ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 20 The Effects of Assessment: a Reflection from within the Economic Worldview in Education Olivier Michaud In a seminar held in my university last fall on education policy research, I was, like many of my colleagues, as-tonished at the dominance of the economic worldview in the shaping of American education.1 The question of what good education is was surprisingly absent from the book used in the seminar: Handbook of Education Policy Research (Sykes, Barbara, & Plank, 2009). The majority of the authors of this book were more interested in how we could make the educational system more efficient. More to the point, they were interested in how efficiency can be measured. This is why the main methodology used by the researchers was quantitative. Indeed, the assumption is that only quantitative methods allow for a more accurate judgment of reality at large, in this case the educational system, and show possible links between causes and effects. The circle is complete: the economic worldview dominates the discussion on all aspects of policy research. The economic worldview in education gives us, as its main advice, to treat the educational system with the medi- cine of the market. We have to make the educational system efficient through the use of the business model. The prerequisite of this model is a simple outcome, because only a simple outcome can be measured or assessed clearly. Finally, we evaluate the model through the logic of the model, i.e., the predominance of quantitative methods. Everything is now understood through the lens of quantity and numbers. I don’t know for what reason my colleagues and I were surprised by the dominance of the economic world- view in education policy research. Aren’t we already aware that this discourse is shaping most of the public sphere and, by extension, invading even the private sphere? Bourdieu states it in a simple way: “everywhere we hear it said, all day long—and this is what gives the dominant discourse its strength—that there is nothing to put forward in opposition to the neo-liberal view, that it has succeeded in presenting itself as self-evident, that there is no alternative” (cited in Hursh, 2001, p. 3). There is no reason why the same logic that is taking over the pub- lic domain will stop in face of the sacred role of education in our democracies. According to Milton Friedman, father of all the theories that aim at reshaping education through economy (Belfield & Levin, 2009, p. 516; Carnoy, 2009, p. 31; Witte, 2009), education is not a different good than other goods, there is no essential dif- ference between food and education: “The role of the government, in market-education, says Friedman, would be limited to insuring that schools met certain minimum standards, such as inclusion of minimum common content in their programs, much as it now inspects restaurants to insure that they maintain minimum sanitary standards” (quote of 1962 cited in Witte, 2009, p. 491). Therefore, the invasion of higher education by issues of assessment and measurement has to be put in a larger perspective. First, we have to recognize that higher education is poised to succomb to the same economic logic that has taken over virtually every other dimension of education. Second, much of what determines educa- tion policy is largely a response to the market ideology that shapes the public domain. The goal of this article is to offer a rationale against this dominant economic model by analyzing one aspect of it: high stakes testing. Ac- cording to Jenlink and Austin (2004), “the standards movement now dominates discussions about all aspects of education – teaching and learning, curriculum, and assessment – as well as all aspects of educator preparation” (p. 3). Consequently, the choice to look more closely into high stakes testing, rather than some other feature, is not a trivial one, but speaks to the core elements of the economic ideology in education. Rather than see this ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 21 situation as hopeless, I aim to show, through several qualitative studies on K-12 education, that we can use this type of economic logic against itself. But before we do that, let’s review a short history of the economic worldview in education. A short history and presentation of the economic worldview in education The educational situation in which we encounter ourselves today did not pop up accidentally in recent years; on the contrary, it has a long history. As said earlier, the theoretical roots of this movement at least go back to Milton Friedman in the sixties. However, it is with the publication of A Nation at Risk in 1983 that economic educational theory took a new direction that became predominant in the shaping of American education. Ac- cording to Cuban (1998), all the changes in educational policy in the United States since the publication of A Nation at Risk have been superficial, since all the politicians of the different parties and policymakers have adhered to the main principles of this work (p. 463). If there has been a change in the actors and the names of educational reforms, there has been no change in the basic principles. This trend continues up to the present: No Child Left Behind (NCLB) is the continuation and the intensification of the changes proposed in A Nation at Risk (Olsen & Sexton, 2009, p. 9; Shoen & Fusarelli, 2008, p. 190). The development of the economic view in education is based on the idea that the American schooling system is failing children and American society. First, the educational system does not prepare the American workforce for the global economy. Second, it is failing because scores on various tests have not improved in the last decades even if the investment per student has increased significantly. The educational economists took a diagnostic of why the American schooling system has been failing. On one hand, it is failing because the laws of the market are inexistent in education. As a consequence, an important part of the solution to reform the failing system