Rational Responses to High-stakes Testing and the Special Case of Narrowing the Curriculum
Publication Date: 2009-09-15
Citation: Paper presented at the International Conference on Redesigning Pedagogy, National Institute of Education, Nanyang Technological University, Singapore, June 1, 2009.
This is the perfect quote to begin a paper about curriculum for American and Singaporan schools at the start of the 21st century. For Adams, the ultimate goal of education in a free society was to foster the arts. War was a necessity so that his children had freedom. Freedom was necessary to generate wealth by means of the commercial and agricultural skills that the new nation possessed and to do so independent of European powers. Adams knew, however, that the ultimate goal of education for a free person was the opportunity to appreciate and participate in the arts.
This view of the importance of what is more broadly called the liberal arts is derived from our Greek heritage. The root of the term Ă˘€śliberalĂ˘€ť as used in the "liberal arts" actually has little to do with the term "liberal" in the political realm. Instead, the root of the term "liberal" when referring to the arts is in liberty, indicating that art, music, government, rhetoric, philosophy and the like, are the arts of free men and woman. Benjamin Barber (1994) contrasted the liberal arts against what were called the servile arts: business and law enforcement, food preparation and medicine, mechanics and technology. The servile arts encompass the knowledge and skills needed to run the country. This was exactly what the Greeks wanted their slaves and servants to do while they engaged in the kind of grand political and philosophical thinking that gave the Western world its intellectual heritage.
America being a practical country, a land for the promotion of Yankee ingenuity and dominated by the interests of business, never was in awe of the liberal arts or the humanities. There never was a golden age in America in which the liberal arts and humanities flourished while the servile arts were looked down upon. In fact, for at least the first hundred years of widespread public schooling in the US, the practical was deemed appropriate to teach most middle-class white students and was considered even more desirable as the curriculum for the poor and minorities, were they afforded any education at all. The exception to this focus on the practical was for a small and select group of wealthy, white, male students who did get to study the liberal arts in upper high school grades and college.
Today may actually be worse for poor children in the US than at any time in the last half century. This is because the lower classes are being kept from the liberal arts and humanities curricula by design. Using the argument that we must get their test scores up, we in the US are designing curriculum for poor children, often poor children of color but certainly, numerically, for poor white children, that will keep them ignorant and provide them with vocational training, at best. Their chances of entrance to college and middle class lives are being diminished, and this is all being done under the banner of "closing the gap," a laudable goal, but one that has produced educational policies with severe and negative side effects.
What we find today is that America's students are primarily enrolled in courses Adams understood were needed for his children. Our youth are not enrolled in great numbers in the courses Adams wanted for his grandchildren. The theme of this paper is that the curriculum balance, long in favor of the practical and the servile arts, has tipped even further in that direction, particularly in the schools that serve poor and minority children. America apparently has developed an apartheid-like system of education.
The newest difficulty in promoting the arts and humanities in the curriculum is due to the use of high-stakes testing. Such testing is found in all 50 states as a function of the No Child Left Behind (NCLB) act that went into effect in 2002. I will make clear, however, that the decrease in exposure to certain curricula is a rational response to high-stakes testing. But this decrease in exposure to a varied curriculum is of great concern as we contemplate what the 21st century might have in store for our youth. Compared to the past, the future is likely to be more Volatile, Uncertain, Complex, and AmbiguousĂ˘€"A VUCA world (Johansen, 2007) for our children to face. I think adaptation to such a world requires a citizenry with the broadest possible curriculum, not a narrow one that constricts the skills of the youth because of a need to demonstrate accomplishments on a small set of assessments.
Curriculum time and curriculum content are changing and narrowing in the USA as function of over reliance on a small set of assessments. My data come from the Center on Education Policy (2008a), a non-partisan monitor of the effects of NCLB. Table 1 presents their data on curriculum changes from a nationally representative sample of school districts.
Table 1. Changes Since 2001-2002 in Instructional Time for Elementary School English Language Arts and Mathematics, in Districts Reporting Increases (Center for Education Policy, 2008a).
These data show that changes in the time allocated for teaching reading and mathematics in elementary schools were quite dramatic between 2002 and 2007. These are the years of the NCLB act and mandated high-stakes testing. The time allocated to reading has been increased, on average, over two and a third hours a week, while mathematics time has been increased, on average, about an hour and a half a week. What needs to be kept in mind when interpreting this table is that the Ă˘€śaverageĂ˘€ť masks relevant information. It is likely that many school districts increased time in these subjects a great deal more than the average, because the average includes districts serving high-income children, who typically score well on the tests used to satisfy NCLB requirements. Those districts probably changed their time allocations very little. On the other hand those serving low-income students probably changed their time allocations a lot. In a previous study of the curriculum by the Center for Education Policy (2006), 97 percent of the school districts not making Adequate Yearly Progress (AYP) had changed their curriculum times and content. These are, of course, the districts where the pressures to score higher on the tests are greatest and where you expect rational responses to that pressure.
