begin to adopt similar policies to bring their secondary and postsecondary systems into greater alignment.66 THE CASE AGAINST EXIT EXAMS THE CHANGES: STATES’ STUDENT ACCOUNTABILITY POLICIES As states are making dramatic changes to their high school assessments, and facing increasing levels of uncertainty over what those assessments will be and when they will be administered, it would be easy for policymakers to ignore questions about how their tests of college and career readiness will be used, or push these decisions off to a later time. After all, it seems counterintuitive to determine how an assessment should be used before selecting that assessment, administering or piloting it, and examining the results. But these policy choices will be essential to the future success of whatever college- and careerready assessment a state selects—and to successful implementation of college- and career-ready standards overall. State policy is likely to have an outsized influence on how the new tests are received by teachers, students, parents, and the public. What are the stakes involved, and for whom? Are there any punitive consequences for poor performance, and how and when would they be applied? These decisions, and how they are explained to stakeholders, will be critical in shaping the environment in which the new tests will eventually operate. In this way, it is precisely the states that have made the most changes, or that face the greatest uncertainty, that should now pay greatest attention to their policies around the use of assessments. Even if legislators or policymakers cannot definitively say what test will be given in 2015 or 2016, they could help stabilize and clarify the transition for educators, students, and families by addressing key policies that will apply regardless of which test is selected. But high school exit exam policies will not just affect individual students. Because they operate at the crucial transition between high school and higher education, they also affect states’ larger efforts to increase college and career readiness through the adoption of new standards, like the Common Core. On their own, the college- and career-ready assessments only aim to determine who is ready for college. But when used as an exit exam, they could now also determine who is able to go to college by earning a high school diploma. States that continue to use exit exams as they shift to collegeand career-ready standards will have to consider the trade-offs between these two uses of assessments. While both higher standards and higher educational attainment are important policy goals, they could work against one another if states’ new college- and career-ready tests are used as exit exams without careful thought and a deliberate transition strategy. Past, Present, and Future: State High School Exit Exam Policies In 2012, the Center on Education Policy (CEP) found that 25 states required students to pass an exit exam to graduate high school, and in future years, two states planned to phase out their policies, while one state planned to add an exit exam. Further, two states had recently dropped their exit exam policies for the 2011–12 school year, North Carolina and Tennessee. When CEP asked specifically about how the transition to the Common Core affected their exit exams, 14 of the 22 responding states indicated that they planned to maintain an exit exam requirement, and just six were unsure. However, our analysis reveals the extent to which states’ policies have—and have not—changed in the last two years. Today, many states still have their exit exams in place, but the number of states that could potentially continue these policies moving forward is higher than the 14 reported in 2011–12. Based on a scan of current state policy and assessments in all subjects, 24 states had an exit exam requirement in place for the class of 2014, and as many as 21 states could have an exit exam as they transition to college- and career-ready standards and assessments.67 In these states, officials will be confronted with the dilemma of how to hold students accountable for higher academic standards while making higher education an option for as many of them as possible. Since the CEP survey, only Arkansas has eliminated its exit exam. Moving forward, Alabama, Alaska, Arizona, Georgia, Minnesota, and South Carolina will also eliminate their exit exams, while Connecticut, Pennsylvania, and Rhode Island may adopt an exit exam policy (see Figure 9). 21 22 THE CASE AGAINST EXIT EXAMS Figure 9. Past, Present, and Future: State High School Exit Exams THEN (Class of 2012) 25 states have an exit exam requirement Alabama, Alaska, Arkansas, Arizona, California, Florida, Georgia, Idaho, Indiana, Louisiana, Maryland, Massachusetts , Minnesota, Mississippi, Nevada, New Jersey, New Mexico, New York, Ohio, Oklahoma, Oregon, South Carolina, Texas, Virginia, Washington 14 states plan to maintain an exit exam requirement after adopting Common Core Arizona, Arkansas, California, Florida, Idaho, Louisiana, Massachusetts, Mississippi, Nevada, New Mexico, New York, Oklahoma, Oregon 2 states plan to eliminate an exit exam requirement Alabama, Georgia NOW (Class of 2014) 24 states have an exit exam requirement Alabama, Alaska, Arizona, California, Florida, Georgia, Idaho, Indiana, Louisiana, Maryland, Massachusetts, Minnesota, Mississippi, Nevada, New Jersey, New Mexico, New York, Ohio, Oklahoma, Oregon, South Carolina, Texas, Virginia, Washington 9 states use high school tests as a portion of students’ final course grades Florida, Kentucky, Louisiana, Missouri, North Carolina, Pennsylvania, South Carolina, Tennessee, Texas FUTURE As many as 21 states could have an exit exam requirement California, Connecticut, Florida, Idaho, Indiana, Louisiana, Maryland, Massachusetts, Mississippi, Nevada, New Jersey, New Mexico, New York, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, Texas, Virginia, Washington At least 11 states plan to use high school tests as a portion of students’ final course grades Alabama, Arizona, Florida, Georgia, Kentucky, Louisiana, Missouri, North Carolina, Pennsylvania, South Carolina, Tennessee 7 states will have eliminated an exit exam requirement Alabama, Alaska, Arizona, Arkansas, Georgia, Minnesota, South Carolina 6 states undecided on a future exit exam requirement after adopting Common Core Indiana, Maryland, New Jersey, Ohio, South Carolina, Washington Notes: Italicized states are those that plan to eliminate their policy in the future, while states in bold are those that plan to adopt the policy for the first time in the future Georgia is eliminating all of its exit exams, with the exception of its comprehensive writing test. Source: 2011–12 data are from the Center on Education Policy’s 2012 state survey “State High School Exit Exams: A Policy in Transition.” Questions about states’ exit exam plans related to Common Core did not include states that had not adopted the standards (e.g., Texas). Newer data are from a New America analysis, which does not include state surveys but is inclusive of all 50 states, plus Washington, D.C. THE CASE AGAINST EXIT EXAMS The shift to college- and career-ready standards is one reason states are reconsidering their exit exam policies. For example, South Carolina is removing its exit exam requirement, and instead will require students to take, but not earn a certain score on, the ACT WorkKeys and a college readiness test. As Melanie Barton, director of the state’s Education Oversight Committee, put it, “the [exit exam] doesn’t give students any information to move forward. It’s a bare-minimum criteria to get a diploma . . . The bar has been raised. The diploma is no longer enough.”68 The assessment policy changes have also been welcomed by local educators across the state as a way to help and encourage all students to succeed after high school. As one district superintendent explained, it “means that we have to look at all options to help students be successful and not look at selecting and sorting students into being unsuccessful.”69 Unlike requiring students to earn a particular test score to graduate, some states use standardized tests as final exams and incorporate performance on them into course grades. Nine states currently use high school assessments toward students’ final grades, and at least 11 states plan to do so in the future. This is one alternative to high-stakes exit exams, and it works particularly well in states that administer end-of-course exams, as opposed to comprehensive ones that can cover content taught in several grades or subjects. When EOC tests are used for course grades, rather than as an exit exam, the assessments can still be used to inform students, families, higher education, and employers of an individual student’s postsecondary readiness, and there is still an incentive for students to work hard and perform well on the test, since there are consequences for poor performance. But rather than deny students a diploma, this poor performance is reflected in their grade point averages. Where High School Exit Exams and the Consortia Collide States’ exit exam policies are evolving, however, in tandem with their efforts to adopt college- and careerready assessments. Will states use consortia-designed tests to both determine students’ preparedness for higher education and their ability to access it by earning a high school diploma? If so, efforts to increase high school graduation rates and college and career readiness could collide with, rather than complement, one another. If a particular cut score is tied to graduation, most states, at least initially, will try to designate a score that the vast majority of students can meet—which will likely be below the college- and career-ready determination, or even in-between the performance level descriptors set by the consortia. This least-disruptive strategy is preferable to requiring students to suddenly meet college- and career-ready benchmarks to graduate, but it does make it even more critical to communicate clearly with