is to apply the laws of the market to education, which means improving competitiveness. This is the basis of what is called “public choice” in education (West, 2009). Charter schools and voucher programs (Belfield & Levin, 2009; Vergari, 2009; Witte, 2009) are some of the direct effects of a policy that aims to create competition and choice in the American educational system. The creation of competition inside American schooling is twofold: first, it facilitates the creation of private schools and, second, it creates competition inside the public system (Hess, 2009). On the other hand, the economic view in education also affirms that the American educational system can be managed as a business. The central concept of this theory lies in the idea of “accountability.” People working for the government have to be accountable to it. This is why the second concept that follows from the business model is that of “standards.” People can only be accountable if they know what they are accountable for: this is the role of standards. Educational standards represent what the student is expected to learn at any given stage of education as well as how the student should progress through the education system. “The market-oriented education” is the belief that fair competition inside the American schooling system will result in a better education for all and, ultimately, a better society. Indeed, this improvement of the educa- tional system is necessary to establish fair competition among individuals. The democratic ideal that sustains this vision is the one that we usually refer to as the “American Dream” (Beach, 2007): there is natural inequality between people, but this can only be fully appreciated and assessed from the standpoint of fair competition. All individuals need some common standpoint or measure from the beginning of the competition, which ensures it is a fair competition. The idea is that a good education is something all require in a democratic meritocracy. Furthermore, the public good will be naturally enhanced from the competition of all: everyone can share the burden of any democratic inequalities. I want to focus my attention on a particular aspect of the economic worldview in education, an aspect that is logically and naturally related to “accountability” and “standards,” namely, “high-stake testing.” Testing is the tool that policy makers and politicians use to verify if educational actors (administrators of all levels, teachers and students) have attained the standards that were assigned to them. “Test scores were favoured partly because ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 22 they were the most readily available measure of school outcomes” (Carnoy, 2009, p. 31). Without testing there is no accountability, because this is the most effective way to verify that standards are reached. Moreover, to make accountability real there must be consequences attached to the people that do or do not achieve the goals given to them. Depending on whether the standards are attained or not, there can be rewards or sanctions. A reward, for example, could be that certain teachers receive monetary bonuses, or that students receive a diploma, or that administrators receive public praise in seeing their school labelled as “successful.” The sanctions are more diverse: students jeopardize their chances to graduate, schools are publicly labelled as “failing” and, in some extreme cases, even closed and restructured. High-stakes testing performs various functions for those who advocate its use. First, as representing common standards, they give a reference point to all educational actors of what has to be achieved. Second, they can give a description and evaluation of the education situation for not only policy makers and politicians, but also for teachers and administrators of a school and district. They can also provide data that can be used to inform peo- ple about a given educational situation, which allows for future actions to be taken and the potential success of these actions assessed. Third, because the testing is related to positive and negative consequences, it is supposed to motivate actors to attain the goals assigned to them (Herman & Baker, 2009, p. 179). My next goal is to show the negative consequences of the widespread policy of high-stakes testing in educa- tion, referring to some recent qualitative studies that directly tackle this issue. My assumption is that most of the problems and criticisms with high-stakes testing at K-12 education also hold for higher education in general insofar as both buy into the same market ideology. Perverse effects of high stakes testing Teaching to the test One of the collateral effects often attributed to high-stakes testing is that curriculum is narrowed, which al- lows for teachers to teach to what the test assesses, therefore, narrowing significantly the education of children. This phenomenon is usually referred to as “teaching to the test” (Herman & Baker, 2009, p. 182). Mcneil, Cop- pola, Radigan and Heilig (2008) report in their research that teaching has been significantly affected by testing. Teachers limit the curriculum to what will be assessed in the test. There is also an effect on the pedagogy used inside the classroom, increasing the use of drills to make the students learn the knowledge and skills they need to pass the test. Narrowing the curriculum affects at risk students more than those in little danger of failing the test (McSpad- den McNeil, Coppola, Radigan, & Vasquez Heilig, 2008, p. 28). The former lose their elective classes in order to focus on math and English (Sipple, Killeen, & Monk, 2004, p. 154). For the same reason, other schools may ask their at risk students to stay after school (Seachore Louis, Febey, & Schroeder, 2005, p. 193). “Teaching to the test” also narrows what it means to be educated. Students will have to memorize a large amount of knowledge without seeing the usefulness of what they learn. Learning then becomes disconnected from student lives (Rigsby & DeMulder, 2003, pp. 19-20). Lipman (2003) argues that, thanks to this process, at risk students learn obedience and powerlessness; the students are not recognized as agents of their learning but as passive actors that need to be filled with information and skills. This vision of how learning is degraded through “teaching to the test” is largely voiced by teachers. They feel that education becomes not only meaningless for the students, but also for themselves as teachers (Rigsby & DeMulder, 2003, p. 17). For most teachers, teaching involves much more than “teaching to the test,” but is a passion that has to be shared with the students. Students do not learn to read just for the sake of reading, they must learn to read texts that are meaningful (Gerstl-Pepin, 2006, p. 153). ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 23 A variety of different studies report that teachers feel high-stakes testing is attacking their professionalism, limiting their autonomy as a teacher (e.g. Seachore Louis, Febey, & Schroeder, 2005, p. 184). What is more, the problem is far more serious for those teachers who have students at risk of failing tests (Lipman, 2003, p. 337). Teachers find it difficult to teach something that they believe is important and educational but yet is not explic- itly required for test preparation (Mabry & Margolis, 2006, p. 12). The consequence is that teachers are losing their sense of agency in the educational process. Some teachers conclude that they are no longer educators or teachers but, rather, technicians (Rigsby & De- Mulder, 2003, p. 17). They do not have to think, but rather apply mechanically what others have thought. This is well expressed in the following quote from a teacher: “It makes me feel, like, then, you don’t really need trained teachers; you just need trained monkeys” (cited in Olsen & Sexton, 2009, p. 23). As is evident, such testing poli- cies clearly impact the morale of teachers, in which they feel unrecognized as professionals while being forced to do things they do not believe in. It undermines other essential dimensions of teaching as well, such as reflectivity and compassion (Gerstl-Pepin, 2006; Rigsby & DeMulder, 2003). For example, some teachers report that they are thinking about quitting the profession because of the climate created by high-stakes testing. Quite often high-stakes testing only compounds the anxiety and pressure felt by participants. Students fail the tests because they are under stress. Teachers and administrators at schools that are “high risk” usually feel more pressure. Moreover, the pressure is higher when teachers and administrates believe there is no way to reach the objectives proposed to them, especially with the limited resources often available (Mabry & Margolis, 2006, p. 155; Sipple, Killeen, & Monk, 2004). Pressure comes not only from all levels of administration, but also from the public, since schools labelled as “failing” are publicly shamed. Even schools that perform well are under constant pressure, as they want to keep their status of successful schools (Seachore Louis, Febey, & Schroeder, 2005, pp. 183, 186). Equity One of the goals of high-stakes testing is efficiency. This is seen as a tool to make sure that the goals settled by policymakers and politicians are attained. However, there is another goal behind this policy as important as that of efficiency, namely, social justice or equity. The very name ‘No Child Left Behind’ points out that goal. “The catch-phrase, ‘no child left behind,’ refers to federal efforts to remedy persistent gaps in student achievement re- sults in comparison among majority student performance and groups of minority, low-income, English language learners, and students with disabilities” (Lindle, 2009, p. 319). NCLB aims to remedy the various achievement gaps in education among students of different groups by offering an education of quality to all. Seen as such, it is the continuation of the democratic ideal where the success of an individual in school – and therefore in society – is not determined by his or her cultural and economic background. A lot of studies put in doubt the notion that the policy of high-stakes testing narrows achievement gaps. Diamond (2007) concludes that such testing is ineffective as a means to narrow achievement gaps. This is also a finding of Seashore Louis, Febey and Schroeder (2005) in which they note that in one of the three schools they studied: “Almost all teachers mentioned that poor, immigrant students’ needs were given inadequate considera- tion by state-level policymakers representing suburban and rural interests” (p. 192). The results of many studies corroborate the conclusion that high-stakes testing is not helping the communities that most need it. Other researchers are far more radical in their estimation and affirm that high-stakes testing is increasing the very achievement gaps they were supposed to remediate (Lipman, 2003; McSpadden McNeil, Coppola, Radi- gan, & Vasquez Heilig, 2008). Lipman (2003) concludes that the Latino and Black communities in low income neighbourhoods are more at risk of being victims of the “teaching for the test” style of pedagogy. A school in a white middle class community can avoid this, since they are not in danger of failing the test (p. 338). For Lipman, that situation prevents Latino and Black youths from low income communities from receiving an education that aims to teach them the higher critical thinking skills that could help them to resolve real life problems. Lipman ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 24 claims that not only is high-stakes testing not resolving the achievement gaps, but that it is actually increasing them, preventing youth from learning those very skills needed to transform their situation. McSpadden McNeil, Coppola, Radigan and Vasquez Heilig (2008) arrive at the same conclusion, bringing to light another hidden side-effect of high-stakes testing. They studied a high school in a high-poverty urban district of Texas and brought to light how this school improved their test score results by keeping at risk students from graduating into tenth grade. By keeping