Clearly, the changes documented above may not be all bad. We can all agree that it is highly desirable that students read and do mathematics well. Thus we might normally welcome such changes unless there is reason to believe that the increased time is used poorly and is having either no effects, or detrimental effects, on students. This is where things get interesting: Evidence exists that this may indeed be the case.
If reading and English language arts consists of too much phonics practice; too much drill and test preparation; too many worksheets for practicing reading skills; not enough writing to express complex thoughts; not enough reading for enjoyment; and not enough reading of academic material to increase vocabulary in order to aid comprehension; then the reading is more to foster the goal of basic literacy and not literacy for its pleasure, or for its value in exploring the arts, the sciences and the humanities.
We are learning that reading that is more critical and emancipatory is not stressed as frequently when tests are highly consequential for schools. Instead, drill on test formats and items suspiciously like those on the test constitutes too much of the reading curriculum (Nichols and Berliner, 2007). Sadly, evidence exists to support the hypothesis that the increased time spent on reading and mathematics is not helping us make better readers and mathematicians. Table 2 presents gains made in reading at the 4th and 8th grade on the National Assessment of Educational Progress, tests we call the NAEP tests. These are considered by many to be the best tests the USA has, and they show quite clearly that the gains made before NCLB were greater than the gains made after NCLB. Despite large increases in time allocations for reading, greater growth in reading is not observed.
Table 2. 4th and 8th grade reading gains and rate of gain before and after NCLB became law.
A second look at reading achievement and the effects of greatly increased reading instruction on the performance of the various US states comes from the Educational Testing Service (Barton & Coley, 2008) and is presented as Table 3. What is obvious is that average scores are not increasing, and many states have actually done worse since the enactment of NCLB.
Grade 8 reading change, 2002-2007* in: States Improving States Unchanged States Doing Worse
Average Score 0 32 12
Percent Proficient 0 40 3
Average score for students in the top quartile 5 27 11
Average score for students in the bottom quartile 5 24 14
*Includes data from states participating in both assessments, as well as the District of Columbia and Department of Defense schools.
Table 3. Changes in average reading score and percent proficient in reading for grade 8 US students on the NAEP tests, 2002-2007.
The evidence is that the schools with the poorest children, and therefore the schools with the greatest likelihood of being sanctioned under NCLB, are those where the reading curriculum in now often of the most basic type. While such a level of literacy might have been good enough at the beginning of the 20th century, it is hard at the beginning of the 21st century to defend the forms of instruction used and the kinds of literacy attained by the children in many of our poorest schools.
Some of our psychological theories are appropriate to interpreting these data. For example, although we have argued in our journals over the power of the undermining hypothesis, it does appear to be a replicable experimental finding. We now know, therefore, that for many children the motive to engage in activities found pleasurable for their own sake is diminished when those same tasks are rewarded. This suggests that a significant number of poor and minority children who really do enjoy reading for pleasure and edification are much more likely to be turned off of reading because reading has become a task governed by extrinsic rewards. In many schools with the poorest students stars are awarded for rather trivial multiple-choice questions answered correctly about books just completed. There are other rewards for competitive reading, which appears to me to be akin to a new school sport, where rewards are given for the number of books read per unit of time. Other schools have class parties for high numbers of books read collectively per unit of time. None of these approaches is wrong from a behaviorist theory, yet all of these short-term motivational strategies are likely to have a negative influence on continuing motivation to read. We don't know this, of course, because we usually do not study the long-term effects of these programs. But there is good reason to believe that continuing motivation to read will suffer under some of these instructional programs.
There is another theory in our field that comes to mind when looking at these data. It is related to time and learning. I did some of that research myself. From all the research, and from the common sense that is found in the humblest of homes, we have been able to derive a sound educational law, namely, that the more time students spend studying in some area of the curriculum, the more likely they will have learned more in that area. Time and learning are believed to be, and are empirically found to be, causally related. But this principle of learning is directly challenged by the reading data we have. Significantly more time spent in reading is leading to less improvement on the high quality assessments of reading that are used, the NAEP tests. This suggests that students may be studying the wrong things, or that their motivation is being undermined, or both. This is not good.
