Common Standards for K-12 Education?: Considering the Evidence: Summary of a Workshop Series (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

7 Perspectives on Common Standards Q uestions about the effects a move to common standards might have and about which issues would deserve the highest priority if this policy initiative were undertaken may look different to people with different perspectives on education policy and practice. The commit- tee thus invited discussants representing three important perspectives— researchers, elected officials, and implementers—to reflect on the advan- tages and disadvantages of the options for common standards, asking them to use the options and evaluation framework described in Chapter 3 to structure the conversation. This discussion was preceded by a look at two examples of what discussant Lynn Olson described as an existence proof that shared standards can be successful. TWO EXAMPLES The idea of common standards across ages and subjects for the nation is thus far largely hypothetical, but two programs provide examples of efforts to share standards. The end-of-course exam in algebra II being developed by Achieve as part of the American Diploma Project, and the New England Common Assessment Program (NECAP) are both cases in which groups of states are implementing shared standards in the context of assessment programs. Matthew Gandal described the Achieve project and Peter McWalters described NECAP. 59 

60 COMMON STANDARDS FOR K-12 EDUCATION? Achieve Achieve did not set out to develop common standards, Gandal explained, although the organization was founded as a way for states to compare notes and benchmark their efforts as they pursued standards- based reform. As part of that work, Achieve has provided evaluations of states’ standards and assessments, comparing them in terms of their rigor and other characteristics. Achieve collaborated with the Fordham Foundation, the Education Trust, and the National Alliance of Business to launch the American Diploma Project (ADP), which initially focused on defining the core skills young people need to succeed in college and the workplace. Ultimately, the group developed specific benchmarks in English and mathematics as a more concrete guide to the states that joined the ADP network. That network, which now includes 32 states, asks each member state to sign on to a program of aligning standards, assessments, graduation requirements, and data and accountability systems at the K-12 level with postsecondary and workplace expectations, as exemplified in the ADP benchmarks. Achieve has worked with 20 of the participating states to update their standards and align them with the ADP benchmarks and with the expectations of the postsecondary institutions and employers in the states. They have found that a common core of knowledge and skills is increas- ingly shared across these states. Individual states may vary in the ways they present their standards, in the weight they attach to different ele- ments, and in other ways, but the core has become increasingly easy to see, Gandal explained. Recognizing the critical role played by assessments, Achieve has also worked with 13 states to oversee the development of an algebra II assess- ment designed to be used at the end of the course. The participating states recognized that they were emphasizing the same material and that it would be logical to work together to develop a high-quality test. Inter- estingly, Gandal noted, ironing out the procurement procedures across the states so that they could join forces in hiring a testing contractor was in some ways more challenging than hammering out the content and performance expectations. For Gandal, the experience has been a heartening proof of what is possible: “If states, working collaboratively, can agree on common stan- dards and develop assessments in common, they could actually accom- plish some of what you have been talking about in this room, which is ultimately coming up with a common expectation and measuring it.” 