students. If one score is used for graduation requirements, while different scores are used for college placement, school accountability, or other purposes, achieving clarity about what signifies college and career readiness will require more nuance and explanation. High school assessments have not always been used for these purposes. In 2004, only Georgia indicated to CEP that it used exit exams to ensure students were prepared for college and careers, but twelve states did so in 2012: Florida, Georgia, Idaho, Indiana, Massachusetts, New Mexico, New York, Oklahoma, Oregon, Rhode Island, Virginia, and Washington. This shift is likely due, at least in part, to the increased urgency and attention toward improving students’ preparation for college and the workforce. Along these lines, 18 states in the 2012 survey reported that they planned to replace their current exit exam with a new assessment aligned to the Common Core standards. Only California planned to continue administering its current exit exam rather than update it to align with the new standards.70 “ States’ exit exam policies are evolving in tandem with their efforts to adopt collegeand career-ready assessments Even though the number of states with exit exams has not changed dramatically since 2012, our analysis shows that the tests states are using—and plan to use—in English Language Arts and math for these purposes have changed (Figure 10). This reflects both greater certainty on the part of some states, who finalized their testing decisions over the last two years, as well as greater uncertainty in other states over the use of the Common Core and/or the consortia-developed tests. Half of the states that were planning to use PARCC or Smarter Balanced as their exit exam in 2012 have reversed those plans, or are uncommitted to transitioning to the consortia tests in the future. This includes seven of the 13 states that were previously planning to use the PARCC tests, and one of the three that were planning to use Smarter Balanced as exit exams. Two states (South Carolina and Washington) that were unsure of their plans in 2012 have now solidified their exit exam policies, but they have been replaced by more states on the fence about how they will use tests for graduation or course grades. 23 24 THE CASE AGAINST EXIT EXAMS Figure 10. States’ Changing Plans to Use PARCC or Smarter Balanced as Exit Exams THEN (2011–12) 16 states planned to replace their exit exam with a consortia-developed test (PARCC or Smarter Balanced) Arizona, Arkansas, Florida, Idaho, Indiana, Louisiana, Maryland, Massachusetts, Mississippi, Nevada, New Jersey, New Mexico, Ohio, Oklahoma, Oregon, Rhode Island 3 states planned to use their existing exit exam or replace it with a new state-developed test Alabama, California, New York NOW (2013–14) 10 states plan to replace their exit exam with a consortia-developed test (PARCC or Smarter Balanced) Connecticut, Idaho, Maryland, Mississippi, New Jersey, New Mexico, Ohio, Oregon, Rhode Island, Washington 8 states plan to use their existing exit exam or replace it with a new state test California, Florida, Indiana, Nevada, Oklahoma, Pennsylvania, Texas, Virginia 6 states plan to use their existing exam or a new state test toward students’ final course grades Alabama, Florida, Georgia, Kentucky, Missouri, Pennsylvania 2 states were uncertain about what their future exit exam will be South Carolina, Washington 3 states are uncertain about what their future exit exam will be Louisiana, Massachusetts, New York 5 states are uncertain about what their future endof-course exam toward students’ final grades will be Arizona, Louisiana, North Carolina, South Carolina, Tennessee Notes: Italicized states are those that changed their plans since the CEP 2012 survey. Source: 2011–12 data are from the Center on Education Policy’s 2012 state survey “State High School Exit Exams: A Policy in Transition.” Questions about states’ exit exam plans related to Common Core tests did not include states that had not adopted the standards (e.g., Texas). Newer data are from a New America analysis, which does not include state surveys but is inclusive of all 50 states, plus Washington, D.C. THE CASE AGAINST EXIT EXAMS 25 Figure 11. The Overlap Between States’ Assessments Choices and Exit Exam Policy Choices The Honor Roll States that have adopted Common Core and plan to use PARCC or Smarter Balanced for grades 3–8 and high schools.1 43% Plan to use their high school assessment (PARCC or Smarter Balanced) in a high-stakes capacity for students: Connecticut, Idaho, Maryland, Mississippi, New Jersey, New Mexico, Ohio, Oregon, Rhode Island, Washington 57% Do not plan to use their high school assessment (PARCC or Smarter Balanced) in a highstakes capacity for students: Arkansas, Colorado, Delaware, Hawaii, Illinois, Maine, Montana, New Hampshire, North Dakota, South Dakota, Vermont, West Virginia, Washington, D.C. The Exchange Students States that were planning to give common tests, but ultimately, went back to their own countries.2 80% Plan to use their high school assessment in a high-stakes capacity for students: Alabama, Florida, Georgia, Indiana, Kentucky, Oklahoma, Pennsylvania, South Carolina 20% Do not plan to use their high school assessment in a high-stakes capacity for students: Kansas, Utah The Varsity Athletes States that are part of consortia for grades 3–8 but may make an exception and give special treatment (like using other tests) to high schools.3 83% Plan to use their high school assessment in a high-stakes capacity for students: California, Massachusetts, Missouri, Nevada, New York 17% Do not plan to use their high school assessment in a high-stakes capacity for students: Wisconsin The Loners States that stayed to themselves and avoided the common standards and tests altogether.4 40% Plan to use their high school assessment in a high-stakes capacity for students: Texas, Virginia 60% Do not plan to use their high school assessment in a high-stakes capacity for students: Alaska, Minnesota, Nebraska The Drama Club States that may be involved with consortia but face legislative and executive conflicts around the new tests, for all grades or for high schools only.5 57% Plan to use their high school assessment in a high-stakes capacity for students: Arizona, Louisiana, North Carolina, Tennessee 43% Do not plan to use their high school assessment in a high-stakes capacity for students: Iowa, Michigan, Wyoming 43+57 80+20 83+17 40+60 43+57 43% Have High Stakes Policy 80% Have High Stakes Policy 83% Have High Stakes Policy 40% Have High Stakes Policy 57% Have High Stakes Policy Notes: Italicized states are those that use their high school assessments as a portion of students’ final course grades, as opposed to using the assessment as an exit exam. 26 THE CASE AGAINST EXIT EXAMS Additionally, there appears to be a relationship between states that plan to continue using assessments as an exit exam (or toward a final course grade) and the uncertainty of their assessment choices (Figure 11). Four out of ten PARCC and Smarter Balanced states are considering using those tests in high-stakes ways in English Language Arts and math, but the proportion for states that have abandoned the consortia for their own state-developed assessments is much higher. Eight out of ten states in the “exchange student” and “varsity athlete” profiles deploys their high school tests in a high-stakes way for students. While there are many reasons for states to choose another assessment, on top of the deepening ideological and political fault lines surrounding the Common Core, the stakes involved with these testing decisions for students—and not just schools or educators—only add to the importance of selecting a college- and career-ready test and transitioning to it in a smart, intentional way. And it is hardly surprising that states would be especially cautious about switching their assessments if those tests determine whether students graduate from high school, setting up a conflict between measuring preparedness for college and careers and helping students to get there via earning a high school diploma. State Strategies for Maintaining Student Accountability while Transitioning to College- and Career-Ready Assessments Given the stakes, states have made several kinds of changes to their exit exam policies as they begin to administer college- and career-ready assessments. While the three states that are planning to adopt exit exams in the future (Pennsylvania, Rhode Island, and Connecticut) may not have to grapple with reconciling an exit exam policy across two different assessments, the 24 states with existing exit exams will. College- and careerready exit exams require states to balance the tensions between the higher expectations and higher educational attainment: the more prepared for college and careers students must be to pass the exit exam, the less likely it is that most students will be able to meet that standard and graduate from high school. And much as states are pausing or modifying their accountability systems for teachers and schools by suspending the designation of new low-performing schools or delaying when teacher evaluations will inform personnel decisions, most states are taking a careful, deliberate approach to their exit exams, adopting various timelines and transition plans as the more rigorous expectations take effect (Figure 12).71 How States Plan to Lower the Stakes on High School Tests Over the next three years, 10 states with high-stakes tests for students plan to ease those stakes as full implementation of college- and career-ready assessments begins. Two states plan to adopt end-of-course exams and use the results toward students’ grades as a replacement for their exit exams (Strategy #2: Alabama, Georgia), while four have passed legislation to eliminate the exit exam altogether (Strategy #1: Arkansas, Alaska, Minnesota, South Carolina). And although the state is not participating in the Common Core, Texas has also modified its exit exam policy by reducing