Kauffman, J. M., & Konold , T. R. (2007).

Making Sense in Education: Pretense (Including No Child Left Behind) and Realities in Rhetoric and Policy About Schools and Schooling. Exceptionality, 15(2), 75-96. doi:10.1080/09362830701294151.

One wrong assumption about reality is sufficient to ensure the failure of education policy. For a variety of reasons, too many people have ignored realities in policies such as the No Child Left Behind Act. Individuals who are serious about reforming or improving education, general or special, must confront realities in at least the following areas: measurement, statistical distributions, measurement error, labels, special services, teacher responsibilities, and teacher training. Enacting or supporting education policy that is known to be seriously flawed is unacceptable and tantamount to the criminal conduct of someone who markets goods with known safety defects or harmful effects or of someone who ignores critical realities in the conduct of business. Life is full of absurdities. Rhetoric about education should not be added to them. When someone suggests that all children will be able to perform at ____ (> 0) level, that all children will succeed, or that no child will be left behind, he or she is contributing to needless and unhelpful silly talk about schools and schooling. Some educators, as well as people in other lines of work, may argue, for example, that the No Child Left Behind Act (NCLB) at least focuses attention on education and motivates educators to work harder to make sure that all students achieve all that they can. They are misguided about NCLB, as it is a prime example of absurd education policy. Under NCLB (actually a reauthorization of the Elementary and Secondary Education Act, passed by Congress in 2001), schools are required to demonstrate adequate yearly progress (AYP). Although measuring and defining levels of proficiency that define AYP is determined at the state level, states are required to increase their proficiency standards gradually over time in order to ensure that all students (i.e., 100%) are functioning at the level of proficiency within 12 years (i.e., by 2014, as the bill was signed into law by President George W. Bush in 2002). NCLB legislation stipulates that progress be measured in the areas of reading, mathematics, and science. Currently, children must be tested in reading and mathematics each year from Grade 3 through Grade 8, and additionally at least once between Grades 10 and 12. Science assessments are mandated to be in place by 2007, at which time students will be assessed three times. These assessments are required to take place sometime between Grades 3 through 5, 6 through 9, and 10 through 12. Regardless of the unique challenges that may present themselves within certain school districts, schools failing to make AYP for two consecutive years must use at least 5% of their Title I funds to provide students the opportunity to attend another public school, inclusive of transportation. The continual failure to meet AYP standards places an even greater burden on schools through sanctions of increasing severity. Test scores for various subgroups (e.g., ethnic groups, students with disabilities) must be disaggregated, and the failure of any group to meet AYP is sufficient to define a school as failing. Allowances have been made for a small percentage of students with the most severe disabilities, but the vast majority of students with disabilities served under the Individuals With Disabilities Education Act are expected to perform, as a group, similarly to those without identified disabilities. Although a full description of the voluminous documentation and

requirements of the NCLB legislation is beyond the scope of this article, interested readers are encouraged to consult the law itself for additional details (the law can be found at http://www.ed.gov/nclb). Policies like NCLB, though conceptually flawed, may produce some benefits for some students, and perhaps even for many. In fact, just about any policy enacted with good intentions, regardless of how conceptually vacuous, will benefit some students. However, policies like NCLBbased on irrational premises and that ignore important realitiesnot only are doomed to collapse but in the longer term hurt the cause of education more than they help (Kauffman, 2005b). We are not the only ones to note the irrationality of this law, its blithe avoidance of realities (e.g., Rothstein, Jacobsen, & Wilder, 2006). However, NCLB is not the only silly education policy. It is only a convenient and popular example. Senseless statements and policies are common in education but incompatible with its goals. Especially in the case of education, which at its most fundamental level is helping people make sense of things, senseless rhetoric runs against the grain of what one is trying to accomplish (Kauffman, 2002, p. 284). Senseless rhetoric describes much of what has been written and said about education, not only by educators but by captains of industry, politicians, newspaper columnists, and others. Oxymoron may be an accurate description of major provisions of NCLB (Rothstein et al., 2006). Rhetoric may safely be judged senseless when it ignores any reality. Most people ignore relatively few realities because they know that by doing so they are courting disaster. In fact, when someone ignores realities, most observers conclude that he or she is mentally ill or at least is engaging in very dangerous behavior. Adults try to teach progeny and students about realities so that children do not harm themselves or others. If children ignore realities, adults apply restrictions and warn them about the dangers of assuming that realty does not affect them. Adults want their children and students to talk and act as if they were in a real world in which some things are universally, unalterably true, whether anyone likes it or not. For example, adults want them to think and act as if gravity and inertia and other similar principles of physics apply to them. Adults want them not to play with fire based on an understanding of realities. Adults want them to understand that having casual, unprotected sex is risky because they know what causes HIV/AIDS and how it is transmitted. Adults want them to understand that killing someone has dire social consequencesdire for the person killed, that person's loved ones, the community, and the person doing the killing. Too often in thinking and talking about education, educators and others seem to assume that realities are distractions, unimportant or avoidable in their proposals to reshape, restructure, or improve general and special education. Their assumption may be based on one or more false premise. Sometimes it is based on the philosophical proposition that what one consider realities are socially constructed by one's culture, and therefore not universal. Sometimes it reflects ignorance, misunderstanding, or presumed political finesse. Sometimes it is expressed as a desire to think outside the box, as if the box of realities were an artificial constraint that needlessly inhibits creative thinking. It may sometimes reflect extreme frustration in trying to change education policy, such that an unrealistic policy may be supported as the best one can hope forat least a beginning of change in the right direction. Or it may reflect the assumption that moral decisions can be made in the absence of reliable evidence (see Brantlinger, 2006;