PERSPECTIVES ON COMMON STANDARDS 61 New England Common Assessment Program The process Achieve has gone through with algebra II bears some striking similarities to NECAP, which is a collaboration among New Hampshire, Rhode Island, and Vermont. The departments of education in these states have developed grade-level expectations and test specifica- tions in reading, writing, mathematics, and science. In New Hampshire, NECAP is administered in grades 3-8 and 11 in mathematics and reading, and in grades 5, 8, and 11 in writing. Science was administered for the first time in May 2008, in grades 4, 8, and 11. Results are reported on a common scale. McWalters, the commissioner of education for Rhode Island, explained that his state was eager to collaborate with its neighbors because it is very small and demographically disadvantaged in comparison to its neigh- bors. Moreover, he joked, “in anything we do, we always come out the lowest in New England.” Education leaders in Rhode Island were not sat- isfied with the state’s existing standards. The collaboration was relatively easy to establish, he added, because in all three of the states the depart- ment of education had the authority to modify or replace the academic standards. In McWalters’s view, one of the most important benefits of the pro- gram has been the political cover it has provided for the tensions that arise when any state attempts to raise expectations for students. The NECAP assessments were more challenging in many respects than the ones Rhode Island had used previously, and as a result some of the early results, particularly those for secondary mathematics, were disappointingly low. However, as these results came out, he explained, “nobody blinked.” The leaders from the three states were resolute in resisting pressure to lower the proficiency cut points. McWalters observed that the states’ decision was vindicated, since their NECAP results now line up well with those from the National Assessment of Educational Progress assessments. Based on the success of NECAP thus far, McWalters is optimistic about the potential benefits of incremental collaboration of this sort, argu- ing that states should “keep sharing, keep finding those opportunities when systems are about to go through their review anyway. But use sys- tems like Achieve to find others that are close enough—don’t drag in the lowest one on the national ranking and the highest one.” Moreover, he believes that the process has highlighted the biggest unaddressed prob- lem in New Hampshire and around the country, which is the reality that different sets of students are held to different expectations and provided with different sets of opportunities. “Once you get past the sixth grade, and even on our NECAP test, you can see the cohort starting to flatten, because we are starting to track [students] in middle school. Once you start tracking in this high-standards environment, you have students who 

62 COMMON STANDARDS FOR K-12 EDUCATION? have never seen the content on the test.” An accountability system that is not also a capacity-building system cannot solve that problem. RESEARCHERS Three researchers, Andrew Isaacs, Brain Rowan, and William Schmidt, offered their perspectives. Andrew Isaacs focused on challenges to some of the assumptions he thought had undergirded much of the discus- sion. He described as “wobbly” the proposition that “a more centralized national curriculum would lead to higher student achievement, and that higher student achievement in turn would lead to increased economic competitiveness.” He pointed out that the United States is one of only a handful of countries that does not have a national curriculum—and that not only countries that outperform it by whatever measure, but also most countries that perform less well, have a national curriculum. Moreover, a variety of other factors—such as teacher quality, family influences, and respect for education—seem likely to have just as powerful an effect as the curriculum. Isaacs also argued that the policies of the Federal Reserve Bank and other factors are likely to have far more significant effects on the nation’s economic performance than the nature of its standards and curricula. He also challenged the idea that the variability in grade level expecta- tions is a significant problem, because he believes it generally reflects the lack of a research basis, in many cases, for placing a particular topic at a particular grade level. Despite that, there is “an awful lot of uniformity already in the math curriculum.” His fear was that if the curriculum is very tightly controlled, innovation will be discouraged. Moreover, he suggested, the so-called math wars are continuing, so if there is just one curriculum, there is the chance that it could go in an unproductive direc- tion, as he suggested it has in California. With regard to the framework, his view was that if the nation were to pursue common standards, a trusted national group, such as the National Governors Association, not the federal government, should lead the effort, and it should be allowed to develop slowly, to provide time for natural collaboration. In his opinion, “other subjects need this a lot more than mathematics—in mathematics we have a huge amount of agreement about what should be taught.” Brain Rowan organized his observations of public education around inputs, outputs, and process, arguing that the nation has been focusing on the first two at the expense of the last. Since the nation’s education bureaucracy was founded, he argued, people have been working on input standards: “we have adequacy studies and rules about teacher certifica- 