the number of required assessments. Originally, Texas students in the Class of 2015 would have had to pass 15 end-of-course exams to graduate, but the state legislature changed this requirement before it had even taken full effect, maintaining just five of the tests (Strategy #3).72 Idaho, Nevada, and New Jersey have taken a slightly different approach. Rather than lowering the stakes for students, they plan to pause them to allow schools to teach the new standards for several years before incorporating them into exit exam policies (Strategy #4). For example, New Jersey students in the class of 2014 still needed to pass the comprehensive High School Proficiency Assessment in reading, writing, and math to graduate, but this was the last year the test was administered—and it will likely be at least 2019 before the PARCC exam is used in a similar way.73 Idaho will have a one-year pause: students in the class of 2016 will have to pass neither the Smarter Balanced exam, nor the existing Idaho Standards Achievement Test, to graduate from high school.74 Nevada is developing its own end-ofcourse exams for high schools, which will be launched in the 2016–17 school year to replace the Nevada High School Proficiency Exam, but will not use the new coursespecific tests for graduation purposes until the class of 2019.75 These pauses allow states to effectively manage the transition to new tests in the short-term, but they still need to indicate the passing score students will need to attain in the long-term—weighing the potential benefits of holding students accountable for college- and careerready standards against the potential costs of preventing them from graduating from high school. Other states are experimenting with competencybased requirements, allowing some or all students to demonstrate that they are proficient on the new standards in more flexible ways (Strategy #5: Colorado, Maine, New Hampshire, Vermont). Here, New Hampshire is the trailblazer, abandoning the use of seat time to measure students’ learning in 2005, so that all high schools were awarding course “credits” based on competencies for the class of 2012. While these systems are still a work in progress, the state has subsequently updated its competencies to reflect college and career readiness and is relying on local school districts to ensure that there will be appropriate assessments in place for high school students to show their mastery, in addition to the state’s Smarter Balanced test.76 Because the consortia tests share some elements with competency-based approaches, like performance tasks, Paul Leather, New Hampshire’s Deputy Commissioner of Education, sees Smarter Balanced as a “bridge” between the old system of standardized, end-of-year tests and a new competency-driven one, and other states seem to agree.77 Colorado, Maine, and Vermont have THE CASE AGAINST EXIT EXAMS Figure 12. How States with High-Stakes Testing Policies are Transitioning to College- and Career-Ready Assessments 7a 7b 7c 7c 5 1 4 6b 6a 5 6a 1 7c 1 4 7a7b 7c 4 3 6a 7a 7b 7c 6a 2 2 6a 6b 6b 7a 7b 55 1 New Policies to Lessen the Stakes on High School Assessments 1 Strategy #1: eliminate high school exit exams 2 Strategy #2: replace exit exam with EOC exams for final grades1 4 Strategy #4: pause the exit exam requirement 5 Strategy #5: move toward competency-based requirements2 3 Strategy #3: require fewer end-ofcourse exams to graduate Transition Strategies to Keep the Stakes on High School Assessments 6a Strategy #6a: use the same exam with the same cut scores3 6b Strategy #6b: use the same exam with higher cut scores4 7b Strategy #7b: use a new exam but only for lower-level courses 7c Strategy #7c: use a new exam but a two-cut-score approach 7a Strategy #7a: use a new exam but offer old exam during transition Other Transition Strategies Uncertain Transition Plans5 States Implementing Exit Exam Policies for the First Time 1. Georgia is eliminating all of its exit exams in favor of end-of-course exams, with the exception of its comprehensive writing test. 2. The four states moving toward competency-based approaches do not currently use their assessments as exit exams, nor do they plan to do so in the future. 3. California and Missouri plan to continue their current graduation testing requirements and have not announced a timeline for retiring their current assessments, although California will also be administering the Smarter Balanced exam. Louisiana, Massachusetts, and North Carolina may retire their current high school graduation tests, but these changes would affect the classes of 2019 or 2020, at the earliest. 4. Virginia established higher cut scores on its Standard of Learning tests for the 2011–12 and 2012–13 school years to better reflect college- and career-ready expectations, and Kentucky is touted as the first state to adopt college- and career-ready tests (produced by the ACT) in 2012. New York will not require graduating students to meet college- and career-ready benchmarks on its Regents exams until the class of 2022, at the earliest. These tests could be updated Regents exams, or the PARCC exams. 5. There are competing accounts as to whether Florida will use its new state tests, developed by AIR for the 2014–15 school year, as an exit exam, and if so, how the state will navigate the transition. Arizona, Nevada, and Tennessee have not articulated high-stakes testing transition policies, likely because they have not selected a vendor for their college- and career-ready end-of-course assessments, although only Nevada will use this test as an exit exam (the others will likely use the tests toward final course grades). Similarly, Indiana may be phasing out its end-of-course exams, a current graduation requirement, and extending its comprehensive testing system into ninth and tenth grades in 2015–16, but these plans are not finalized. Finally, given problems with its existing vendor for end-of-course exams, in addition to its rejection of the Common Core standards in June 2014, Oklahoma’s future high school testing policies are particularly volatile. 27 28 THE CASE AGAINST EXIT EXAMS also articulated plans to develop competency-based systems, including graduation requirements, in the future. Although none of these leading states have exit exams, the competency-based approach could become a prominent alternative to exit exams, as several states with graduation test requirements are considering this kind of system.78 Achieve’s Competency-Based Pathways State Partnership includes states with high-stakes high school assessment plans, like Kentucky, Ohio, Oklahoma, Oregon, and Rhode Island. The Partnership is assisting these states in competency-based approaches to college and career readiness, including graduation requirements.79 Further, the Carnegie Foundation for the Advancement of Teaching has noted that only nine states do not offer districts the option to define the credit hour more flexibly.80 How States Plan to Keep the Stakes on High School Tests While seven states are easing high-stakes accountability permanently for students during the transition to collegeand career-ready tests, twice as many states are planning to maintain their current policies without a pause. They are facing an array of possible choices as they seek to implement rigorous standards and assess students’ preparedness faithfully, without stifling access to higher education or lowering students’ educational attainment (Figure 12). One possible choice, however, is more of a non-choice: five states are maintaining their exit exams by default— remaining noncommittal to new assessments and keeping their current testing regimes for the foreseeable future (Strategy #6a). California, for example, is unwavering in its plans to use the Smarter Balanced tests and even won a special testing waiver from the federal government to fully transition to Smarter Balanced a year earlier than planned.81 But the state continues to use the California High School Exit Exam as a graduation requirement, even though it is not aligned to the Common Core standards being taught in schools. While the state has contemplated using part of the Smarter Balanced test to replace the exit exam and is considering possible alternatives, an official plan has not been announced.82 Louisiana, Massachusetts, Missouri, and North Carolina are also maintaining their current tests, and hence, their high-stakes policies, for the time being. Kentucky, New York, and Virginia are similarly avoiding the consortia tests, except they have already started to use other assessments that are aligned to more rigorous standards (Strategy #6b). Rather than an exit exam, Kentucky adopted end-of-course tests developed by the ACT in 2012 as part of students’ final course grades, and these tests are linked to the ACT’s college-ready benchmarks. Virginia phased in new cut scores on its state-developed tests in 2012 and 2013.83 It is unclear, however, how rigorous unique state college- and careerready determinations, like those in New York and Virginia, will be, as they lack the external pressure from other states to set their scores at a certain level, and are not linked to an established college-ready benchmark used by postsecondary institutions. It is particularly telling that when Virginia first established its new scores, proficiency rates for elementary schools fell more significantly than at the secondary level.84 And just as the new tests were launched, New York lowered the cut scores on its college- and career-ready version of the Regents exams, worried about the effects of lower passing rates. The new scores are so low that students can answer two-thirds of the Algebra I questions incorrectly, and still pass.85 This lower proficiency score may be appropriate for a graduation requirement, especially as the new tests are first introduced, but it should not be confused with a sign of college and career readiness if that is not what the lower cut score is measuring. In some ways, states that have opted to keep their existing tests have an easier path