those students in ninth grade, they prevented them from taking the national test, which is given in tenth grade. The school could do this by using a waiver administered by the state education agency that permits schools to apply for special rules for graduation to some students. “It was a waiver from the traditional method of basing grade promotion on the number of accrued credits; a school under this waiver could base grade promotion on different criteria, such as having to pass four courses rather than gaining credits” (p. 21). The final consequence of this strategy was that a lot of students were just dropping out of school instead of repeating a year. The authors of the study argue that the decision to drop out of school is not the result of one cause, but the convergence of many, such as: students were repeating a year, students were identified as those who will underachieve in the state test, students report the experience of school as meaningless, useless and boring, and that the curriculum/pedagogy of their courses were completely shaped by the test. This strategy of the administrators and teachers of Edgeview, the school studied by McSpadden McNeil, Cop- pola, Radigan and Vasquez Heilig (2008), is directly linked to the Texan policy of high-stakes testing. The previ- ous principal of the school had been fired for lack of results, even though he had improved the test scores from 15 to 20 % (p. 20). The teachers of Edgeview had, for their part, to live with the public shame of being a failing school and facing the danger of being closed. The stakes were clear for the administrators and for the teachers of Edgeview School. Administrators and teachers saw, in the waiver, the most efficient way to improve their test scores. Moreover, they knew that other schools in the district were improving their scores through the waiver. The administrators and the teachers were morally divided. On the one hand, they knew it was bad for the students. On the other hand, they had to avoid the negative consequences, such as, losing their jobs and being labelled a failing school. “Ethically, I think it [the waiver] is wrong. But if I’m going to lose my job if my scores don’t go up, do I roll over and forget it?” (the principal of Edgeview high school, cited p. 21). By using the waiver strategy the school was able to earn the title “recognized with exemplary progress” (p. 22). “This jump [in the test scores] earned the school its exemplary progress rating on the district’s accountability matrix and another star on the marquee in the front of the school for becoming a Recognized school.” This ex- ample clearly shows that positive consequences and recognition can be won at a considerable expense, namely, the loss of many students from the schooling system. This loss did not appear in the official number of Edgeview dropouts in 2002, which was reported at zero. It is because of problems like the one outlined above that some authors affirm that the social justice discourse in education is false and misleading. Indeed, the rhetoric of NCLB is so closely focused on teachers and admin- istrators in accounting for achievement gaps that deeper social inequalities are easy to miss. Thus, the inequality that pervades many a school is never put in the larger context of the structural inequality of our society. There is an economic gap at the root of the educational achievement gap that is shadowed by the rhetoric on educational accountability. (Anderson, 2001, p. 324; Gerstl-Pepin, 2006, p. 143; Lipman, 2003, p. 344). The validity problem Various authors put in question the validity of the tests. “Validity is thus a matter of accumulating and integrating evidence that demonstrates the extent to which inferences drawn from an assessment’s results are justified for particular uses or purposes” (Herman & Baker, 2009, p. 180). The general claim against high-stakes testing is that they are not valid because they are not giving an accurate representation of student learning. There- fore, the studies reviewed do not put in question the reliability and the consistency of tests, but they put in doubt ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 25 that those tests achieve what they are made for: an accurate representation of the knowledge and the learning of a child at a precise moment. “Reliability and accuracy are necessary but not sufficient prerequisites for valid assessment” (Herman & Baker, 2009, p. 180). At the root of the critiques against the validity of high-stakes testing is the issue and tension between uni- formity and particularity. The tests are constructed on the frame of a “one-size-fits all standard” (Lipman p. 343). However, students are not all at the same place at the same moment. The nature of our education system is such that we are attempting to artificially civilize the whole spectrum of the population, including those who enter school with major disparities, to a common level. It is a race to the top of the hill that was chosen by the ruling powers to be the goal of all children. The problem is that there are unforeseen handicaps in the race. Some children have to carry heavy baggage. Some children have been coached in racing tactics. Some children were unable to read the racing instructions. Yet all children are assumed to be equally able to race (teacher quoted in Rigsby & DeMulder, 2003, p. 17) Testing is based on an ideal situation where all students would be at the same place at the moment of the test. As stated earlier, one of the goals of a democratic education is to create a situation where all the individuals could engage in a fair competition, fair because they all start at the same point. However, as this teacher tells us, high-stakes testing is not fostering such a situation; rather, it is fostering the same kind of inequalities. What was supposed to be part of the solution is in fact now part of the problem. Moreover, testing does not take into consideration the fact that all children are fundamentally different. Testing makes sense from the view of the policymakers: all students are data, they have no