D. J. Gallagher, 2004; Gartner & Lipsky, 1989, for examples of evading or denying realities while calling for reform). But regardless of why anyone thinks and talks as if realities do not exist in education, this gambit has adverse consequences. Also too often, any statement about education or any call for reform, regardless of its failure to conform to the real world, goes unchallenged. Some seem to assume that every statement or suggestion about education should be taken seriously. The consequences of this assumption are dire: Actual improvement of education is stymied, whereas possibilities that are only imaginary are pursued. Long ago, people learned that some things (e.g., turning lead to gold) are flatly impossible. For many things, proof is demanded in the scientific tradition before someone is allowed to market an idea or product. For example, an antigravity device, cold fusion, an antiaging substance, perpetual motion, and so on are not assumed to be realities until skeptics are satisfied by demonstrations that pass scientific muster. Nevertheless, as a society Americans have not only allowed but supported and sometimes even enshrined in law preposterous propositions about education. Silly talkpretensehas too often been allowed to share a place at the table with serious talk about education. This produces babble, not constructive conversation. In many ways, education is rocket sciencecomplex, difficult to understand, and disastrous when one seemingly small detail is forgotten or one fundamental reality is overlooked. Space shuttle disasters occur because little realities are neglected. Getting things mostly right about space flight but wrong in one critical assumption is an invitation to disaster. Getting things mostly right in education is not good enough either. In education, experts too often forget or ignore realities that may be considered inconvenient but that bring about an inevitable policy failure when they are assumed to be trivial. Substantive improvement of education is essential. However, such improvement will not occurindeed, cannot occuruntil thinking and talking about education are judged legitimate only if they contain no pretense, no pretending that a reality does not exist or can safely be ignored. Education as an enterprise is supposedly designed to help students separate fact from fiction about the world and deal productively with the confusing world in which they live. Teaching based on the premise that fact cannot be distinguished from fiction will always be derelict. So will talk of education that does not frankly acknowledge realitiesand reject as frivolous the fictions to which people cling implicitly when their thoughts and actions reflect ignorance of or disregard for realities. Most people understand how projects in the physical or social sciences are doomed ultimately to failure by ignorance of or disregard for realities. For example, no one is allowed to construct and sell airplanes if he or she has ignored any reality of gravity or aerodynamics in the process. No view of criminality that ignores the necessity of negative consequences for criminal behavior is tolerated. Talk of education that does not recognize important realities should not be accepted, simply because any proposal for reform, legislation, or regulation that does not start from the premise that realities must be incorporated into the plan is doomed to failure. Reality-evading proposals may be appealing to the masses, may be sold to the unsuspecting, and may seem initially to be successful. Ultimately, they crash and burn, as does any reality-evading proposal in the natural sciences. NCLB is a case in point of education policy doomed to failure because it ignores realities.

NCLB legislation mandates that schools failing to make AYP standards for the first time must provide students the opportunity to attend a better public school provide transportation, and use at least 5% of its Title I funds for this purpose (NCLB, 2001). Failure to meet AYP standards in subsequent years is met with sanctions of increasing severity. How, then, are these schools expected to succeed and meet the needs of their students? The threatened removal of resources from schools that fail to meet AYP goals is both counterintuitive and contrary to the way in which things work. Athletes do not stop training for months prior to a race, farmers do not stop feeding their livestock prior to auction, and schools cannot be expected to perform better through the imposed removal of educational resources after assessments of student learning have shown their failure to meet expectations. These schools and the students that compose them are placed in even greater danger to fail, thereby contributing to a widening of the achievement gap, rather than a narrowing of it. Moreover, the successful schools to which students are transferred from failing schools will inevitably encounter the very same statistical realities and will, predictably, fail in future years. The very policy itself is internally destructive to the goals it has laid out. This is because schools that fail to meet annual AYP goals are in even greater danger of being left behindthe very problem NCLB is intended to defeat. In fact, many have suggested that NCLB promotes segregation to the extent that was realized prior to Brown vs. Board of Education (see Associated Press, 2006; Lawrence, 2006), have commented on this and other moral concerns that NCLB raises (National Council of Churches, 2006), or have noted the negative and unrealistic statistical effects of the law (Basken, 2006). Even Marian Wright Edelman, whose Children's Defense Fund first used the slogan Leave No Child Behind (see Kauffman, 2002, pp. 2224), has called for President George W. Bush to come clean on the intellectual dishonesty of the way the slogan has been used in NCLB (Edelman, 2006). Rhetorical excess in the heat of a political campaign, such as the promise to put two chickens in every pot, can perhaps be understood and may even be forgiven. Unforgivable, howeverat least to those who favor reality over fantasyis the translation of such rhetorical excess into policy, such that by law one must examine pots regularly and those found to have fewer than two chickens in them must be punctured until such time as they are found to contain two or more chickens. Supposing that the rhetoric were somewhat less excessive, simply a chicken in every pot, and the law required puncturing any pot without a chicken until it contained a chicken, the rhetoric-to-policy remains an example of an absurdity. But our examples are apropos to the absurdity of NCLB. Note this, too: The rhetoric in our examples is absurd; the policy is not only absurd but reprehensible. Rhetoric that distorts reality is merely silly; policy based on such rhetoric is destructive of the very idea the rhetoric is thought to advance. Ignoring realities is implicit or explicit in many discussions of education and its improvement. These realities include but are not limited to those mentioned in following paragraphs. Critics of what we suggest may advance the view that these realities either do not exist or are not universal, or they may argue that these realities may be ignored for a variety of reasonsreligious, philosophical, political, or some combination of these. We disagree. Rhetoric ignoring any reality is ill advised; policy ignoring any reality is even worse. REALITIES THAT MUST BE ACKNOWLEDGED

Jump to section REALITIES THAT MUST BE ACKNOWLEDGED IMPLICATIONS FOR POLICYMAKERS AND... CONCLUSION Education, like medicine or engineering, is at least in part an applied science. True also, as is the case in medicine or engineering (or any other applied science), it can be practiced clumsily and artlessly or artfully and with exquisite skill. To deny that education includes applied science is to deprive it of a foundation of common (i.e., universal) knowledge and to classify it as a purely idiosyncratic endeavor or craft. Although teaching may have its highly individualistic features, it is also based onand only evaluated legitimately bymethods that have a scientific base. The call for scientifically validated methods of teaching is not misguided. However, one cannot reasonably call for scientifically validated methods of instruction and then ignore realities on which science is based. The fact that educators, politicians, or others sometimes have embraced the notion of scientifically validated instruction while ignoring realities, including scientific principles, merely illustrates the way in which discussions of education and proposals for its reform can go awry. Measurement Is Essential Measurement is essential to any science (see Gould, 1996; Ruscio, 2002; Shermer, 1997, 2001). Without measurement, no scientific venture can be seen as legitimate, much less succeed. Educators must have measurement to know how their students are doing. Otherwise, they have no way of judging whether teaching is having any effects on their students. Teaching without wanting to know how students are doing as a result of the teaching is what Sasso (2001) called willful ignorance. Some may oppose measurement, including classification and ranking, yet deny that accountability implies measurement (e.g., Obviously, I am not eschewing accountability for what we do. I am breaking the particular equation of accountability = measuring/ranking, Heshusius, 2004b, p. 295). But people can write explicitly self-contradictory statements, make a claim and then refute it in the next sentence, make explicit statements that are the opposite of what is implied in another of their statements, and write or speak gibberish that others of similar bent defend as understandable or even describe as particularly insightful (see Kauffman, 2002; Kauffman & Sasso, 2006a, 2006b; Mostert, Kavale, & Kauffman, in press). Nevertheless, opposition to measurement is opposition to accountability. Measurement in Education Requires Testing Testing can be accomplished in a variety of ways, but any type of test is a way of checking to see whether students have acquired particular skills. Tests may be essays, performances, portfolios, or those types most often criticized: multiple item, more objectively scored tests that may be true/false statements, fill-ins, or multiple choice. Tests are sometimes misused or abused. More often, they are simply misunderstood. The most dangerous misunderstanding is that testing can be avoided. Those who