PERSPECTIVES ON COMMON STANDARDS 63 tion, and about how kids get to school, and so forth.” More recently, the nation has developed “a fairly elaborate, if imperfect, system of output controls,” particularly through the assessment requirements associated with the Elementary and Secondary Education Act and the No Child Left Behind Act. However, he argued, “we don’t have any professional or even bureau- cratically consensual rules about how to turn these inputs into outputs.” In Rowan’s view, what is missing is an adequate investment in evidence- based practice, as used in the practice of medicine. If a policy shift is to be pursued, he would advocate reducing spending on assessment and accountability systems and using the savings to build knowledge of how to improve outputs—outcomes for students. This could be done, for example, by relying more on adaptive testing and sampling—collecting fewer data overall and focusing on the information that could be useful in specific ways to the enterprise of improving student learning. In Rowan’s view, “scientifically, the worst-case scenario would be for the United States to settle on a single national test.” He believes that hav- ing a variety of tests makes it possible to evaluate evidence on particular hypotheses. He believes that “a diversity of standards, of processes, and of tests would generate more hypotheses and better evidence than would a system that only had one way of doing business.” William Schmidt used the elements of the evaluation framework— quality, equity, feasibility, and opportunity costs—to reinforce several points. First, he stressed the critical importance of the sequence in which topics are covered, saying, “it is not arbitrary. You can’t just flow through topics around and across grades.” He argued that there are very big dif- ferences in the degree to which state standards are coherent in the way that mathematicians would define coherence. Furthermore, “no other country in the world that I am familiar with has this conversation about alignment, because they don’t even think there would be any alternative to [an aligned system].” The result is a lack of equity. He argued that it is unreasonable to expect to eliminate black-white gaps and gaps between students with low and high socioeconomic status, until the causes of the inequity in students’ educational opportunities are addressed. He finds it appalling that the United States tolerates the reality that different groups of children are offered very different content, at least in mathematics and science. Thus, rather than worrying about the cost of moving toward com- mon standards, he suggested considering instead the cost of not doing so. “There are huge costs there. We face them every day. We hear about them from the business world.” 

64 COMMON STANDARDS FOR K-12 EDUCATION? ELECTED OFFICIALS Julie Bell, Rae Ann Kelsch, and Roy Romer commented, each from a different vantage point, on the way elected officials may look at the question of common standards. Julie Bell began by explaining that her organization, the National Conference of State Legislatures (NCSL) rep- resents not just individual representatives but the 50 legislatures, and that they are entirely bipartisan. Thus their mission is not to advocate policy positions to the states, but to support the legislative process by providing information and analysis. Nevertheless, the NCSL has opposed NCLB because its members took the position that the law’s provisions were not an appropriate role for the federal government in education policy. Thus, with respect to the question of common standards the NCSL’s members are very interested in the possibilities for improvements, but they have a great deal of concern about how such a policy could be national but not federal, worrying that the federal government would be “tempted . . . to jump in and take control.” Deep concern about global economic competitiveness is dominating discussion of education in legislatures around the country, Bell explained. State officials are worried about the variability in educational opportunity, but they tend to focus on evening things out in their own states, rather than at the national level. She also has found that standards are not a pri- mary focus for legislators, who are more interested in issues they see as more “tangible . . . teaching quality and spending and top quality school leaders who can make a difference for schools.” For Rae Ann Kelsch, a state representative from North Dakota, the first observation was that legislators are very interested in what they hope to gain from a given investment. She noted that the workshop discussion “zipped right into ‘how do we implement the standards, how do we go about developing the standards?’ before identifying what are we trying to achieve if we go to a common standards approach in the United States.” Will this approach improve graduation rates, raise achievement levels, or get more kids into college, for example? She thought legislators would be very interested in some of the poten- tial benefits she heard described at the workshop. The idea of having consistent information, so that the significance of comparative data about students is clear, is very appealing. She thought her colleagues would be very interested, for example, in the data showing that North Dakota students perform less well on NAEP than on the state’s own assessments. And legislators are likely to be in favor of opportunities to avoid costly mistakes—not to have to “go through what other states went through the first time they developed new standards.” Kelsch also observed that the concept of standards is not particularly well understood in her state—and that the idea that the state needs to 