forward than those that are simultaneously implementing a new collegeand career-ready testing system and attempting to hold students accountable for the results. In these states, their potential policy choices could pit the success of the new consortia exams, and their ability to accurately gauge students’ mastery of the Common Core, against the success of students and their future educational opportunities. To avoid this outcome, our analysis finds that six of the 10 states that are planning to use Smarter Balanced or PARCC assessments for high-stakes graduation decisions have created the most nuanced, intricate, and complicated transition plans: Maryland, Mississippi, New Mexico, Ohio, Oregon, and Washington. Unlike Idaho and New Jersey, where the exit exam will be put on hold as the consortia assessments are launched, these states are trying to manage a multi-year process of phasing-in new tests, retiring old ones, and determining graduation requirements and alternative options for each subsequent cohort in a way that reflects both the urgency of college and career readiness for all students and the need for fairness and thoughtful implementation of higher standards. It is no easy task to create a seamless process for students between the old graduation tests and the new ones coming from PARCC and Smarter Balanced. One common approach is to phase-in the new tests over several years (Strategy #7a: Maryland, Mississippi, Ohio, Washington). This overlap allows states to continue using their existing tests for current students, and only require the consortia tests as an exit exam for younger students, often those that are now in middle school. Another tactic—particularly for states that use end-of-course exams—is to use the exams associated with lower-level courses for graduation requirements, but not the exams associated with more advanced subjects (Strategy #7b). For example, Maryland will offer the PARCC exams in English 9, 10, and 11 and in Algebra I, Geometry, and Algebra II, but only the English 10 and Algebra I tests will be used as exit exams. In this way, the state can avoid pitting the new expectations against students’ graduation prospects, with only the most advanced tests used for measuring postsecondary readiness, including whether students require remediation in college.86 Mississippi, Ohio, and Washington are also taking this approach in tandem with a gradual phase-in of the new tests.87 Put another way, these states will offer two graduation tests (the old state-developed test and PARCC or Smarter Balanced) and will split how the suite of consortia tests are used (some for graduation, others for college and career readiness). But there is another element of bifurcation in their plans. Most are also establishing two cut scores—a college- and career-ready score, and a lower score, or composite score, for graduation (Strategy #7c). Maryland and Ohio plan to establish composite scores on the end-of-course exams that are required for graduation (which do not include English 11 or Algebra II), implying THE CASE AGAINST EXIT EXAMS that the graduation score will not be the equivalent of the college- and career-ready benchmark. Washington’s state board has also implied that it will use two different scores. New Mexico is arguably the clearest, explicitly stating that its passing score on the PARCC exam will be a three out of five (although the state has not indicated if this applies to every PARCC exam, or just a sampling).88 And Oregon is prohibited by state law from establishing a cut score on its new Smarter Balanced exit exam that is more rigorous than its current exam, the OAKS, until 2019—and even then, the state may not choose to increase its passing benchmark at all.89 Mississippi, however, has not yet communicated what its new cut score will be.90 A two-cut-score approach is appealing because it reduces the incentive for states to manipulate and/or lower the college- and career-ready benchmark score in order to meet graduation rate accountability targets and maintain the pipeline of students from high school to higher education. And although there are trade-offs to this strategy, they are relatively minor. With multiple scores used for different purposes, it could become confusing for students why one score is good enough to graduate, but not good enough to escape remediation in college—particularly if the K–12 system makes it appear that a high school diploma signifies college and career readiness, regardless of students’ mastery of the standards by the time they graduate. Communicating these differences clearly should be a priority for states that take the two-score approach. Finally, it is worth noting that even within a two-cut-score system, states still face pressure to choose the “right” scores, especially on the lower one used for graduation. This could have implications for the quality or validity of the new tests at the lower-end of the readiness spectrum, even if the higher college- and career-ready score is set at a rigorous level. The Evergreen State: A Model Transition Plan Washington offers a good example of how these three approaches to the new English Language Arts and math assessments come together in a single state plan (Figure 13). The class of 2014 must pass the reading and writing High School Proficiency Exams (HSPE), along with an end-of-course test in Algebra I or Geometry to graduate. But the classes of 2015 and 2016, who will experience both old and new assessment systems, could graduate meeting the old HSPA requirements, or by passing a number of other exams in their place: a tenth grade English Language Arts test aligned to the Common Core, end-of-course exams in Algebra I or Geometry aligned to the Common Core, or the eleventh grade Smarter Balanced tests in both subjects. Because Smarter Balanced is not developing end-of-course tests, Washington will be developing a Common Core-aligned tenth grade English Language Arts test and math endof-course exam especially for the transition—a kind of bridge assessment between the old HSPE and the comprehensive eleventh grade Smarter Balanced tests. However, the state will also continue to offer the HSPE in 2015 and 2016, maintaining continuity for students and giving them numerous opportunities to meet the testing requirements. In other words, the classes of 2015 and 2016 will have three options for graduation tests (old, bridge, or new assessment), and an option to meet the requirements through tests that only cover the content taught in lower-level courses (English 10, Algebra I or Geometry). Figure 13. Washington State High School Testing Transition Assessment Requirements for Certificate of Academic Achievement (CAA) / High School Diploma Subject Classes of 2013 & 2014 Classes of 2015 & 2016 Classes of 2017 & 2018 Class of 2019 10th-grade ELA Exit Exam based on the Common Core  - OR11th-grade Smarter Balanced ELA Test 11th-grade Smarter Balanced ELA Test English Reading and Language Writing HSPEs* Arts Reading and Writing HSPEs* - OR10th-grade ELA Exit Exam based on the Common Core** - OR11th-grade Smarter Balanced ELA Test** Math Algebra I/Integrated Math 1 EOC - ORGeometry/Integrated Math 2 EOC - ORAlgebra 1/Integrated Math 1 EOC Exit Exam based on the Common Core** - ORGeometry/Integrated Math 2 EOC Exit Exam based on the Common Core** - OR11th-grade Smarter Balanced Math Test** Algebra 1/Integrated Math 1 EOC - ORGeometry/Integrated Math 2 EOC 11th-grade Smarter Balanced Math Test * Reading and Writing HSPEs will be available to 11th and 12th graders in spring and summer 2015 and to 12th graders in spring and summer 2016. ** This test is not available until spring 2015. Source: Adapted from the Washington State Office of Superintendent of Public Instruction website: http://www.k12.wa.us/assessment/ StateTesting/ (accessed June 15, 2014). 29 30 THE CASE AGAINST EXIT EXAMS Meanwhile, the classes of 2017 and 2018 can meet their graduation requirements through any of the Common Core-aligned tests, but the HSPE will no longer be an option. And eventually, the tenth-grade and end-of-course bridge assessment options will also be phased out—the class of 2019 will be the first that must meet the passing standard on the Smarter Balanced exam in both subjects to graduate. Although the Evergreen state has not specified the Smarter Balanced performance level (out of four) that is equivalent to “meeting standard,” it appears likely that the benchmark for graduation will be lower than the Smarter Balanced college-ready distinction. The state board of education’s draft recommendations for using the eleventh-grade Smarter Balanced assessment note that the board is weighing the ramifications of a “two-cut-score system (one requirement for graduation, another for demonstration of college readiness).”91 Using a score of three on Smarter Balanced would also provide consistency during the transition, since Washington designates a score of three out of four as the passing score on its current tests.92 Communication and Clarity as a Remedy for Uncertainty Communicating transition plans like Washington’s clearly to the public and making them easily accessible on state education websites is nearly as important as finalizing the plan, especially if states’ choices could affect how and whether students graduate from high school. The Evergreen State does relatively well on both counts, with an easy-to-read rubric for each graduation cohort and a prominent, easy-to-find location on the state education agency’s assessment webpage. But while each of the six states that are keeping their high-stakes testing policies and using the consortia tests could improve the specificity of their plans, they are, arguably, in better shape in terms of communication than those states that plan to keep their high-stakes policies, but have not settled on their high school assessment choices. The Common Core chaos and/or high school testing uncertainty leaves students particularly vulnerable in Arizona, California, Florida, Indiana, Louisiana, North Carolina, Oklahoma, and