particularities and individualities, they all share common formal characteristics, they are all abstractly similar and interchangeable. The teacher has a completely different view. For him, students are not abstract entities, but concrete individuals with whom he is living every day. He knows the specificities of all his students and will adapt to them. In other words, his teaching and evaluation are informed by his knowledge of the differences between the students of his class. “We are teaching a child. You take out that human piece at the times when you just look at the data” (quote of a principal in Mabry & Margolis, 2006, p. 19). “They [the staff of the school] also shared their school’s ongoing struggles to improve amid the reductionism of their school districts, which tended to treat all schools, teachers, and students as the same” (Craig, 2009, p. 125). The second aspect of the critique of testing validity arises from the issue of time. Tests are temporally situated at a precise moment. They are a snap shot of the learning of the child at a particular time. Testing has therefore not the temporal vision of the teacher, a vision of the progress of the child through time. A teacher evaluates the child by taking in consideration the progress he made during the class. In Mabry and Margolis (2006), a principal affirms that he relies more on the teachers to know the state and progress of student learning than test scores (p. 25). As Cuban (1998) points out, policy makers and researchers have different criteria to judge the validity of an educational policy, and, more fundamentally, of what it means to educate someone and how we can measure such education. Finally, there is the belief across several studies that tests are not valid because they are not constructed to evaluate the learning of the child. As said in a previous section, testing can lead in some places to a narrowing of the curriculum and of teacher pedagogy – what we usually refer as “teaching to the test.” When that happens education comes to have a very limited meaning: a teacher educates when he makes the students learn the skills and the knowledge necessary for the test; a student is educated when he has the skills and the knowledge to pass the test or when he has passed the test. Here is how a teacher describes her experience of “teaching to the test”: “I presented the information to my students in a way in which I was not comfortable: read, recite, review, MEMO- RIZE. I knew the majority of my students would retain little, if any, of it. However, I got caught up in the testing mania” (quoted in Rigsby & DeMulder, 2003, p. 19, majuscules originals). In such cases, testing becomes its own goal: being educated means being able to pass the test. Education no longer means equipping the child with ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 26 skills and knowledge that are relevant for his life and that will help him to transform his reality. The teacher says in the last quote that even memorization may be of little use, because she expects that the students will forget all that they learned through drills. Moreover, this approach prevents the learning of other thinking aptitudes, such as critical thinking, that are obviously more important in today’s world. If we understand education as a more complex phenomenon than only memorization, we have to conclude that much standardized testing in its current forms is not very valuable, as it does not represent and evaluate the learning of the child. Mabry and Margolis (2006) conclude their study by saying that official test scores, even where they are posi- tive, may not give an appropriate image of the real situation of the schools. “Test data indicating that the two participating schools were meeting state and federal achievement targets suggested successful policy implemen- tation. However, the fuller picture of implementation provided by classroom observations and interviews of personnel was much less clear about whether local implementation of NCLB had been successful” (p. 26). The role of the scholar in this debate How to reflect in the era of economic logic I think it is inevitable that the assessment logic that has so dramatically affected the K-12 American edu- cational system will proceed along similar lines in the field of higher education. There may have been a time, a golden age, where the university was seen as an institution that had to be separate from other social institu- tions, as a place that dedicated itself to the search for truth, a place of higher learning free from the mercantile passions that dominate society (Bloom, 1987). If such a time really existed, it is now far beyond us. We should therefore not see our role as university scholars as disconnected from the other levels of the educational system. The question is what should be our role in this debate? I believe that the battle with the economic logic in education is ultimately ideological: between what we think are the goals of education, of what is the role of teachers, on how we can assess learning, on how and if we should help children in difficulty, in what kind of society we wish to live in, and, ultimately, what kind of de- mocracy we want. However, providing alternatives alone does not end our intellectual work. To be successful we need to use the logic of the economic worldview against itself. First, we must continue to do the kind of research that was presented in the previous sections. This means going into the field and showing the real consequences of such economic policies in education, as well as looking at examples of people who have successfully struggled against them. As shown earlier, the policy of high-stakes testing is one of the most important dimensions of the economic worldview that permeates education, and so showing the failure of this approach is one of the best ways of showing the limitations with the economic approach to education. It is now clearly documented that the policy of high-stakes testing is ineffective at aiding those students that the policy was originally designed to help. Indeed, high