PERSPECTIVES ON COMMON STANDARDS 65 prepare to compete economically with other nations is only beginning to take root. Consequently, residents and legislators tend to think their stan- dards are fine as they are and to be relatively uninterested in what other states may be doing. So the preliminary reaction to the idea of common standards may be “it would be okay if we adopted common standards, as long as they were my common standards.” Roy Romer, former governor of Colorado, acknowledged the chal- lenges of building understanding of the challenges the nation faces in improving education for all students, saying that “we have got to bring to America an awareness of the peril that they are in.” He believes that Americans are genuinely worried that “their eighth grader is not going to be able to afford the house they now live in, or send their children to college.” While many may see that the answer to that worry is to build skills and knowledge, consensus about the urgency and the strategy to address it has not yet been forged. What is the best mechanism for making common standards a reality? Romer suggested starting with two or three models as experiments, tak- ing some time to develop evidence about what is successful, and allowing others to observe the benefits. He acknowledged the profound skepticism regarding the federal government’s involvement but believes that the U.S. Department of Education can provide invaluable support, supplying states with the tools they need to move forward on their own. Rather than responding to federal mandates, he argued, states should be benchmark- ing their performance annually against that of the 10 best nations in the world. Here again, the federal government can facilitate that effort and supply resources without serving as an enforcer. Romer agreed with earlier observations that testing is driving the current system in unhelpful ways, but argued that testing still plays a critical role. States need better information from testing and higher qual- ity assessments that have a more positive impact. But states cannot do this alone; “it is just too expensive.” The primary value of a common standards approach for Romer is that the necessary improvements cannot be accomplished at the school or even the district level. State legislatures need to undergo a paradigm shift, to see that collaboration offers them a valuable opportunity, rather than the burden of being told what to do. IMPLEMENTERS The three panelists who were asked to reflect on implementation issues were David Driscoll, James Liebman, and Richard Patz. As commis- sioner of education in Massachusetts, Driscoll oversaw the development of the state’s curriculum frameworks, implementation of the Massachu- setts Comprehensive Assessment System (MCAS), and state accountabil- 

66 COMMON STANDARDS FOR K-12 EDUCATION? ity assessment and development administration of new teacher licensure regulations. He described what he saw as the most important implemen- tation issues, beginning with political consensus. In Massachusetts, he explained, “the stars lined up. The business community led it, the gov- ernor was behind it, the legislature was behind it—Republicans, Demo- crats. The people in Massachusetts were ready. They saw that we had a problem.” With the goal established, the next step was to develop the specific “carrots and sticks” that brought about specific changes. The state was willing to fund the improvements it wanted, allocating $2 billion in new monies between 1993 and 2000. Massachusetts also attached new account- ability measures to the initiatives, starting with a tenth grade test that students would have to pass in order to graduate. Setting the pass score for this test was a delicate task, Driscoll noted: “it is like being a safe- cracker; you have got to get it just right.” There was a push to set a high, meaningful standard, but the fear was that if too many students failed, the initiative would sink. Driscoll agreed with Romer that “we are sleeping through a crisis.” He believes the evidence is very clear that other countries hold their students to higher standards and that the United States will struggle to compete financially so long as that is true. Yet the nation has not had the will to change the schools. “If Horace Mann came down on earth today, the only institutions he would recognize are our schools.” Because no one has rethought the school schedule, for example, during the past century, “we dump millions of kids into the streets of Detroit and Miami at 2:30 in the afternoon with nothing to do and no supervision.” We offer them virtually nothing during the summer, and they come back in September having forgotten all that they learned.” Yet solutions to these problems have already been figured out, and schools could be taking advantage of them. Driscoll believes the U.S. president can begin the work of creating a sense of urgency about this issue, and that a commission, akin to the 9-11 Commission, could develop a specific plan. He believes states are “not all that far apart in their thinking about what kids should know and be able to do. The questions are about how high you set the standards.” He also believes that the cost of setting common standards would not be prohibi- tive. The high cost will come with the changes required to get all students up to that standard. This will include two main elements: supports for struggling students (such as summer school, after school, and special education and psychological services) and real improvements in instruc- tion. Driscoll illustrated the point with an observation about how amazed foreign educators are when they visit U.S. classrooms. School leaders from other countries may have sophisticated systems for guiding teachers and 