What do we know about mathematics? As noted in Table 4, we see almost the same things we noted when looking at reading. The exception is eighth grade gains for Hispanics. All other cells show a pattern of higher gains before NCLB and the additional time that was allocated for mathematics instruction.
Table 4. 4th and 8th grade mathematics gains and rate of gain before and after NCLB became law.
Actually, we learn from the 41 states for which there were complete data at the 8th grade that mathematics scores overall did go up quite a bit over time. Its just that in 24 of those states the gains were larger in the three years before NCLB than in the 4 years after NCLB. So NCLB seems to improve the gains in achievement in fewer than 50% of the states. And in about 10% of the states for which we have data NCLB has no discernable effects at all. So in mathematics, as was true in reading, NCLB does not seem a sensible social policy.
While rarely taught as well as the experts would like it to be taught (c.f. Ball, Lubienski & Mewborn, 2001; Lampert, 2001), mathematics can be even more boring and inadequately taught than ever before under the threat of sanctions. Mathematics, can [be] a subject that is a rich source of discourse and debate, of conjecture and the testing of ideas, and even an important contributor to democratic practices (Ball & Bass, 2008) when taught correctly. But like reading it can be turned into a drill oriented, teacher dominated subject in which the increased time results in increased boredom and dislike of the subject.
Moreover, in a recent analysis by Jaekyung Lee, in RER, he pointed out that the achievement gap between wealthier and poorer students has not been closing at all on the National Assessment of Educational Progress, the best audit test America has (Lee, 2008). This all suggests that something is quite wrong. Increases in allocated time ought to result in increased learning. But if the increased time for learning reading and learning mathematics results in a less interesting curriculum for teachers to teach, and for students to learn, then the results we are getting are actually quite sensible, though certainly quite disheartening. Questions need to be asked about motivation to teach and motivation to learn, and about the attitudes formed toward the subject matter, through the instruction that accompanies the high-stakes testing imposed by NCLB. Such research is specially needed in schools that serve the poor.
I am struck, as always, by DeweyĂ˘€™s prescience and our failure to take him seriously: he said:
Other odd data about contemporary curriculum exist. For example, England has also tried to reform its schools through accountability plans that are heavily test oriented. And it seems to have problems similar to the US. In fact, for the first time since IQ tracking was started, in which an upward trend has been the norm for decades, IQs appear to be declining quite dramatically in UK middle grades. And this has occurred in a relatively short amount of time, according to James Flynn, who gave us the Flynn effect and monitors these trends worldwide (Flynn, 2009). In truth, no one currently has a good explanation for this downward trend. But at least one reason given is that the UK has become a test-oriented culture and this has stunted the intellectual growth and motivation to learn among British students.
Following up his own research of the 1970s, the British classroom researcher Maurice Galton found that teacher-centered pedagogy, characterized by interactions of a very low cognitive level, managerial in their intent, had increased dramatically between 1976 and 1996. Pupils had fewer opportunities to question or to explore new ideas after the tests became the primary instrument that the government used to change the schools (Galton, 2007). Assessment pressures have resulted in 42% of teachersĂ˘€™ time being taken up with whole class teaching, compared to 18% in 1976. In primary schools in England there are now few opportunities for expressing anything that resembles creative reasoning (Galton, 2007).
Galton and McBeath (2002) surveyed primary teachers in England. Teachers in their study regret that time pressures no longer allow them to engage in informal conversations with individual children during lessons, or to allow pupils, at certain times, to pursue their own ideas and interests as part of topic and project work. The British data tell us that since the seventies this time has decreased by nearly 50%. Yet, teachers regarded exchanges of that kind as highly rewarding and motivating because they greatly enhanced the teacher-child relationship and provided what some classroom practitioners described as 'magic moments.'
Galton and McBeath (2002) quote teachers. For example, a female with 23 years experience says:
Another female teacher with 9 years experience says:
Galton and McBeath (2002) report that many teachers noted the creative subjects were being squeezed out, with the consequence that there were fewer opportunities for children to be good at something, to succeed or to excel, and the teachers knew that this was not good for the children and it made classroom management all that more difficult. The emphasis on the core subjects, with increased focus on content, simply meant that there was less space in the school day for less structured activities, though it was in those kinds of activities that some non-academic children excelled.