stakes testing is not narrowing the achievement gap but increasing it. It is the students of low income communities or from specific minority groups that are the biggest targets of “teaching for the test,” and who end up with not only lower test scores, but also impoverished curriculums that do not prepare them for any real life problems. We have to advance the notion that high stakes testing is not, in its current use, improving the quality of education. Such an approach to testing is not valid, because it does not give a valid measure of children’s learn- ing. The meaning of “being educated” ends up terribly impoverished under these assessment methods. The ultimate goal of high-stakes testing was to improve schooling. We can show in various ways that it is working against this goal: learning cannot mean memorizing facts, and teaching cannot be defined only by the use of drills to prepare pupils for tests. Policy makers and politicians should be sensitive to such arguments, as it makes good use of the vernacular they are most comfortable with, that of numerical data. ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 27 How to improve high stakes testing My position is not to reject high stakes testing outright. We probably all share, to some degree, aspects of its ideology. However, the way that such testing is used at the present is doing more harm than good. Consequently, I now want to focus on how I think we can best improve the problem, and do so from within the framework of the economic worldview. The ultimate principle of the reigning economic logic is one of efficiency. As shown in the previous sections, high-stakes testing is not efficient: it is not achieving the goals for which it was established and in many cases even undermining these goals. This is an argument that people who promote current assess- ment policies should be sensitive to. We then have to propose a new form of testing, which demands staying inside the economic logic of education while avoiding its pifalls and limitations. I have focused most of my efforts so far on the negative effects of high-stakes testing. However, there are also successful stories of such testing (e.g. Gerstl-Pepin, 2006; Rigsby & DeMulder, 2003; Seachore Louis, Febey, & Schroeder, 2005). What most of these success stories have in common is that the policy was implemented in a democratic way. There is a clear correlation between the positive perception of the educational actors and the level of democracy they experience. There is also a clear correlation between the positive effects of testing and the democratic experience of the different actors. The problems that are reported by many people who have been involved with high-stakes testing appear to be the result of a top down policy that is rigidly applied. To be sure, we have to be careful in our judgement of whether top down policies are democratic or not. Top down designs are undemocratic according to a particular vision of democracy, yet they also retain important elements of democracy. For example, educational policies are applied by policymakers who themselves have received orders from politicians democratically elected. It is legitimate for the politician to ask for reforms since he receives his legitimacy from the community through a democratic process. It is through the politician that the community can act. I agree that this dimension of de- mocracy is important, but it remains too limited as a model of effective policy implementation. I want to advance another vision of what it means to democratically implement an educational policy and standard, one that is informed by those studies that report a successful implementation of high-stakes testing. This model should not be seen in direct opposition to the hierarchical model mentioned above. I think both approaches can be seen as mutually complementary and informative of what democratic processes entail. I therefore do not promote an either/or approach to this topic. From a minimalist standpoint, democracy occurs when individuals are given some power to discuss the issues in which they are concerned. This appears at the level of high-stakes testing when a certain degree of autonomy is granted teachers, schools and districts when it comes to the implementation of educational policy. However, a fuller picture of democracy occurs when the dif- ferent actors are involved communally not only in the process of implementation but also in the decisions about standards and high-stakes testing in general. This means, for example, that the administration and teachers are collectively engaged in how they will adapt testing reform to meet their particular situation. When top down designs are implemented rigidly participatory democracy disappears, which leads to ineffec- tive and dangerous results. In those cases that interest us, this happens when teachers see themselves as techni- cians who are pressured to teach to the test. The results of this approach are ineffective, since it pits teachers against the very reform they need to implement. In Cuban’s (1998) analysis, there cannot be effective reform without adaptability. A top down design of high-stakes testing is dangerous because it leads to the creation of an education schedule of drills and memorization, wherein education is rendered largely useless for the students. We have also seen that this kind of situation is more likely to happen for those students already in difficulty or from low income communities. Rigidity in top down design leads to a loss of democracy inside the schools. It leads to what Olsen and Sexton (2009) call “threat rigidity”: “threat rigidity is the theory that an organization, when perceiving itself under siege (i.e. threatened or in crisis), responds in identifiable ways: structures tighten; centralized control increases; con- ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 28 formity is stressed; accountability and efficiency measures are emphasized; and alternative or innovative thinking is discouraged” (p. 15). As said earlier, I do not want to fall into an either/or