criticize NCLB because it requires standardized testing are on the wrong track. NCLB is misguided policy, but not because it requires testing. All types of tests have advantages and disadvantages. None is perfect, but some are very good at giving educators important information. Of all tests, experts must inquire about what is being assessed, how the knowledge or skill of interest is being measured, who composes the comparison group, and what the obtained score means. Tests may be criticized on any of these grounds: that they measure the wrong thing or something trivial, that they are of questionable reliability or validity, that they result in inappropriate or invidious comparisons, or that the obtained score is meaningless or misleading. Standardized tests have been the objects of scorn primarily because critics do not understand what they are designed to do and why they are important (see Kauffman, 2002). Aside from education, most people understand the importance of testing. Most believe it is important that products be tested before they are marketed. Most desire services from professionals who have passed tests of their knowledge and skill. Testing is quality control; quality control without testing is impossible. Quality control in education is needed, but children are not products that can be manufactured to specified tolerances. Other aspects of the measurement of human beings preclude treating children as if they can be discarded if they do not measure up to a given criterion, not the least of which is the reality that measurement of human beings always results in what Stephen J. Gould (1996) referred to as a full housea full distribution (in the present discussion, of test scores or other performances), any part of which is ignored only at the cost of denying reality. Distributions may have walls beyond which a score is impossible, and a wall may set the upper or the lower limit of possibility. The one wall known about in human performance related to education is at the lower limitzero. Measurement Is Part of Identifying Failure and Success The words failure and success are nonsensical in the absence of measurement. Failure and success are always defined by a student's performing or not performing some expected task or by a student's reaching or not reaching a given criterion. To suppose that failure is impossible or that a student will be judged successful regardless of performance is to deny the meaning of the terms. Certainly, saying that all fish are large is nonsensical; saying that all those that do not pass through a net are large may make sense. Likewise, to say that all children are successful (or successful at a certain task) is nonsensical, although all who can demonstrate a certain level of proficiency in a particular activity may be judged successful. Herein lies a problem that cannot be solved by ignoring its reality: The measurement (i.e., value) constituting the criterion for judging success and failure is arbitrary (see Kauffman & Hallahan, 2005). One legacy of bad thinking about education is the assumption that failure at one thing means failure in general. The solution to this problem is not to try to pretend that failure can be avoided for all students but to recognize that failure by a given measure signifies only failure on what was measured. True, educators make failure by their measurement, and they cannot do otherwise unless they are to draw no lines or set no expectations for accomplishment for any purpose or for any age. Any measurement

that produces only successes is bogusit does not reflect the real world. Nevertheless, failure by one criterion does not imply failure by every criterion. Measurement Is Necessary for Accountability Regardless of whether the measurement is a subjective judgment of whether an individual has reached a stated level of performance or is a more objective judgment of whether the individual has obtained a certain score on a standardized test, measurement is required if educators are going to be held accountable. In fact, accountability in the absence of measurement is a non sequiturit does not follow, and indeed cannot follow, that accountability has been demonstrated in the absence of measurement of either student performance or teacher behavior. Those who call for accountability and moral judgment but condemn testing and measurement (e.g., Heshusius, 2004a, 2004b) are not describing something that comports with the actual world of accountabilityexcept that a person may be empowered to judge that someone has or has not been accountable or successful by undisclosed criteria (which is a tactic of misanthropes). Measurement Allows Educators to Identify and Work on Gaps Most educators and observers of education are aware of gaps in achievement, knowledge, or performance (regardless what term one uses to describe evidence of learning) among individuals' abilities and among various groups (which may be categories defined by ethnicity, gender, age, class, or disability). Educators may wish to widen the gap between those who are gifted and those who are not, but most gaps between groups are those they want to narrow or close. Without measurement, it is impossible even to make the case that gaps exist, and it is impossible also to know whether gaps are widening or narrowing. Thus, any suggestion that achievement gaps, disability, or diversity can be identified without measurement (e.g., Heshusius, 2004a, 2004b) is nonsense, as is denial that measurement always reveals differences that may be interpreted as gaps. Some gaps may be impossible to close by any legitimate means, but others can and should be narrowed or closed altogether. One of the problems with NCLB is that it ignores realities about closing gaps (Kauffman, 2005b). It does not seem to recognize that (a) there is a mean of means (i.e., that means, as well as scores for individuals, have a distribution), (b) some of the variance in achievement scores is due to factors other than teaching, and (c) the mathematical properties of continuous distributions apply in all cases. The failure of NCLB to recognize realities regarding distributions and gaps in achievement scores is a fatal flawas certain to result in disastrous consequences as the failure of the National Aeronautics and Space Administration to take into consideration the effects of cold temperatures on booster rockets' Orings and the damage done by foam insulation striking surfaces at high velocity. Problems with O-rings and foam insulation were not known to be fatal when the space shuttle was designed. Now experts know, and so now it would be highly irresponsible, and perhaps criminal, not to take those realities into account. Individual Differences Are Real