PERSPECTIVES ON COMMON STANDARDS 67 helping them improve, but principals in the United States tend to focus on external features, or, as Rowan suggested earlier, the inputs: “You are doing a nice job, nice bulletin boards, you vary your instruction nicely, you’ve got some nice little overheads there, you are using the smart board.” Meanwhile, opportunities to figure out better ways to help the students who are falling behind are missed. James Liebman brought to the discussion his perspective as chief accountability officer for the New York City Department of Education, and largely seconded Driscoll’s comments. He described the approach in New York City, which has been to use standards and accountability to centralize the definitions of what students should learn, but to use a vari- ety of strategies to empower schools to accomplish that goal. He believes that the central authority, whether the U.S. Department of Education, a state agency, or a district, should set initial, provisional learning goals and commit to revising the goals, once experience shows what is possible. This balance of power, in which local entities (schools, districts, or states) have the authority to make some of their own decisions but are strictly accountable for the results, works best, he explained, when these entities see that they are being compared with their peers. Thus, in New York, the performance of the best school of a particular type becomes the standard that all other schools of that type need to meet. The other schools believe they can meet it because one of their own has done it. The city grades schools, rewarding those that succeed and sanctioning those that fall behind in various ways. It offers comprehensive training in data management, performance management, and knowledge man- agement, as well as a diagnostic assessment program that helps schools understand the data and use them to make improvements. Liebman observed that “regression analyses don’t work.” The data on which schools (or districts or states) are evaluated has to be transparent so that it is easy for them to calculate and track their performance. In order to compare like entities to like entities without complex regression analyses, he suggested, one can use simple matching strategies using, say, five or six of the criteria that would go into the regression. Focusing on longitu- dinal student progress, rather than proficiency scores, is also important. “Low-performing students tend to progress further,” Liebman explained. So measuring their progress “helps to balance out the proficiency bonus that the states with high-performing kids get.” Liebman’s suggestion for pursuing common standards was that the federal government or some prestigious private entity could set provi- sional standards, perhaps only in a few subjects at first, and then establish a very clear grading system that focuses on both proficiency and growth to compare the performance of similar states. The focus would be on out- comes, not on the way states decide to establish their standards, but Lieb- 

68 COMMON STANDARDS FOR K-12 EDUCATION? man advocated rewarding states that perform well, progressively taking money or support away from those that do poorly and don’t improve. Finally, Richard Patz, vice president of research at CTP McGraw-Hill, offered the perspective of long-time education publisher. He has seen the textbook publishing and test development industry change significantly in response to NCLB. While the states are now all required to focus on reading and mathematics in grades 3 through 8, he explained, the differ- ences from state to state in terms of both content standards and testing practices have increased. The need to respond to this very varied market has meant that the publishers’ work is very inefficient and far less profitable for testing com- panies than many people believe. In Patz’s view, “building custom state tests to custom standards in a completely unique way in each state overall is just spending a lot of money that you would not have to spend if . . . there were some consensus around standards. It is very hard to leverage what we are doing in one state in another context.” Moreover, he believes that “you can do measurement in a much better way than it is currently being done, where we ask all 500,000 California fifth graders the same 40 or 45 questions.” Computer adaptive testing is an excellent example. In most states there are too many students and too few computers to do it on a large scale, as the annual proficiency require- ments of NCLB dictate. However, this technology could allow assessment of much broader domains, which could facilitate benchmark reference assessment, as well as matrix sampling of student populations, to facili- tate more informative comparisons of their progress. Patz is in a position to see the effects of the significant differences in the quality of states’ standards and assessments. He believes that states underutilize the testing information they now have, and that with differ- ent, better information, they could answer more important questions—and better understand not just which teacher has higher performing students, but more about why, which programs and textbooks work best, and so forth. Nevertheless, he stressed that states that want to do these things well must be prepared to spend what it takes.