position by saying that top down designs are necessarily bad and anti-democratic. However, it does seem clear that when top down designs are used rigidly they become undemocratic. Top down designs are democratic when they allow autonomy and local decisions; in others words, where decisions can still be experienced by participants as flowing from the bottom up and not just from the top. It is precisely when high-stakes testing is implemented with an eye to moderating both sides that it will be the most efficient and, unsurprisingly, the most democratic. In that sense, I think we should not separate democracy from efficiency, for both clearly go together on this matter. Again, the language and measures of efficiency are es- sential, because it gives us a common language from which to engage, and persuade, those within the economic paradigm of education. I think the best way to confront the prevailing ideology of high-stakes testing is to accept the criterion of ef- ficiency in order to clearly show why current policies are failing to reach their goals. That said, I hope it remains clear that in taking this approach one is still moving the debate to another level. Indeed, what we mean by de- mocracy is fundamentally an ideological issue, as is our vision of the teacher’s role in cultivating a democratic education, and how political policy should be implemented. In using economic ideology against itself we are at the same time opening up the space for another kind of conceptualization. However, my hypothesis is that, in order for this political strategy to work, we need to be seen as ‘playing the game’ rather than confronting the issue directly as a battle of different ideologies of education. The issue of whether these two types of democracies, top down and grass roots, can be united is one of the most important and complex issues of policy and politics in the contemporary world. It is easy to say that top down design must encounter bottom up movement. However, it is impossible to clearly predict how that can happen concretely. Indeed, how far can a policy design be flexible to local situations? Which points of a policy must be strictly adhered to, and which changed? There is no ultimate answer to these questions. This is an issue that flows out of, and follows, the politics and policy-making of a democracy. Fundamental to the nature of our democracies, is the union of local democracy with national democracy. If nothing else, it is our role as educators to demonstrate the complexity of democracy in action and challenge the simplistic view of participatory democ- racy that inevtiably arises once quantitative methodology, and the positivism that underwrites it, are seen as the order of the day. Moreover, although the complexity of real democracy in action seems impossible to capture, the studies I referred to earlier show how different actors have been able to integrate, in a positive way, the policies of high-stakes testing. The underlying issue In saying we should not confront the economic ideology directly, however, does not mean we should aban- don the ideological discussion altogether. To be sure, we need to engage in the political discussion in order to show what is at stake in the different ideologies of education. I think this is nicely illustrated in the debate be- tween those who acknowledge the complexity of reality, and those who continually affirm its simplicity. Accord- ing to Cochran-Smith (2005), the complexity/simplicity dichotomy is at the root of two worldviews in education. The economic view in education is based on the conception that reality is simple. “Stone suggested that from a market model of policy making, the definition of policy problems is mistakenly regarded as a simple and straight- forward matter of ‘observation and arithmetic’ ” (cited p. 133). Reality is unproblematic in this view, it is easy to judge situations and act to change them. The following statement confirms this worldview: “In simplest terms, if the objective is to improve student performance, student performance should be the focal point of policy… I use a simple definition of teacher quality: good teachers are ones who get large gains in student achievement for their classes; bad teachers are just the opposite” (Hanushek cited in Cochran-Smith, 2005, pp. 183-184). ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 29 The present use of high-stakes testing is a good example of the simplicity of the economic worldview at work. It is based on the belief that student learning and knowledge is easily measurable, which ends up dissolving the complexity of school situations, as well as student lives, in order to emphasize their uniformity and unity. “One of the most basic assumptions embedded in technical standards is a naïve realism that assumes the world is a simple system made up of entities capable of precise empirical descriptions” (Jenlink & Austin, 2004, p. 4). In general, testing risks giving a simple vision of student reality by assuming it is easy to separate the loosers from the winners (Cuban, 1998, p. 456-7). The alternate worldview, as discussed by Cochran-Smith (2005), is one that embraces the complexity of real- ity. Education is something hard to define, there are diverse conceptions of what a good education is, and we cannot subsume these views in one idea. The reality of students and schools is diverse and complex, and policy must be flexible enough to respect this. Therefore, testing alone, in virtue of its simplicity, is unable to give an accurate vision of educational reality, and so also remains unable to effectivley transform it. In the current situation, where the economic view dominates the landscape of policy research, it is our duty as scholars to present the weaknesses of this paradigm as well as present viable alternatives. In doing so, we have to be aware of how the dominant discourse of the day shapes our understanding of reality (Stovall, 2009; Torres & Van Heertum, 2009). It is only by entering in a debate with the dominant worldview that we can offer a more complex and authentic view of