Supporters of NCLB argue that schools have not been successful in educating all students. In fact, the rationale for NCLB is that schools don't work or are failures (see Popham, 2004, for descriptions of how NCLB defines failing schools; criticism of public schools as failures and calls for reform so that schools work have a long history, as Cuban, 1990, recounts). Saying that schools are failing begs the question of what people should expect to see if schools or any of the programs offered therein are working (Kauffman, 1990). Would children's test score distributions be different? Would measures of central tendency (e.g., mean, median) of these benchmarks be higher? Would all students obtain the same score on these measures with no tolerance for any variation in expected student outcomes? Is it really possible to both raise the central tendency of test scores and at the same time reduce the variation in scores? These questions are central to resolving whether the goals of NCLB are obtainable even if public schools are really good. Individual differences are a reality. People differ on so many factors (e.g., interests, attitudes, motivation) that it is difficult to identify invariant characteristics. Student achievement is no exception. To illustrate, standardized achievement tests are typically normed so that the mean of the raw score distribution is set at 100 and the standard deviation is 15. For decades, the resulting scores have been shown to be approximately normally distributed when tests are administered to large populations of students. The reality of such distributions is that 50% of the students measured will have scores below the mean, and 50% will have scores above the mean. In fact, 68% of the tested population will have scores that range in value from 85 to 115, or a 30-point standard score difference; and 95% of the tested population will have scores that range in value from 70 to 130, or a 60-point standard score difference. Children show individual achievement differences, and these differences were pervasive even before experts started measuring them. Researchers have been relatively successful in identifying factors that have a positive effect on children's knowledge. However, there is also much research to suggest that different student ability levels require different instructional methods to provide students the best opportunity to reach their potential. These ideas are in contrast with NCLB initiatives that assume a one-size-fits-all model (Lawrence, 2006). Some students are simply better equipped to accumulate knowledge at a faster rate than are others, and the uniform application of programs across students of different ability levels does little to help groups on either side of the achievement continuumthough it may serve to reduce the existing variability in student achievement. Because of the punitive consequences of NCLB for schools that do not measure up, administrators feel threatened by the looming prospect of losing resources when struggling students fail to reach proficiency. As a result, greater emphasis is placed on those struggling students who are relatively close to but below the expected standard, and fewer resources are made available for gifted education (Winerip, 2006) or for the education of those with cognitive limitations who obviously will not meet the standard. In the end, this may well result in distributions of achievements that show smaller differences between children at opposite ends of the achievement scale. However, these are not the gaps that educators wish to narrow. This unfortunate consequence will also contribute to leaving behind (i.e., far below their potential) those students that hold the greatest academic potential to be successful in achieving domestic and international advances. Most Educational Variables Are Continuous in Nature

Some educational measurement consists only of categories (e.g., male/female, did/did not meet a performance criterion) or ranks (e.g., first in class, second to last to finish). However, most measurements of educational performance (e.g., most standardized tests of achievement) produce a score distribution that lies atop a continuous distribution of outcomes. As Kauffman (2002) and Kauffman and Hallahan (2005) pointed out, continuous distributions are those in which what is measured varies from a little to a lot with fine gradations or increments being possible. There are no natural, inherent, or obvious breaks in a continuous distribution. Height and weight are examples of continuous distributions, as are rate and speed. Whenever human performance (as well as many other physical attributes of humans and many things in the material world) is measured, the results of measurement produce a discrete distribution. This occurs because, regardless of the degree of precision inherent in the measurement tool (tenths, hundredths, thousandths), there will be a break in the scale when moving from one of these measured points to another. Often, however, the data resemble what has come to be known as a normal distribution, in that a graph of the scores approximates the symmetry of a bell by modeling the underlying continuous scale that is inherent to the variable being measured (hence, bell-shaped curve or bell curve). Not all distributions or curves are normal or symmetrical; some are lopsided or skewed. Nevertheless, all continuous distributions have immutable properties. The immutable properties of continuous distributions are referred to by psychometricians as moments, of which there are four: central tendency, variability, skewness, and kurtosis. These are well-established realities in spite of attempts to cast them as mere human inventions (see D. J. Gallagher, 2006) or to ignore their implications for closing gaps in performance (see Kauffman, 2005b). These distributions and immutable qualities apply to all groups, including all groups of averages for schools, districts, states, nations, and subgroups of students. They are as real and as useful as many other human inventions, including houses, arithmetic, roads, laws, and languages. Moreover, criteria that may be used to categorize individuals (e.g., did or did not meet a criterion for reading abilityin short, all criterionreferenced tests) are actually derived from the measurement of continuous distributions of performance; the criterion is based on a continuous distribution. Philosophical objections and observations that the smoothed curve depicting a continuous distribution is actually drawn from a histogram (a bar graph) or that the first such curve was drawn from estimates of measurement error (D. J. Gallagher, 2006) do not mean that continuous distributions are not real. To assume that they can be ignored, changed, or violated at will because they are educational or psychological is to invite disaster. The only way to consider no student or group (or no school, district, or state) to be behind or low performing is to make questionable comparisons or comparisons that at least partially obscure the truth for the unsuspecting observer. For example, one can compare a given score or average to a distribution obtained long ago. Although it may be the case that the United States could have education so good that fewer than 25% of students score below the 25th percentile on a test given 25 years ago (or even on a more recently administered test), it is impossible to have fewer than 25% of students score below the 25th percentile on the test their group has taken for the purpose of describing the distribution.