Tennessee. This is especially true in the five states where the stakes involved are not just final course grades, but possibly high school graduation as well: California, Florida, Indiana, Louisiana, and Oklahoma. It is difficult enough to navigate and meet existing testing requirements without the added insecurity of what those tests will be. Even if these states are unable to choose a new high school assessment in the near-term, they could ease anxiety over the collegeand career-ready transition by clearly articulating policies for how these exams will be used for student promotion, and when. For guidance these states could look to Massachusetts and New York. The Empire State is still uncommitted to giving the PARCC exam in its high schools, but despite this instability, state leaders have firmly established a policy that no student will be held accountable for meeting college- and career-ready standards on whatever assessment is used until the class of 2022. Students, families, and educators could be more assured and better prepared if they also knew what those graduation tests would be, but at least they have a partial understanding of what is expected of students in the coming years and they know that any new requirements will be phased in gradually.93 THE CASE AGAINST EXIT EXAMS THE CASE: MOVING AWAY FROM HIGH SCHOOL EXIT EXAMS Students today cannot afford to be high school dropouts any more than they can afford to enter college unprepared. Luckily, the transition to college- and career-ready standards across the country offers states the opportunity to fully reimagine how they can best ensure students not only graduate from high school, but do so ready to succeed in higher education and in the workforce. The new standards open possibilities for richer instruction, better curricula, and deliberate alignment between secondary and postsecondary learning. And the new assessments will play a critical role in the successful implementation of the new standards. They need to tell public officials whether schools and educators are positively influencing learning and encouraging student growth. They need to tell teachers whether their individual students are making progress and whether their instructional practices are effective. They need to tell families whether their students are on track to college and career readiness. And they need to tell students—especially high school students—whether they are likely to need remediation before starting collegelevel classes. But the new assessments do not need to be exit exams. For starters, the research on the effectiveness of exit exams is murky, at best. These tests have not consistently improved student achievement, high school graduation rates, postsecondary attainment, or workforce outcomes— and have often made vulnerable students even more so. Further, exit exam policies make states’ efforts to introduce college- and career-ready expectations compete against their efforts to ensure more students get those opportunities, instead of complementing those efforts. While the desire to motivate high school students to work hard, to make a high school diploma both a meaningful achievement and a valuable credential to employers and colleges, and to assess college and career readiness or mastery of state standards are all worthwhile goals, each can be accomplished in another way, without relying on exit exams. 1) Students can be motivated to work hard in high schools by using assessments toward final course grades, rather than as graduation requirements. And now that the English Language Arts and math assessments will also measure students’ readiness for postsecondary education, the new tests could be given positive, rather than punitive, stakes.94 Students could be rewarded for scoring at the highest college- and career-ready levels on the assessments in a number of ways, including: a) opportunities to take accelerated coursework like dual enrollment and Advanced Placement, b) access to state merit-based financial aid or scholarship programs for college, and c) automatic placement into credit-bearing courses at in-state public colleges and universities. These sorts of policies give students a “stake” in doing well on the new assessments, without jeopardizing their ability to graduate if they are not suddenly able to meet the higher expectations. And the positive incentives will be most powerful if higher education institutions, in addition to high schools, recognize the college- and career-ready assessments as meaningful measures and integrate common college- and career-ready definitions into their policies, including remediation and course placement. Greater alignment between K–12 and higher education systems should be a top priority for states and for the consortia moving forward. Moreover, these approaches could actually be effective. The National Research Council’s blue ribbon commission, the Committee on Incentives and Test-Based Accountability, found that “several experiments with providing incentives for graduation in the form of rewards, while keeping graduation standards constant, suggest that such incentives might be used to increase high school completion.”95 2) In addition to personal incentives for students, states could add meaning to high school diplomas for colleges and employers by indicating on transcripts that students have earned a college- and careerready distinction by performing at the highest levels on the state assessments, in addition to taking a college preparatory curriculum, completing a series of courses in a career pathway, or mastering other core competencies. Graduation requirements that are entirely competency-based also hold potential to add greater meaning to a high school diploma, providing a clear signal of what students have learned in high school and articulating how these skills transfer to other educational and workforce settings. While not yet realized at-scale in any state, as competency-based assessments are developed further they could offer students multiple, high- 31 32 THE CASE AGAINST EXIT EXAMS quality pathways to demonstrate their proficiency, beyond a single, statewide end-of-year assessment or exit exam. 3) Using the new college- and career-ready assessments as exit exams could also jeopardize their biggest asset: the ability to accurately gauge students’ readiness of college and careers. By tying graduation to a particular score, states will most likely establish that score well below the collegeand career-ready benchmark and be pressured to dilute the rigor of the new standards so that nearly all students can attain the passing score. But instead of offering just a backward view of students’ educational progress—did the student master enough content to be deemed proficient and graduate?—the new assessments would be much more valuable as a forward-looking tool: given the content students have mastered thus far, what should they learn next? By measuring whether students are ready for college and career, the tests could be used in a more diagnostic way for students’ benefit. For high school students already college-ready, accelerated coursework could be the diagnosis, helping them earn college credits before graduating and reducing the time and cost to earn a postsecondary degree. But for those not ready for college and career, high schools could use the test results to offer targeted remediation before students enter higher education and are placed in noncredit-bearing courses. Districts and states must also recognize the need for smarter, and earlier, interventions for struggling students and provide resources and supports to ensure that they are given the opportunity to learn and catch up to their more advanced peers. And given the importance of highquality, effective remediation within the broader effort to see all students college- and career-ready, more research is needed to determine which remediation models are most successful and find ways to execute them at-scale. While some states are already moving to implement these policies and systems, our analysis found that as many as 21 states, including 10 in the PARCC or Smarter Balanced consortia, are planning to have an exit exam requirement, and many states have not yet made a firm decision about what test they will be using, when, and for what purposes. While states are using a multitude of tests, transition strategies, and timelines, policymakers in these states face common issues as they shift toward using college- and career-ready assessments for graduation requirements. 1) Provide clarity. States that have not yet chosen a college- and career-ready assessment should do so as soon as possible, or at minimum, set a timeframe and process for selecting the assessment. In the meantime, state policymakers could still consider whether the new assessment will carry any stakes for students, the timeline for switching to the new testing requirements, what subjects or grades the tests should cover, whether the passing scores should be similar to proficiency benchmarks on current state tests, whether the state would phase out the old test gradually, and the alternatives for students that need testing accommodations or who do not meet the new standards, including retesting options, individual student waivers, or other assessments. 2) Protect test quality. States that plan to use their new college- and career-ready tests, especially those developed by the consortia, as exit exams should prioritize the validity of their assessment systems to accurately measure college and career readiness above all. Rather than tinkering with the college- and career-ready cut scores agreed to by the consortia, states should use a two-cut-score approach. This decreases the likelihood that the college- and career-ready performance level will be manipulated for the sake of maintaining graduation rates. However, this strategy does increase the chances that the lower cut score would face this kind of downward pressure, and states should consider appropriate and effective alternatives for students that cannot meet the required graduation score, no matter where it is set. States will also need to communicate carefully to students and their families what the different cut scores mean and how they are used to avoid conflating graduation requirements with any college- and career-ready requirements. 