reality. That can only happen if we engage in thinking critically about our practice at all levels: our epistemology of education, our understanding of what democracy means, our role as scholars, our teaching practices, and finally our research methodology. Engaging in such a multilevel process helps ensure two things: first, we do not separate ourselves from the other struggles taking place at different levels of educa- tion, second, we do the best we can do in our specific role as scholars in higher education. Endnotes I want to thank Marta Pires, Tyson Lewis and the two reviewers of 1. Analytic Teaching for their careful read- ing and precious comments on the first version of this manuscript. References Anderson, G. L. (2001). Promoting Educational Equity in a Period of Growing Social Inequity the Silent Con- tradicitons of Texas Reform Discourse. Education and Urban Society, 33(3), 320-332. Beach, J. M. (2007). The Ideology of the American Dream: Two Competing Philosophies in Education, 1776- 2006. Educational Studies, 41(2), 148-164. Belfield, C. R., & Levin, H. M. (2009). Market Reforms in Education. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 513-527). New York: Routledge. Bloom, A. (1987). The Closing of the American Mind. New York: Simon & Shulster, inc. Carnoy, M. (2009). Policy Research in Education: The Economic View. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 27-38). New York: Routledge. Cochran-Smith, M. (2005). The politics of teacher education and the curse of complexity. Journal of Teacher Edu- cation, 56(3), 181-185. Craig, C. J. (2009). Research on the Boundaries: Narrative Inquiry in the Midst of Organized School Reform. The Journal of Educational Research, 103(2), 123-136. Cuban, L. (1998). How Schools Change Reform: Redifining Reform Success and Failure. Teacher College Record, 99(3), 453-477. Diamond, J. B. (2007). Where the Rubber Meets the Road: Rethinking Connection Between High-Stakes Test- ing Policy and Classroom Instruction Sociology of Education, 80(4), 285-313. Gerstl-Pepin. (2006). The paradox of Poverty Narratives: Educators Struggling With Children Left Behind. Edu- cational Policy, 20(1), 143-162. Hanushek, E. A. (2009). The Economic Value of Education and Cognitive Skills. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 39-56). New York: Routledge. Herman, J. L., & Baker, E. L. (2009). Assessment Policy: Making Sense of the Babel. In G. Sykes, B. Schneider ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS Vol. 30 No.1 30 & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 176-190). New York: Routledge. Hess, F. M. (2009). A Market of Knowledge? In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Educa- tion Policy Research (pp. 502-512). New York: Routledge. Hursh, D. (2001). Neoliberalism and the Control of Teachers, Students, and Learning: The Rise of Standards, Standardization, and Accountability. Cultural Logic, 4(1), 1-9. Jenlink, P. M., & Austin, S. F. (2004). Ensuring Democracy in an Era of Standards and Accountability: The Scholar-Practitioner’s Work. Scholar-Practitioner Quarterly, 2(4), 3-12. Lindle, J. C. (2009). Assesment Policy and Politics of Information. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 319-332). New York: Routledge. Lipman, P. (2003). Chicago School Policy, regulating Black and Latino youth in the global city. Race Ethnicity and Education, 6(4), 331-355. Mabry, L., & Margolis, J. (2006). NCLB: Local implementation and Impact in Southwest Washington State. Educational Policy Analysis Archives, 14(23), 1-35. McSpadden McNeil, L., Coppola, E., Radigan, J., & Vasquez Heilig, J. (2008). Avoidable Losses: High-Stakes Accountablity and the Dropout Crisis. Education Policy Analysis Archives, 16(3), 1-45. Olsen, B., & Sexton, D. (2009). Threat Rigidity, School Reform, and How Teachers View Their Work Inside Current Education Policy Context. American Educational Research Journal, 46(9), 9-44. Rigsby, L. C., & DeMulder, E. K. (2003). Teachers Voices Interpreting Standards: Compromising Teachers Au- tonomy or Raising Expectations and Performances. Education Policy Analysis Archives, 11(44), 1-33. Seachore Louis, K., Febey, K., & Schroeder, R. (2005). State-Mandated Accountability in High Schools: Teach- ers’ Interpretations of a New Era. Educational Evaluation and Policy Analysis, 27(2), 177-204. Seashore Louis, K., Febey, K., & Schroeder, R. (2005). State-Mandated Accountability in High Schools: Teach- ers’ Interpretations of a New Era. Educational Evaluation and Policy Analysis, 27(2), 177-204. Shoen, L., & Fusarelli, L. D. (2008). Innovation, NCLB, and the Fear Factor. Educational Policy, 22(1), 181-203. Sipple, J. W., Killeen, K., & Monk, D. H. (2004). Adoption and Adaptation: School District Responses to State Imposed Learning and Graduation Requirement. Educational Evaluation and Policy Analysis, 26(2), 143-168. Stovall, D. O. (2009). Race(ing), Class(ing), and Gender(ing) Our Work. Critical Race Theory, Critical Femi- nism, Epistemology, and New Directions in Educational Policy Research. In G. Sykes, B. Sneider & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 258-266). New York Routledge. Sykes, G., Barbara, S., & Plank, D. N. (Eds.). (2009). Handbook of Education Policy Research. New York: Routledge. Torres, C. A., & Van Heertum, R. (2009). Education and Domination: Reforming Policy and Practice through Critical Theory. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Hanbook of Education Policy Research (pp. 221-239). New York: Routledge. Vergari, S. (2009). Charter School Policy Issues and Research Questions. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 478-490). New York: Routledge. West, M. (2009). Public Choice and the Political Economy of American Education. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Education Policy Research (pp. 362-371 ). New York: Routledge. Witte, J. F. (2009). Vouchers. In G. Sykes, B. Schneider & D. N. Plank (Eds.), Handbook of Education Policy Re- search (pp. 491-501). New York: Routledge. Address coresspondence to: Olivier Michaud michaudo1@mail.montclair.edu