Thus, questions about the comparison group and the comparison distribution are always pertinent. But of even greater importance for the present discussion is the fact that there will bealways, every single timea distribution, including a portion below average, a bottom quartile, a lowest score, and, if Gould's (1996) full house is considered, those whose disabilities are profound and whose scores are zero. NCLB blithely disregards this reality and is, consequently, doomed to disgrace as education policy. No Measurement is Error Free Regardless of the type of measurement employed (whether categorical, ranking, or continuous; whether personal judgment of performance or a standardized multiple-choice test), it is not absolute or perfect. This is true also in the physical world: All measurements, even the most precise ones in physics, have a margin of error. Thus, measurements are always estimates, although some are more reliable, more valid, more precise, or more accurate (i.e., less variable) than others. Generally speaking, the less the margin of error, the more desirable the measurement. Those seeking to improve a particular field of study are always desirous of more precise and less variable measures. It is wise to recognize two realities about the measurements used in education: (a) They are often highly useful, and (b) they have a margin of error. The fact that educational measurement contains error is not justification for ignoring or refusing to use it. Error just needs to be recognized and taken into account in educational decisions based on measurement. Measurement tools that are known to have a margin of error (e.g., speedometers, thermometers, tests of various kinds) are nonetheless useful. Margins of error, possible sources of error, and judgments based on measurement must simply be recognized for the realities that they are. Measurement may be essential, but it does not preclude judgment. All measurement produces false positives and false negatives Mistakes include false positive and false negative decisions. That is, error can go either way: One can identify something that does not really exist (a false positive) or fail to identify something that really does exist (a false negative). In any given circumstance, it is wise to acknowledge both (a) the usefulness of the measurement and (b) the probability that the true value or score is different from the one obtained. In all cases there must be judgment about which is worsean error known as a false positive or an error known as a false negative. In applied sciences, such as medicine and education, there is often uncertainty about what measurements mean. The medical practitioner, for example, may wonder whether a patient does or does not actually have cancer. A false positive means needless treatment because the cancer does not exist; a false negative means treatment is not given when the person actually has cancer. An educator may wonder whether the student does or does not have a learning disability. A false positive means a needless label and procedures to deal with a nonexistent disability; a false negative means that the student's actual disability is unidentified and not responded to with any special procedures. Neither false negatives nor false positives are desirable, so measurement creates a dilemma for the practitionerwhich is the worse kind of error to make? The dilemma cannot be avoided (see Ruscio, 2002). More prevention results in more uncertainty

The idea of prevention is very attractive until those who praise the idea encounter the realities that go with prevention. When the realities set in, then prevention is usually avoided, except in rhetoric (see Kauffman, 1999, 2003, 2004, 2005a). One of the realities that people dislike most about prevention is that in practice it requires dealing with higher levels of uncertainty regarding individuals. This is simply because of the mathematical reality that moving the criterion for identification (and, therefore, for preventionyou cannot prevent what you will not anticipate and identify) closer to the average always results in greater uncertainty and a higher risk of false positives, because more cases are involved and the chances of error do not decrease in proportion to the number of cases encountered. No rhetorical maneuver can make this reality go away. With preventive action comes the higher risk of making the errors known as false positives. If one argues that children are misidentified as having disabilities in unacceptable numbers using current practices, then one is unlikely to embrace the actual practice of prevention, which would require risking even higher numbers of false positives. Thus, prevention requires that the false negative be seen as worse than the false positive and also requires acceptance of the reality that the percentage of false positives will increase. The Measured Value That Should Trigger Identification Is Arbitrary There is no nonarbitrary means of identifying students who need special education or other special attention. Just as is the case with driving age, voting age, short stature, obesity, low birth weight, and all other categories derived from continuous distributions, an arbitrary criterion is picked (which may or may not be defensible and which may be changed at will) that triggers a right, category, or concern. Arbitrary does not mean ill-considered or wrong, only that the criterion could be moved up or down, depending on one's interpretation of data or one's sociological preferences. Thus, as Kauffman and Hallahan (2005) pointed out, the criterion for definition of any disability or giftedness may be changed because the nature of disability and giftedness is not an immutable fact (as is, for example, the speed of light, the nature of a species, or the specific gravity of a material). Disability and giftedness are always judgmental. Thus, there is redundancy in speaking or writing of judgmental disabilities, as if some are not. True, a disability may be based on realities about what a person can or cannot do, but it is an arbitrary judgment. So is the judgment that a child is gifted. Recognizing an inability to do something as a disability or judging a superior ability to perform to reflect giftedness may, indeed, have important social consequences. Nevertheless, disability is a judgment rendered about the meaning of an inability, and it is usually a judgment made for good (defensible) reasons. Giftedness is a judgment rendered about the meaning of superior ability, also usually for good reasons. Yearly Measurement Is Useless for Preventing Failure Yearly measurement (like that demanded by NCLB for assessing AYP) has its place in saying how groups and individuals are doing compared to a standard. However, such measurement is useless for avoiding failure. Prevention of failure requires anticipating it or at least catching it in its incipient stages, not pointing it out long after the fact (see Kauffman, 2002, 2003, 2004, 2005a). The kind of testing demanded by NCLB can tell educators that something went wrong (or right), but it cannot tell them

precisely what went right or wrong or even when it occurred, and it is always too late to avert disaster for an individual. The kind of measurement that is useful for averting prolonged failure is curriculum basedfrequent, if not daily, assessment by the teacher of a student's progress in the curriculum. Moreover, no one has invented a measurement device that is reliable and valid, yet produces no variance in what is measured (probably such a device cannot be invented, as it contradicts what one knows about measurement). And when a designated group of students is measured, it is simply a reality that there will be a lower or bottom part of the distribution that will include those who are lowest or last and some whom will be judged to have failed. This is why setting a universal standard for performance, as suggested by NCLB, is unrealisticunless universal is taken to mean all who do not fail, in which case it is clearly nonsensical. Labels Are Essential The suggestion that services can be provided without labels is another evasion of reality. Although the labels may not be those that are in vogue, labels of some nature are required to designate the matter of concern. Labels for educational realities are required as certainly as they are for material things. Insisting that labels should be avoided is just another way of suggesting that something should be unmentionable. Moreover, statements such as We do not believe a person has an intellectual disability; rather, the person is defined by others as having the condition (Kliewer, Biklen, & KasaHendrickson, 2006, p. 188) are merely attempts to absolve oneself of responsibility for naming unpleasant realities, to place blame on others, and to pretend that definitions of disabilities are inappropriate constructions of realities. Clear Communication Demands Words (i.e., Labels) Labels cannot be avoided because words are required to describe things, including people's characteristics. One knows that people have learned discriminations among things when they can label them accurately. Emotional disturbance or deafness or mental retardation, for example, cannot be designated (i.e., discriminated from one another or from the absence of them) without a label of some kind. In short, labels for the things someone wants to talk about are indispensable. The alternative to labeling is refusing to talk about the thing one wants to avoid labeling (see Kauffman, 2003). Unwillingness to talk about something implies not knowing about it or wanting to avoid it or pretending that it does not exist except as someone else's (morally suspect) judgment. Labels for Treatments Label the Individuals Who Receive Them Some have suggested using labels for treatments but not for individuals. The reality is that individuals are categorized by the fact that they receive a given program or treatment. Inevitably, they are those who receive _____, a category that carries a label for anyone who receives it. Every attempt to avoid all labels for individuals does two things in reality: (a) It makes communication difficult or impossible, and (b) it makes whatever label one settles upon all the more distasteful to those who use it and increasingly stigmatizing to those who are labeled. Thus, attempts to avoid special education because it involves labeling are misguided, as students are labeled merely by recognition that they need something that