3) Communication is key. States with exit exams should post transition timelines for their assessments clearly and prominently on their state education agency websites, including the specific assessments and scores students must attain in each graduating class for varying purposes (graduation, college and career readiness, etc.). Further, states could develop materials for districts and schools to share with affected students, and schools and districts could conduct more direct outreach to explain these changes, using in-person conferences and forums, direct mail, social media, and other marketing tools. If particular choices have not yet been made because states are waiting for more information on the new assessments, the time frame for settling these outstanding questions should also be communicated and included in the transition plan. These plans should also include the last administration dates for exams that will no longer be offered, and states should consider whether they want to administer these assessments as long as current students could theoretically use them to meet graduation requirements. 4) Gather data and reassess. States that continue to use exit exam policies should gather student-level data on how these policies affect students once the new standards and assessments are in place. If exit exams are meant to ensure that students are college and career ready in English Language Arts and math when they leave high school, are students’ scores strongly related to their postsecondary outcomes? Are students who pass the exam on the first try any better off than those who take the exam multiple times? Are students who do poorly on the exit exam getting pushed out into GED programs or dropping THE CASE AGAINST EXIT EXAMS out at higher rates, or are the interventions they receive in high school working? These questions should also be asked of exit exams that states administer in other subjects, including science and social studies. In addition to adding to the research literature on the effects of high school exit exams, this information could also encourage better policymaking. If exit exams are not meeting their stated goals, states should consider other policy alternatives. While states are grappling with when and how to use new college- and career-ready standardized test scores for evaluating schools and educators, it is time for this conversation to extend to the stakes placed on test scores for students. The new English Language Arts and math tests hold great potential to help states measure whether students are prepared for college or the workforce. But this promise could be squandered if the new assessments are also used as exit exams, forcing states to choose between using the tests to determine whether students are prepared for college or whether students are able to go to college by earning a high school diploma. And states do not have to make this choice. Unlike exit exams, there are alternative policies that encourage and reward students for meeting higher expectations, without jeopardizing their ability to graduate from high school. There is no doubt that students leaving the K–12 system need to be better prepared for college and the workforce, but they will never get a fair chance to succeed in college or on the job without a high school diploma. Given this dilemma, states need policies that allow them to simultaneously pursue college- and career-ready academic standards—and accountability for meeting them—and college and career attainment for more students. High school exit exams just don’t measure up. NOTES 1 Anthony P. Carnevale, Nicole Smith, and Jeff Strohl, “Recovery: Job Growth and Education Requirements Through 2020,” Georgetown University Center on Education and the Workforce, June 2013, accessed March 15, 2014, http://cew.georgetown.edu/ recovery2020. 2 National Center for Education Statistics, “Table 302.10 Recent High School Completers and Their Enrollment in 2-Year and 4-Year Colleges, by Sex: 1960 Through 2012,” National Digest of Education Statistics, 2013, accessed May 19, 2014, http://nces. ed.gov/programs/digest/d13/tables/dt13_302.10.asp. 3 D. Shapiro, A. Dundar, M. Ziskin, X. Yuan, and A. Harrell, Completing College: A National View of Student Attainment Rates—Fall 2007 Cohort, Signature Report No. 6, (Herndon, VA: National Student Clearinghouse Research Center, December 2013) accessed March 18, 2014, http://nscresearchcenter.org/ signaturereport6/. 4 E. Lauff and S.J. Ingels, Education Longitudinal Study of 2002 (ELS:2002): A First Look at 2002 High School Sophomores 10 Years Later (NCES 2014-363) (U.S. Department of Education. Washington, DC: National Center for Education Statistics, 2013), retrieved March 20, 2014 from http://nces.ed.gov/pubsearch. 5 Lindsey Tepe, Common Core Goes to College (Washington, D.C.: New America Foundation, July 2014). FINAL%20WORKING%20DRAFT%20Education%20Counsel%20 High%20Stakes%20Guidance%20060614.pdf. 10 Naomi Chudowsky, Nancy Kober, Keith S. Gayler, and Madlene Hamilton, “State High School Exit Exams: A Baseline Report,” Center on Education Policy, August 2002, accessed June 16, 2014, http://www.cep-dc.org/displayDocument.cfm?DocumentID=250. 11 Shelby McIntosh, “State High School Exit Exams: A Policy in Transition,” Center on Education Policy, September 2012, accessed January 17, 2014, http://www.cep-dc.org/displayDocument. cfm?DocumentID=408. 12 Data collected by author from state education agency websites, official press releases, media reports, and other public sources. As high school exit exam policies must be communicated clearly and transparently to stakeholders, including adequate notice to families and students, internal conversations or strategic plans among state officials were not counted as evidence of adopting or changing an assessment use policy. 13 Thomas S. Dee, “The ‘First Wave’ of Accountability,” in No Child Left Behind? The Politics and Practice of School Accountability, ed. Paul E. Peterson and Martin R. West (Washington, D.C.: Brookings Institution Press, 2003): 215–241. 6 Complete College America, “Remediation: Higher Education’s Bridge to Nowhere,” April 2012, accessed March 15, 2014, http:// www.completecollege.org/docs/CCA-Remediation-final.pdf. 14 National Commission on Excellence in Education, A Nation at Risk: The Imperative for Educational Reform (U.S. Department of Education, 1983) accessed April 2, 2014, http://www2.ed.gov/ pubs/NatAtRisk/index.html. 7 Judith Scott-Clayton and Olga Rodriguez, “Development, Discouragement, or Diversion? New Evidence on the Effects of College Remediation,” National Bureau of Economic Research Working Paper No. 18328, August 2012. 15 National Research Council, Incentives and Test-Based Accountability in Education (Washington, DC: The National Academies Press, 2011). 8 Complete College America, “Remediation: Higher Education’s Bridge to Nowhere,” April 2012, accessed March 15, 2014, http:// www.completecollege.org/docs/CCA-Remediation-final.pdf. 16 Mark Vanhoenacker, “Don’t Mess With Massachusetts,” Slate, May 14, 2012, accessed June 19, 2014, http://www.slate.com/ articles/news_and_politics/politics/2012/05/massachusetts_is_ the_best_state_in_the_union_.html. 9 “State Transitions to College- and Career-Ready Assessments: A Policymakers’ Guide to Decisions Regarding High-Stakes Student Assessments,” EducationCounsel LLC, June 6, 2014, accessed June 29, 2014, http://www.educationcounsel.com/docudepot/ 17 “Board to set MCAS passing score,” Associated Press, November 23, 1999, accessed June 19, 2014, http://www. southcoasttoday.com/apps/pbcs.dll/article?AID=/19991123/ NEWS/311239958. 33 34 THE CASE AGAINST EXIT EXAMS 18 Linda Shaw, “Lessons from another state’s high stakes test,” The Seattle Times, February 26, 2006, accessed June 19, 2014, http://seattletimes.com/html/localnews/2002829580_ massaschusetts26m.html. 19 John P. Papay, Richard J. Murnane, and John B. Willet, “The Consequences of High School Exit Examinations for LowPerforming Urban Students: Evidence from Massachusetts,” Educational Evaluation and Policy Analysis, March 2010 (32): 5-23. doi: 10.3102/0162373709352530 20 Jennifer Jellison Holme, Meredith P. Richards, Jo Beth Jimerson, and Rebecca W. Cohen, “Assessing the Effects of High School Exit Examinations,” Review of Educational Research, December 2010 (80): 476–526. doi: 10.3102/0034654310383147 21 Olesya Baker and Kevin Lang, “The Effect of High School Exit Exams on Graduation, Employment, Wages and Incarceration,” National Bureau of Economic Research Working Paper No. 19182, June 2013. 22 Steven W. Hemelt and Dave E. Marcotte, “High School Exit Exams and Dropout in an Era of Increased Accountability,” Journal of Policy Analysis and Management, Spring 2013 (32): 323–349. doi: 10.1002/pam.21688. 23 Tom Ahn, “A Regression Discontinuity Analysis of Graduation Standards and Their Impact on Students’ Academic Trajectories,” Economics of Education Review, February 2014 (38.C): 64–75. 24 Shelby McIntosh, “State High School Exit Exams: A Policy in Transition,” Center on Education Policy, September 2012, accessed January 17, 2014, http://www.cep-dc.org/displayDocument. cfm?DocumentID=408. 25 Amanda Ripley, The Smartest Kids in the World: and How They Got that Way (New York: Simon and Schuster, 2013), 185–186. 26 Shelby McIntosh, “State High School Exit Exams: A Policy in Transition,” Center on Education Policy, September 2012, accessed January 17, 2014, http://www.cep-dc.org/displayDocument. cfm?DocumentID=408. 