The British and US experience is exactly what Hong and Youngs (2008) report happened to curriculum in their study of Chicago and Texas, as that district and that state responded to high-stakes testing. In Chicago the researchers found that high-stakes testing seemed to narrow the curriculum and make it harder for students to acquire higher-order thinking, writing, and problem-solving skills. In Texas, it was found that schooling changed in ways that emphasized rote learning, not broad intellectual skills (Hong & Youngs, 2008; McNeil, 2000). Lipman (2004) also studied the Chicago schools and reports that the accountability program insured that the more affluent students in Chicago received a much richer and more intellectually challenging curriculum than did the poor children in Chicago. Poor minority children, in particular, were required to memorize fragmented facts and information, and they were constantly taught simple test-taking techniques. Lipman is probably quite right when she says that this differential access to high-quality curriculum will have significant consequences in terms of the social inequalities we will observe in the future. White students who possess a great deal of the cultural capital valued by schools are going to be much more likely to get to college and thus more likely to attain higher status through higher paying jobs. But low SES and minority students in Chicago's schools are much more likely to end up in lower-skilled and lower-paying jobs. The decisions about curriculum and instruction in Chicago and other urban districts results in access to rigorous curriculum for some, but not for others, thus allowing for the continuation of the current unequal social structure. Apartheid is not a word I like to use, but it is descriptive of what has happened in many schools across the nation.
Let me give you an example of how this plays out in a high school language arts curriculum for low income Latinos. This is from our book, Collateral Damage (Nichols and Berliner, 2007). This is a transcript of a lesson recorded by my student Sandra Foster (2006), in a Texas high school as the students prepare for the TAKS testing-- the Texas Assessment of Knowledge and Skills. The setting is an English class with 16 Latino juniors. A young Anglo teacher has handed out an essay similar to those that are to be written on the test.
T: Okay, this is last minute work for TAKS. You can pass the test. You donĂ˘€™t want to take it again, right?
S: No response
T: You donĂ˘€™t want to have to all practice again, right?
S: No response
T: Please say yes.
S: No response.
T: You are brilliantĂ˘€Â¦.The test is not hard. Take your time; in fact take all the time you need.
S: No response.
T: Remember, be here for the test and give it all you gotĂ˘€"take youĂ˘€™re time and do your best.
S: No response
T: YouĂ˘€™re awesome. If someone tells you differently, donĂ˘€™t listen to them. DonĂ˘€™t be afraid to use those dictionaries. If you donĂ˘€™t know a word, look it up. If you want to use another word, look it up
S: No response.
T: Okay, there will be three types of open-ended questions and three types of literary selections. What does literary mean?
S: No response
T: Is it fiction, non-fiction or biography?
S: No response
T: Are you going to talk to me or you donĂ˘€™t know?
S: No response
T: Its fiction you all (in an angry voice). (Pause) First thing you do is answer the question. It must be insightful and thoughtful. Do not restate the question. You have five lines to fill in. Then you have to support a response. If you summarize in an open-ended question you get a zero. But if you use support for the passage, you get points. Look at this essay. Do you see how this student used textual support?
S: No response
T: Come on! (in an angry voice and shaking her head signaling "no"). Oh, before I forget, it is important that you must stay inside the box and you must use all five lines.
S: No response.
T: See how this student answered the question with insightful evidence?
S: No response.
And so it goes. Another exciting day at school marked only by passive resistance to what is accurately perceived to be an inferior education by these students. Performance on high-stakes tests determines the curriculum choices being made and those choices, paradoxically, may actually result in lower gains on the tests.
In many schools, particularly schools like this one that serve AmericaĂ˘€™s poor, an impoverished system of instruction is now the norm. Galton reports that things in England are no different. In year 6, when national high-stakes tests are given in May, at a school with two classes, the teachers broke up the two classes into three classes, the children who they thought would do well, those who need help, and the "no-hopers," those who teachers believed would not do well. One teacher took the high performers, another took those that needed help, and they worked a lot on test preparation materials while subjects like the arts were cancelled for them. The "no-hopers" were assigned a classroom aide for the months leading up to the test. Social class distinctions in the three different school classes were as you might expect.
Figures 1 and 2 present data related to the British, Chicago and Texas studies, from another report by the Center for Educational Policy (2008b). These data come from a case study of NCLB implementation in Illinois. What is obvious from the two graphs is that the nature of school work differs quite a bit in schools that are under scrutiny for failing to make Adequate Yearly Progress (schools needing to make improvements), and those that are not working under such pressure.
Sadly, in assessment oriented cultures like the US, it is convergent thought and mimetic processes that are valued and it is these styles of teaching and learning that we see a lot of in schools that are not scoring at the appropriate levels. But as we think deeper about this, the narrow, test-compatible curriculum extant in many classes, particularly the classes for the poor, may actually be least likely to change the constructs underlying the tests of achievement (or intelligence) that we use to assess the outcomes of schooling. While you in Singapore are aware that you may have to "teach less to learn more," we in the USA are demonstrably teaching more to learn less! It appears to be self-destructive.