other students do not. The problem of stigma is not the fact that there are labels for conditions or for their treatment but the fact that people react negatively to the label. Euphemisms Are Unhelpful and Lead to Ridicule People soon figure out the euphemisms educators use for the phenomena called disabilities, failure, or superior achievement. Referring to a disability as a challenge sets up people with disabilities for eventual ridicule (see Kauffman, 2002, 2003). Euphemisms fool no one for long, but they do confuse communication for a while and ultimately make whatever is referred to appear more negative or less worthy of respect than the original term. They also contribute mightily to humor, eventually making whatever is called by the euphemism a laughingstock (see Carlin, 2004). Special Services Are Essential Suggestions that all education should be special education or that general education should be so flexible and inclusionary that it meets the needs of all learners are representative of the world of makebelieve, not the world of realities. Although it is no doubt the case that general education should be improved and could be made more flexible, neither it nor any other system can be made infinitely flexible. Infinite flexibility is like perpetual motionnot something the real world allows. Both children with disabilities and gifted children are cheated by the absence of special services to meet their unusual needs (see Kauffman & Hallahan, 2005). Good General Education Is Necessary But Never Sufficient for All Learners The improvement of general education is highly desirable. However, no improvements in general education will or can eliminate low achievers or the need for special education. Better general education may well raise the mean achievement for the population, and it may even raise the achievement of many relatively low achievers. Nevertheless, education of no description can eliminate the distribution of outcomes or detach the left or lower tail of the curve from zero. Improved general education may very well be necessary, but it is not and never can be sufficient to address the needs of all learners (Kauffman & Hallahan, 2005). Special Education Is Designed to Deal with General Education's Failures No matter how good general education may become, it will always produce failures. This is partly because of the fact that learners differ enormously in their ability to acquire particular knowledge and skills and also because general education must be thatgeneral, designed for the masses, not for the outliers in a distribution. Special education is specifically designed to accommodate the outliers. True, special education may indeed be an integral part of public education, but it cannot be invisible or indistinguishable from general education. Regardless how good general education becomes, its failure to meet the needs of some students is predictable, and it should not be castigated because special education is needed as a supplementary service to students at the extremes of the distribution of achievement. Moreover, making the right comparisons is critically important. Condemnation of special education because students with disabilities do not perform similarly to students without disabilities is

grotesquely wrong. The appropriate comparison for evaluating the success or failure of special education is how students with disabilities perform with special education versus how they perform without it (Kauffman & Hallahan, 2005). Other professions face the same needfor specialists to address the needs of clients for whom generalists' skills are insufficient. Education is not peculiar in this regard, and the reality of the need for special education to address the needs of students for whom good general education is inadequate must be recognized. Good Teaching Is Not Simply Intuitive Some who comment on education, particularly those who are not in the education profession, seem to assume that teaching comes naturally and does not require any special training. To be sure, some natural geniuses can be found in teaching as well as in other pursuits. Nevertheless, teaching is not something that comes naturally or intuitively to many. The same is true for playing musical instruments, decoding the law, flying airplanes, practicing medicine, and so on. Schools of education are often criticized, and often for good reason, because they do not concentrate on teaching what is known about instruction and behavior management (see Bateman, 2004; Engelmann, 1997; Kauffman, 2002; Kauffman & Landrum, 2006; Kauffman, Mostert, Trent, & Pullen, 2006; Landrum & Kauffman, 2006; Zigmond, 1997). However, that does not mean that schools of education are dispensable. Schools of education should teach instructional finesse. If they teach only or primarily philosophical propositions or dispositions toward social issues and offer comparatively little or nothing about how to provide effective instruction, then they are derelict. Research on teaching is needed to discover which instructional methods produce better learning outcomes. Instructional methods cannot be empirically sound without scientific tests of their effects. Science demands confronting realities for what they are. There is a science of instruction and behavior management, albeit one that requires a great deal of work. Good Teaching Is Not Enough In many ways, accountability contributes to the productivity of any organization. It is important, however, that this burden be imposed at all relevant levels that can contribute to success. NCLB places the accountability of student-demonstrated achievement growth, and a narrowing of measured achievement heterogeneity, squarely on the shoulders of teachers and schools. Holding teachers and schools entirely responsible for increasing student achievement and closing gaps is not consistent with reality. Classroom quality has indeed been shown to have positive influences on student achievement (LoCasale-Crouch et al., 2007). However, it is equally important to consider that children only spend about 17% of their time in school (Berliner, 2005). Placing the burden of accountability on the shoulders of teachers fails to account for the complex network of factors that contribute to student achievement growth. In addition to the better known influences on learning (e.g., motivation, ability), a growing and compelling body of literature demonstrates the importance of parental involvement (in its various forms) on children's education (Epsteing & Sanders, 2000; Fan & Chen, 2001). Holding teachers solely

responsible for student learning is a disservice to children and an insult to teachers. The National Council of Churches (2006, Concern 6) raised this as a moral concern, although it is also an empirical issue. IMPLICATIONS FOR POLICYMAKERS AND PRACTITIONERS Jump to section REALITIES THAT MUST BE ACKNOWLEDGED IMPLICATIONS FOR POLICYMAKERS AND... CONCLUSION As J. J. Gallagher (2007) noted, educational policy casts long shadowsit influences services, personnel preparation, and research. Policy is not a trivial matter for the practitioner. It affects what teachers are expected to do and what and how they teach, and a given policy is likely to affect educational practice for a generation or more. Thus, it is incumbent upon policymakers and those who advise them to acknowledge all realities, as ignoring any given reality will neutralize policy initiatives, no matter how well intended and no matter whether they are grounded in reality in most respects. NCLB is an example of a well-intentioned policy that ignores realities about groups and gaps in educational performance and that is therefore headed for eventual ignominy (Johns, 2003; Kauffman, 2004, 2005b). It ignores the many realities involving distributions of test scores, the reasons for success and failure, and the nature of the differences among groups. The fact that the law was passed with overwhelming and bipartisan support does not make it rational or defensible. Votes may indicate the political palatability of a policy, but they do not make a policy that ignores realities viable. That is, votes do not change all of the realities of the world, only the political reality that they represent. The implications of this discussion for policymakers and practitioners are simple and straightforward: Before you enact or support a policy, make sure that it is firmly and thoroughly grounded in realities about the world of education. Resist all attempts, for whatever reasons, to enact or support a policy that you know skirts any reality. Do not succumb to the temptation to engage in rhetoric that does not conform to every known reality about education. Demand of advisors the kind of data and thinking that will convince reasonable skeptics of the policy's workability. Some practitioners do not acknowledge realities, partly because their teachers or their policymakers (or both) do not. They have been convinced that realities can be ignored without disastrous consequences. Most practitioners do know fantasy from reality about education. Nevertheless, many acquiesce in efforts to implement policies that they know are unworkable because they believe that refusing to go along with pretense will cost them their employment.