27 Achieve, “Closing the Expectations Gap: 2013 Annual Report,” November 2013, accessed April 20, 2014, http://www.achieve. org/files/2013ClosingtheExpectationsGapReport.pdf. 28 Liana Heitin, “Fewer than 40 Percent of Seniors are Prepared for College, NAEP Analysis Finds,” Education Week, May 14, 2014, accessed June 11, 2014, http://blogs.edweek.org/edweek/ curriculum/2014/05/fewer_than_40_percent_of_senio.html. 29 U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, National Assessment of Educational Progress (NAEP), various years, 1992–2013 Mathematics and Reading Assessments, accessed June 11, 2014, http://nationsreportcard.gov/reading_math_g12_2013/#/whatknowledge. 30 “The Condition of College & Career Readiness 2013,” ACT, Inc., 2013, accessed May 11, 2014, http://www.act.org/research/ policymakers/cccr13/pdf/CCCR13-NationalReadinessRpt.pdf. 31 See, for example, Andrew Ujifusa, “Scores Drop on Ky.’s Common Core-Aligned Tests,” Education Week, November 2, 2012, accessed May 11, 2014, http://www.edweek.org/ew/ articles/2012/11/02/11standards.h32.html; and Andrew Ujifusa, ”Tests Aligned to Common Core in New York State Trigger Score Drops,” Education Week (blog), August 7, 2013, accessed May 11, 2014, http://blogs.edweek.org/edweek/state_edwatch/2013/08/_ one_interesting_aspect_of.html. 32 Bandeira de Mello, V. (2011), Mapping State Proficiency Standards Onto the NAEP Scales: Variation and Change in State Standards for Reading and Mathematics, 2005–2009 (NCES 2011-458). National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education, Washington, DC: Government Printing Office. 33 See, for example, John Perry, Heather Vogell, Alan Judd, and M.B. Pell, “Cheating our children: Suspicious school test scores across the nation,” The Atlanta Journal-Constitution, March 25, 2012, accessed May 10, 2014, http://www.ajc.com/news/news/ cheating-our-children-suspicious-school-test-sco-1/nQSTS/, Greg Toppo, “Memo warns of rampant cheating in D.C. public schools,” USA Today, April 11, 2013, accessed May 10, 2014, http:// www.usatoday.com/story/news/nation/2013/04/11/memowashington-dc-schools-cheating/2074473/, and Michael Winerip, “Ex-Schools Chief in Atlanta Is Indicted In Testing Scandal,” The New York Times, March 29, 2013, accessed May 10, 2014, http:// www.nytimes.com/2013/03/30/us/former-school-chief-inatlanta-indicted-in-cheating-scandal.html?pagewanted=all&_ r=0. 34 Howard Blume, ”All L.A. Unified students must pass collegeprep courses,” Los Angeles Times, May 9, 2012, accessed May 11, 2014, http://articles.latimes.com/2012/may/09/local/la-me0509-lausd-20120509. 35 Stephanie Simon, “The School Standards Rebellion,” Politico, February 14, 2014, accessed May 12, 2014, http://www.politico. com/story/2014/02/education-standards-reform-high-schoolcollege-103510.html. 36 Jill Barshay, “New Common Core high school tests set a low bar for passing in New York,” The Hechinger Report, accessed June 4, 2014, http://hechingerreport.org/content/new-commoncore-high-school-tests-set-low-bar-passing-new-york_16245/. 37 “State Transitions to College- and Career-Ready Assessments: A Policymakers’ Guide to Decisions Regarding High-Stakes Student Assessments,” EducationCounsel LLC, June 6, 2014, accessed June 29, 2014, http://www.educationcounsel.com/ docudepot/FINAL%20WORKING%20DRAFT%20Education%20 Counsel%20High%20Stakes%20Guidance%20060614.pdf. 38 National Center for Education Statistics, “Table 122 High School Graduates, by Sex and Control of School: Selected Years, 1869–70 Through 2021–22,” National Digest of Education Statistics, 2012, accessed May 20, 2014, http://nces.ed.gov/ programs/digest/d12/tables/dt12_122.asp. 39 M. Stetser and R. Stillwell, Public High School Four-Year On-Time Graduation Rates and Event Dropout Rates: School Years 2010–11 and 2011–12. First Look (NCES 2014-391) (U.S. Department of Education, Washington, DC: National Center for Education Statistics, 2014), retrieved May 20, 2014 from http:// nces.ed.gov/pubsearch. 40 Robert Balfanz, John M. Bridgeland, Joanna Hornig Fox, Jennifer L. DePaoli, Erin S. Ingram, and Mary Maushard, “Building a Grad Nation: Progress and Challenge in Ending the High School Dropout Epidemic,” America’s Promise Alliance, April 2014, accessed May 20, 2014, http://gradnation.org/resource/buildinggradnation-progress-and-challenge-ending-high-schooldropout-epidemic-2014. 41 “National Governors Association and State Education Chiefs Launch Common State Academic Standards,” Council of Chief State School Officers, June 2, 2010, accessed June 6, 2014, http:// www.ccsso.org/News_and_Events/Press_Releases/NATIONAL_ GOVERNORS_ASSOCIATION_AND_STATE_EDUCATION_CHIEFS_ LAUNCH_COMMON_STATE_ACADEMIC_STANDARDS_.html. 42 See, for example, Michele McNeil, “Racing For an Early Edge: States Jockey for Position as the U.S. Education Department Readies Billions of Dollars in ‘Race to the Top’ Awards—the Stimulus Program’s Grand Prize,” Education Week, July 9, 2009, http://www.edweek.org/ew/articles/2009/07/09/36stimulus_ ep.h28.html; and Michele McNeil, “All But 10 States Throw Hats Into Race to Top Ring,” Education Week, January 19, 2010, accessed May 20, 2014, http://www.edweek.org/ew/ articles/2010/01/20/19rtt_ep.h29.html. 43 Nancy Kober and Diane Stark Rentner, “States’ Progress and Challenges in Implementing Common Core State Standards,” Center on Education Policy, January 2011, accessed June 29, 2014, http://www.cep-dc.org/displayDocument.cfm?DocumentID=343. 44 Data collected by author from state websites, the archived THE CASE AGAINST EXIT EXAMS websites for the Partnership for Assessment of Readiness for College and Careers and the Smarter Balanced Assessment Consortium, official press releases, and media reports. 45 See, for example, Mark Bauerlein, “Common Core vs. Great Literature,” New York Daily News, July 10, 2013 http://www. nydailynews.com/opinion/common-core-great-literaturearticle-1.1394249; The Hechinger Report, “The Common Core Math Standards: Content and Controversy,” U.S. News and World Report, February 25, 2014, http://www.usnews.com/news/ special-reports/articles/2014/02/25/the-common-core-mathstandards-content-and-controversy; Martha Dalton, “Common Core Assessment Too Expensive for Georgia?” WABE, July 9, 2013, http://wabe.org/post/common-core-assessment-too-expensivegeorgia; Benjamin Herold, ”Tech Challenges Lead Oklahoma to Opt Out of PARCC Exams,” Education Week, July 3, 2013, http://blogs.edweek.org/edweek/DigitalEducation/2013/07/ tech_challenges_lead_oklahoma_.html; and Conor P. Williams, “The Vast Network of Common Core Conspiracy Theories,” Talking Points Memo, April 29, 2014, accessed May 25, 2014, http:// talkingpointsmemo.com/cafe/the-vast-network-of-commoncore-conspiracy-theories. 46 Catherine Gewertz and Andrew Ujifusa, “National Landscape Fragments as States Plan Common-Core Testing,” Education Week, May 20, 2014, accessed June 22, 2014, http://www.edweek. org/ew/articles/2014/05/21/32assessment_ep.h33.html. 47 See, for example, Catherine Gewertz, “Revoking the Common Standards: An Idea Under Consideration in Several States,” Education Week (blog), May 19, 2014, http://blogs.edweek.org/ edweek/curriculum/2014/05/the_herald_rock_hill_south.html; and Andrew Ujifusa, “S.C. Governor Signs Bill Requiring State to Replace Common Core,” Education Week (blog), June 4, 2014, accessed June 10, 2014, http://blogs.edweek.org/edweek/state_ edwatch/2014/06/south_carolina_gov_haley_signs_bill_to.html. 48 Catherine Gewertz, “Massachusetts Won’t Require All Schools to Give PARCC Test in 2015,” Education Week (blog), November 19, 2013, accessed May 29, 2014, http://blogs.edweek.org/edweek/ curriculum/2013/11/massachusetts_wont_require_all.html. 49 New York Board of Regents, “October 2013—Agenda and Materials, PARCC Update,” New York State Education Department, October 2013, accessed May 29, 2014, http://www.regents.nysed. gov/meetings/2013Meetings/October2013/1013monthmat.html. 50 Julie O’Donaghue, ”Bobby Jindal Has a Harder Time Getting Out of Common Core Than Other Governors,” The TimesPicayune, June 19, 2014, accessed June 21, 2014, http://www. nola.com/politics/index.ssf/2014/06/bobby_jindal_louisiana_ common.html. 51 John White, “Teachers Deserve Clarity and a Long-Term Plan on Common Core Standards,” Common Core Watch (blog), June 5, 2014, accessed June 20, 2014, http://edexcellence.net/articles/ superintendent-john-white-teachers-deserve-clarity-and-a-longterm-plan-on-common-core. 52 Danielle Dreilinger, “National Common Core Test Not for Louisiana High Schools Until 2016—If At All,” The TimesPicayune, March 10, 2014, accessed June 20, 2014, http://www. nola.com/education/index.ssf/2014/03/national_common_core_ test_not.html. 53 North Carolina State Board of Education, “Meeting Agenda,” February 5, 2014, accessed June 15, 2014, https:// eboard.eboardsolutions.com/Meetings/ViewMeetingOrder. aspx?S=10399&MID=1164. 54 Mark Binker, “Senate Approves Version of Common Core Repeal Plan,” WRAL, June 5, 2014, accessed June 15, 2014, http:// www.wral.com/senate-approves-version-of-common-corerepeal-plan/13704539/. 55 See, for example, James Chilton, ”New Multi-State Assessment Could Prepare Wyoming Juniors for College,” Wyoming Tribune Eagle, March 19, 2014, http://trib.com/news/local/education/ new-multi-state-assessment-could-prepare-wyoming-juniorsfor-college/article_db7f25a1-d10e-5fe1-858f-5601197ffa2e. html; and Mike Wiser, “Iowa Legislators Find Fault with Common Core,” The Gazette, March 24, 2014, accessed April 15, 2014, http://thegazette.com/2014/03/24/iowa-legislators-find-faultwith-common-core/. 56 See, for example, Brian Smith, “Michigan Teachers Scramble to Create Lesson Plans, Now that MEAP Test is Back On,” mlive. com, June 29, 2014, http://www.mlive.com/lansing-news/index. ssf/2014/06/meap_changes_what_to_expect.html; and Andrew Ujifusa, “Smarter Balanced Remains Best Common Core Testing Option, Michigan Dept. Says,” Education Week, December 2, 2013, accessed June 29, 2014, http://blogs.edweek.org/edweek/state_ edwatch/2013/12/smarter_balanced_remains_best_common_ core_assessment_option_michigan_dept_says.html. 57 Andrew Ujifusa, “Michigan Drops Common-Core Test, But Might Still Use Its Questions,” Education Week, June 30, 2014, accessed July 1, 2014, http://blogs.edweek.org/edweek/state_ edwatch/2014/06/common-core_test_dropped_.html. 58 Brian Smith, “MEAP Test ‘Is Not an Option’ for Next School Year, State Superintendent Mike Flanagan Says,” mlive.com, April 23, 2014, accessed May 12, 2014, http://www.mlive.com/lansingnews/index.ssf/2014/04/meap_test_is_not_an_option_for.html. 59 Shelby McIntosh, “State High School Exit Exams: A Policy in Transition,” Center on Education Policy, September 2012, accessed January 17, 2014, http://www.cep-dc.org/displayDocument. cfm?DocumentID=408. 