We are not under the illusion that policy must be shown to be flawless before it can be supported. However, supporting a policy known to be conceptually faulty is in our opinion unethical. Finding flaws not known to exist at the time a policy is enacted and fixing them is inevitable. Gillon (2000) pointed out how policies often have unintended consequences, and we do not believe that people can be held responsible for the unpredictable effects of their behavior. Nevertheless, it is also the case that in American society when a person knows (or should know) the negative consequences of an action but performs that action anyway, citizens consider that person negligent or malicious. American society does not countenance constructing and marketing goods that are known to be defective. Neither do Americans fail to hold people responsible for negative outcomes when those people knew or should have known that their actions would harm others. For example, people accept the fact that it is malicious to manufacture and sell vehicles that people know or should have known are unsafe. The courts generally excuse people from responsibility for the unknowable, but not from the known or knowable. Neither our colleagues in education nor those in politics are free of responsibility when they embrace policies with known deficiencies. We find it incomprehensible that people of average or better intelligence produce or defend education policies like NCLB that so clearly ignore important realities. In most civil endeavors other than education, criminal penalties are exacted or people are found insane for pretending that fundamental realities do not exist. The education of children is too important to be approached with sloppy thinking, and the consequences of bad thinking about education should be no less serious than the consequences for bad thinking about planes, trains, automobiles, and other things that are important to society. CONCLUSION Jump to section REALITIES THAT MUST BE ACKNOWLEDGED IMPLICATIONS FOR POLICYMAKERS AND... CONCLUSION The improvement of education, general or special, demands the acknowledgement of basic realities. These realities are things that happen regardless of anyone's wishes or statements; they are features of the world that humans may have discovered and described (or constructed through their language), but they will not change because of philosophical preferences or denials. Any philosophy, pronouncement, or policy based on the pretense that these realities do not exist or will not be encountered is doomed to eventual failure, regardless of its social acceptability, emotional appeal, or apparent initial success. Most people in American society seem to understand that any project in the material world that is launched with even a single flawed assumption regarding physical realities will come to naught and may result in very negative consequences for those involved in it. A car may be well designed in every other way, but if a design assumption about the brakes is wrong, the result will be extremely undesirable consequences. Educators and other people in this society must understand that something similar occurs in the world of educationthat any educational project, reform, or policy launched with a flawed

assumption regarding reality is similarly headed for inevitable failure. Rhetoric about education that does not account for basic realities must be labeled the siren call that it is and be resoundingly rejected. NCLB need not be honored. In fact, it should be recognized as another example of silly talk about education. Written or spoken proposals that ignore realities should prompt imitation of the statement of the late Senator Daniel Patrick Moynihan regarding proposed changes to the Aid For Dependent Children program: Are there no serious persons in the administration who can say, Stop, stop right now? No, we won't have this? (Moynihan, 1995, p. A31). Of many reforms and proposals, more people need to say, Stop! Stop right now! We won't have this! Frustration with the powers that be is not a sufficient excuse to go along with ideas that are not fully grounded in reality (Kauffman, 2002, 2005b). Neither is the suggestion that a proposal is known to be partly or mostly right and therefore worthy of support. Only one assumption not grounded in reality is sufficient to render a project or proposal inoperable or disastrous. NCLB is no exception. Enacting or supporting legislation that is known to be or should have been known to be conceptually flawed is, in our opinion, highly irresponsible. ACKNOWLEDGMENTS A shorter version of this article focused on labeling and the nature of special education will be published by Learning Disabilities: A Multidisciplinary Journal under the title Labels and the Nature of Special Education: We Need to Face Realities. REFERENCES 1. Associated Press. 2006. No child raises school segregation fear http://www.ap.washingtontimes.com/dynamic/fronts/HOMERetrieved April 22, 2006, from 2. Basken, P. March 29 2006. States have more schools falling behind. In The Washington Post March 29, A17 3. Bateman, B. C. 2004. Elements of successful teaching: General and special education students, Verona, WI: IEP Resources. 4. Berliner, D. 2005. Our impoverished view of educational reform http:///abwww.tcrecord.org/content.asp?contentid=12106Retrieved April 19 from 5. Brantlinger, E. A., ed. 2006. Who benefits from special education? Remediating (fixing) other people's children, Mahwah, NJ: Lawrence Erlbaum Associates, Inc.. 6. Carlin, G. 2004. When will Jesus bring the pork chops?, New York: Comedy Concepts. 7. Cuban, L. 1990. Reforming again, again, and again. Educational Researcher, 19(1): 313. 8. Edelman, M. W. 2006. Mr. President: We want our slogan back http://www.ncccusa.org/poverty/commentary-edelman-august.htmlRetrieved April 22, 2006, from