60 Catherine Gewertz, “States Grapple With Common Test-Score Cutoffs,” Education Week, December 10, 2013, accessed June 11, 2014, http://www.edweek.org/ew/articles/2013/12/11/14naep. h33.html. 61 “Colorado High School Graduation Guidelines,” Colorado Department of Education, May 2013, accessed May 19, 2014, http://www.cde.state.co.us/sites/default/files/ adoptedgraduationguidelines2013.pdf. 62 Michele McNeil, “Many States Left Key NCLB Flexibility on the Table,” Education Week, April 10, 2014, accessed June 1, 2014, http://www.edweek.org/ew/articles/2014/04/10/28multiple.h33. html. 63 “Early Assessment Program,” California Department of Education, January 30, 2014, accessed June 1, 2014, http://www. cde.ca.gov/ci/gs/ps/eapindex.asp. 64 John Fensterwald, “What’s Next for Standardized Testing in California?” EdSource (blog), October 22, 2013, accessed June 1, 2014, http://edsource.org/2013/whats-next-for-standardizedtesting-in-california/40454#.U5TXDPmwLYg. 65 “College Readiness Indicators,” Kentucky Council on Postsecondary Education, April 3, 2014, accessed June 20, 2014, http://cpe.ky.gov/policies/academicinit/deved/. 66 Lindsey Tepe, Common Core Goes to College (Washington, D.C.: New America Foundation, July 2014). 67 Data collected by author from state education agency websites, official press releases, media reports, and other public sources. As high school exit exam policies must be communicated clearly and transparently to stakeholders, including families and students, internal conversations among state officials were not counted as evidence of adopting or changing an assessment use policy. 68 Seanna Adcox, “New South Carolina Law Deletes High School Exit Exam Mandate,” Associated Press, April 21, 2014, accessed June 20, 2014, http://www.postandcourier.com/article/20140421/ PC1610/140429883. 69 Emily Pace, “SC Legislature Approves Bill To Drop Exit Exam As Must For Graduation,” WSPA, April 10, 2014, accessed June 20, 2014, http://www.wspa.com/story/25218743/sc-legislatureapproves-bill-to-drop-exit-exam-as-must-for-graduation. 35 36 THE CASE AGAINST EXIT EXAMS 70 Shelby McIntosh, “State High School Exit Exams: A Policy in Transition,” Center on Education Policy, September 2012, accessed January 17, 2014, http://www.cep-dc.org/displayDocument. cfm?DocumentID=408. 71 Andrew Ujifusa, ”Teacher, School Accountability Systems Shaken Up,” Education Week, June 10, 2014, accessed June 22, 2014, http://www.edweek.org/ew/ articles/2014/06/11/35accountability.h33.html. 72 Will Weissert, “81 percent of Class of 2015 Has Passed All Exams,” Associated Press, June 11, 2014, accessed June 18, 2014, http://www.houstonchronicle.com/news/texas/article/81percent-of-Class-of-2015-has-passed-all-exams-5544924.php. 73 See, for example, John Mooney, “Major Changes on Way for New Jersey’s High School Tests,” NJSpotlight.com (blog), November 20, 2013, http://www.njspotlight.com/ stories/13/11/19/new-jersey-s-high-school-tests-facingmajor-changes-to-come/; and Christopher D. Cerf, “Educator Evaluation/Common Core/PARCC Facts,” memorandum to Chief School Administrators and Charter School Lead Persons, March 5, 2014, accessed June 8, 2014, http://education.state.nj.us/ broadcasts/2014/MAR/05/10943/Cerfmemo_cc_parcc%20(2).pdf. 74 Bill Roberts, “Follow the Testing Maze: Common Core will Bring Plenty of Changes,” Idaho Statesman (blog), October 2, 2013, accessed May 20, 2014, http://blogs.idahostatesman.com/ follow-the-testing-maze-common-core-will-bring-plenty-ofchanges/. 75 Paul Takahashi, ”State Board of Education Lays Out New Rules for Graduation Requirements,” Las Vegas Sun, April 17, 2014, accessed June 5, 2014, http://www.lasvegassun.com/news/2014/ apr/17/state-board-education-lays-out-rules-new/. org/2013/future-of-high-school-exit-exam-unclear-as-californiarevamps-testing-requirements/39146#.U55cY5StkvE. 83 Linda Wallinger and Shelley Loving-Ryder, “Virginia’s College and Career Readiness Initiative,” Virginia Department of Education, June 13, 2012, accessed May 29, 2014, http://www. schev.edu/council/VAG/Wallinger-SCHEV-June2012.pdf. 84 Michael Alison Chandler, ”Virginia Reading Scores Drop on More Difficult SOL Test,” The Washington Post, August 20, 2013, accessed May 29, 2014, http://www.washingtonpost. com/local/education/virginia-reading-scores-drop-by-doubledigits-on-new-sol-test/2013/08/20/3ecf1726-09a3-11e3-b87c476db8ac34cd_story.html. 85 Jill Barshay, “New Common Core High School Tests Set a Low Bar for Passing in New York,” The Hechinger Report, accessed June 4, 2014, http://hechingerreport.org/content/new-commoncore-high-school-tests-set-low-bar-passing-new-york_16245/. 86 Lillian M. Lowery, “Transition from Current High School Assessments to PARCC High School Assessments,” memorandum to Members of the State Board of Education, February 25, 2014, accessed May 20, 2014, http://marylandpublicschools. org/stateboard/boardagenda/02252014/Tabs_J1_J2_J3_J4_ MemoBoardTransitionfromHSA_to_PARCC.pdf. 87 See, for example, “Testing Students in Washington State,” Washington State Office of Superintendent of Public Instruction, May 15, 2014, http://www.k12.wa.us/assessment/statetesting/; and Patrick O’Donnell, “Ohio Students Must Now Pass Common Core Exams or Other Tests to Graduate, State Legislature Decides,” The Plain Dealer, June 4, 2014, accessed June 14, 2014, http://www.cleveland.com/metro/index.ssf/2014/06/ohio_ students_must_pass_common.html. 76 See, for example, Chris Sturgis, “New Hampshire Rocks Competency Education Policy,” CompetencyWorks (blog), November 25, 2013 http://www.competencyworks.org/2013/11/ new-hampshire-rocks-competency-education-policy/; and “College and Career Ready Competencies Aligned with Common Core State Standards,” New Hampshire Department of Education, February 2013, accessed June 14, 2014, http://www.education. nh.gov/competencies/. 88 “Graduation Requirements FAQs,” New Mexico Public Education Department, accessed June 14, 2014, http://ped.state. nm.us/ped/Graduation_FAQ.html. 77 Catherine Gewertz, “N.H. Schools Embrace CompetencyBased Learning,” Education Week, February 7, 2012, accessed June 14, 2014, http://www.edweek.org/ew/ articles/2012/02/08/20proficiency_ep.h31.html. 90 See, for example, Catherin Gewertz, “Is Mass. Leading the Way to a More Cautious Route on Common Tests?” Education Week (blog), November 20, 2013 http://blogs.edweek.org/edweek/ curriculum/2013/11/states_taking_cautious_route_o.html; and “Higher Expectations, Higher Achievement: CCSS-aligned Assessments,” Massachusetts Department of Education, 2012, accessed June 13, 2014, http://www.mde.k12.ms.us/ms-collegecareer-standards/ccss-faq/lists/common-core-faqs/ccss-alignedassessments. 78 See, for example, Noel K. Gallagher, “Maine’s New Graduation Mandates Challenge School Districts,” Portland Press Herald, June 10, 2014, http://www.pressherald.com/2014/06/10/mainesnew-graduation-requirements-challenge-school-districts/; Josh O’Gorman, “Vt. Schools to Create New High School Proficiency Standards,” Valley News, June 7, 2014, http://www.vnews.com/ news/12274494-95/vt-schools-to-create-new-high-schoolproficiency-standards; and Todd Engdahl, ”Board Pushes Ahead on Grad Guidelines,” Chalkbeat Colorado (blog), May 15, 2013, accessed June 14, 2014, http://co.chalkbeat.org/2013/05/15/ board-pushes-ahead-on-grad-guidelines/. 79 “Competency-Based Pathways State Partnership,” Achieve, 2014, accessed June 14, 2014, http://www.achieve.org/CBP. 89 “Statewide Assessments & Essential Skills Transition Plan,” Oregon Department of Education, February 2012, accessed June 12, 2014, http://www.ode.state.or.us/wma/teachlearn/ commoncore/essential-skills-oaks-to-sbac-transition.pdf. 91 “Draft Recommendations for Use of the 11th Grade Smarter Balanced Assessment,” The Washington State Board of Education, March 2014, accessed June 15, 2014, http://www.sbe.wa.gov/ documents/BoardMeetings/2014/March/02ResponseToSBAC.pdf. 92 “Frequently Asked Questions about State Testing,” Washington State Office of Superintendent of Public Instruction, accessed June 15, 2014, http://www.k12.wa.us/assessment/StateTesting/ FAQ.aspx. 80 Taylor White, “Giving Credit Where Credit’s Due: A 50-State Scan of Course Credit Policies,” Carnegie Foundation for the Advancement of Teaching, August 5, 2013, accessed June 14, 2014, http://commons.carnegiefoundation.org/wp-content/ uploads/2013/08/CUP_Policy_MayUpdate.pdf. 93 “Regents Adjust Common Core Implementation,” New York State Education Department, February 10, 2014, accessed May 29, 2014, http://www.oms.nysed.gov/press/regents-adjustcommon-core-implementation.html. 81 Michele McNeil, “California Wins Prized NCLB Testing Waiver,” Education Week (blog), March 7, 2014, accessed June 1, 2014, http://blogs.edweek.org/edweek/campaign-k-12/2014/03/ california_wins_prized_nclb_te.html. 94 See, also “Measures that Matter, Making College and Career Readiness the Mission for High Schools: A Guide for State Policymakers,” Achieve and The Education Trust, November 2008, accessed June 30, 2014, http://www.achieve.org/files/ MakingCollegeandCareerReadinesstheMissionforHighSchool.pdf. 82 Kathryn Baron, ”Future of High School Exit Exam Unclear as California Revamps Testing Requirements,” EdSource (blog), September 18, 2013, accessed June 1, 2014, http://edsource. 95 National Research Council. Incentives and Test-Based Accountability in Education (Washington, DC: The National Academies Press, 2011). New America Education Policy Program 1899 L Street, NW Suite 400 Washington DC 20036 Phone 202 986 2700 Fax 202 986 3696 www.newamerica.org www.edcentral.org