9. Engelmann, S. 1997. Theory of mastery and acceleration. In Issues in educating students with disabilities, Edited by: Lloyd, J. W., Kameenui, E. J. and Chart, D. 177195. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.. 10. Epsteing, J. L. and Sanders, M. G. 2000. Connecting home, school, and community: New directions for social research. In Handbook on the sociology of education, Edited by: Hallinan, M. T. 285306. New York: Kluwer Academic/Plenum. 11. Fan, X. and Chen, M. 2001. Parent involvement and students' academic achievement: A metaanalysis. Educational Psychology Review, 13: 122. [CrossRef], [Web of Science ] 12. Gallagher, D. J., ed. 2004. Challenging orthodoxy in special education: Dissenting voices, Denver, CO: Love. 13. Gallagher, D. J. 2006. If not absolute objectivity, then what? A reply to Kauffman and Sasso. Exceptionality, 14: 91107. [Taylor & Francis Online] 14. Gallagher, J. J. 2007. Driving change in special education, Baltimore: Brookes. 15. Gartner, A. and Lipsky, D. K. 1989. The yoke of special education: How to break it, Rochester, NY: National Center on Education and the Economy. 16. Gillon, S. M. 2000. That's not what we meant to do: Reform and its unintended consequences in twentieth-century America, New York: Norton. 17. Gould, S. J. 1996. Full house: The spread of excellence from Plato to Darwin, New York: Three Rivers Press. 18. Heshusius, L. 2004a. From creative discontent toward epistemological freedom in special education: Reflections on a 25-year journey. In Challenging orthodoxy in special education: Dissenting voices, Edited by: Gallagher, D. J. 169230. Denver, CO: Love. 19. Heshusius, L. 2004b. Special education knowledges: The inevitable struggle with the self.. In Challenging orthodoxy in special education: Dissenting voices, Edited by: Gallagher, D. J. 283309. Denver, CO: Love. 20. Johns, B. H. 2003. NCLB and IDEA: Never the twain should meet. Learning Disabilities: A Multidisciplinary Journal, 12(3): 8991. 21. Kauffman, J. M. . What happens when special education works: The sociopolitical context of special education research in the 1990s. Annual Meeting of the American Educational Research Association. Boston, MA. Invited address, Special Education Special Interest Group, April 19, 22. Kauffman, J. M. 1999. How we prevent the prevention of emotional and behavioral disorders. Exceptional Children, 65: 448468. [Web of Science ]

23. Kauffman, J. M. 2002. Education deform: Bright people sometimes say stupid things about education, Lanham, MD: Rowman & Littlefield Education. 24. Kauffman, J. M. 2003. Appearances, stigma, and prevention. Remedial and Special Education, 24: 195198. [CrossRef], [Web of Science ], [CSA] 25. Kauffman, J. M. 2004. The president's commission and the devaluation of special education. Education and Treatment of Children, 27: 307324. 26. Kauffman, J. M. 2005a. How we prevent the prevention of emotional and behavioural difficulties in education. In Handbook of emotional and behavioural difficulties, Edited by: Clough, P., Garner, P., Pardeck, J. T. and Yuen, F. K. O. 429440. London: Sage. 27. Kauffman, J. M. 2005b. Waving to Ray Charles: Missing the meaning of disability. Phi Delta Kappan, 86: 520521. 524 [Web of Science ] 28. Kauffman, J. M. and Hallahan, D. P. 2005. Special education: What it is and why we need it, Boston: Allyn & Bacon. 29. Kauffman, J. M. and Landrum, T. J. 2006. Educational service interventions and reforms. In Handbook of intellectual and developmental disabilities, Edited by: Jacobson, J. W. and Mulick, J. A. 173188. New York: Kluwer. 30. Kauffman, J. M., Mostert, M. P., Trent, S. C. and Pullen, P. L. 2006. Managing classroom behavior: A reflective case-based approach, , 4th ed., Boston: Allyn & Bacon. 31. Kauffman, J. M. and Sasso, G. M. 2006a. Certainty, doubt, and the reduction of uncertainty: A rejoinder. Exceptionality, 14: 109120. [Taylor & Francis Online] 32. Kauffman, J. M. and Sasso, G. M. 2006b. Toward ending cultural and cognitive relativism in special education. Exceptionality, 14: 6590. [Taylor & Francis Online] 33. Kliewer, C., Biklen, D. and Kasa-Hendrickson, C. 2006. Who may be literate? Disability and resistance to the cultural denial of competence. American Educational Research Journal, 43: 163192. [CrossRef], [Web of Science ] 34. Landrum, T. J. and Kauffman, J. M. 2006. Behavioral approaches to classroom management. In Handbook of classroom management: Research, practice, and contemporary issues, Edited by: Evertson, C. M. and Weinstein, C. S. 4771. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. 35. Lawrence, M. April 7 2006. The impact of No Child Left Behind, April 7, San Francisco, CA: Presidential invited speech to the American Educational Research Association. 36. Lo Casale-Crouch, J., Konold, T. R., Pianta, R. C., Howes, C., Burchinal, M., Bryant, D., Clifford, R., Early, D. and Barbarin, O. 2007. Profiles of observed classroom quality in state-funded pre-kindergarten

programs and associations with teacher, program and classroom characteristics. Early Childhood Research Quarterly, 22: 317. [CrossRef], [Web of Science ] 37. Mostert, M. P., Kavale, K. A. and Kauffman, J. M., eds. Challenging the refusal of reasoning in special education, Denver, CO: Love. in press 38. Moynihan, D. P. September 21 1995. I cannot understand how this could be happening.. In The Washington Post September 21, A31 39. National Council of Churches. 2006. Ten moral concerns in the implementation of the No Child Left Behind Act http://www.ncccusa.org/pdfs/LeftBehind.pdf.Retrieved April 22, 2006, from 40. No Child Left Behind. 2001. Executive summary of the No Child Left Behind Act http://www.ed.gov/print/nclb/overview/intro/execsumm.htmlRetrieved March 19, 2006, from 41. Popham, J. W. 2004. America's failing schools: How parents and teachers can cope with No Child Left Behind, New York: Routledge Farmer. 42. Rothstein, R., Jacobsen, R. and Wilder, T. . Proficiency for all: An oxymoron. Paper presented at the Examining America's Commitment to Closing Achievement Gaps: NCLB and Its Alternatives symposium. November, Manhattan, NY: Teachers College, Columbia University. 43. Ruscio, J. 2002. Clear thinking with psychology: Separating sense from nonsense, Pacific Grove, CA: Wadsworth. 44. Sasso, G. M. 2001. The retreat from inquiry and knowledge in special education. Journal of Special Education, 34: 178193. [CrossRef], [Web of Science ], [CSA] 45. Shermer, M. 1997. Why people believe weird things: Pseudoscience, superstition, and other confusions of our time, New York: Freeman. 46. Shermer, M. 2001. The borderlands of science: Where sense meets nonsense, New York: Oxford University Press. 47. Winerip, M. April 5 2006. No child left behind? Ask the gifted. In The New York Times April 5, B7 48. Zigmond, N. 1997. Educating students with disabilities: The future of special education. In Issues in educating students with disabilities, Edited by: Lloyd, J. W., Kameenui, E. J. and Chard, D. 377390. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.