The rest of the curriculum.
Time added for reading and math, in the typical 6.5 hour day, and the
typical180 day year, needs to come from somewhere else in the curriculum.
* not enough data
Table 5. Decreases in instructional time for various curriculum areas to accommodate increases in time for English language arts and mathematics. (Center for Education Policy, 2008a)
Table 5 makes clear where that time has come from. Again it is important for interpreting this table to remember that the average masks the bigger cuts that some districts have made in these subjects. With this caveat in mind we see that the teaching of social studies, intended always to be part of youth development for responsible citizenship, is down, on average, over an hour a week.
Yet Americans of all political persuasions ask that the schools help to develop citizenship. So this trend in curriculum is in opposition to the aspiration that all Americans have for their school curriculum. Furthermore, school programs that might foster citizenship have been cut because of the need for more time in reading and mathematics, so cleaning up neighborhoods and parks, visiting nursing homes, going on field trips to the legislature, projects that examine pollution at a local level, and so forth, all have been jettisoned. Although this nation has never been great at teaching community involvement as a responsibility of citizenship, I expect we will do worse in the future.
Science, a field that probably will be even more important in the 21st century than in the 19th and 20th centuries, is down, on average, over an hour a week as well. Although science is now one of the areas tested under the NCLB law, and is a privileged curriculum area, scores on the science tests do not count toward AYP. Thus, lack of progress in science, and/or low performance on science tests, can safely be ignored by schools and districts since no sanctions attach to the test. Science, like social studies has been robbed of minutes to expand time for reading and mathematics. Thus curriculum that might help insure American economic competitiveness in the future, and surely will contribute to intelligent citizenship in our science- and technology-rich future, has been sacrificed.
Table 5 also documents that time for physical education is down, despite the fact that our youth are more sedentary than they should be, are quite overweight, and Type 2 diabetes is becoming more common. It is easy to argue that physical education is more important today than ever before, and it is acknowledged as one of the most important ways to keep medical costs down as we slowly move to universal medical coverage. Yet physical education is sacrificed for the possibility of a few more points on state tests that have to rise continuously to satisfy the requirements of NCLB.
Not enough data were obtained in this study about the time allocation for lunch. But lunch is obviously wasted time for those who feel the pressure of testing under NCLB. Anecdotally, therefore, it was not surprising that a teacher at a Massachusetts district reported her concern that lunch at her elementary school was less than 15 minutes on many days "so that more time could be put in on the rigorous curriculum areas." "Rigorous curriculum areas" is code, meaning the areas that are tested. Anything else (social studies, history, government, art, music, physical education) has been defined in her school as inherently a non-rigorous subject. The school she reported on had actually abandoned traditional luncheon meals and started serving finger foodĂ˘€"wraps and chicken nuggetsĂ˘€"to get the students in and out of the cafeteria faster! (Nichols and Berliner, 2007).
Nationally, as seen in Table 5 recess was found to be down, on average, about an hour a week. We even discovered, in Maryland, that naps for preschoolers and kindergartners were forbidden by one county school superintendent.
Art and music, nationally, are down an average of an hour a week. This is particularly troublesome because the nation never spent a lot of time in these subjects. You might think that California would be a place that countered this trend, because the arts provide a large source of employment for the state and a great deal of wealth for the nation through the huge overseas earnings from films and videos that greatly reduce our trade imbalance. Yet 89% of California K-12 schools fail to offer a standards-based course of study in all four disciplinesĂ˘€"music, visual arts, theatre, and danceĂ˘€"and thus fall short of that states own goals for arts education. At the elementary level, arts instruction is often left to regular classroom teachers, who rarely have adequate training. In fact, 61% of California schools do not have even one full-time-equivalent arts specialist. Secondary schools are much more likely than elementary schools to employ such specialists, but even when they do, participation rates in secondary arts subjects are low (Woodworth, Gallagher & Guha, 2007).
The defense of the arts can be made on many grounds, but one stands out in terms of the needs of the 21st century, namely, that the arts are alternative ways to represent reality. Ideas expressed through the visual arts, dance, and music are not presented in the verbal or mathematical symbol systems that are in everyday use. So by cutting the arts we limit the ways our students can represent the world in which our students live and about which they may choose to comment. A diminution in curriculum for learning the arts, therefore, restricts our studentsĂ˘€™ ways of thinking. It limits the possibility for creativity. Yet, according to many pundits, it is creativity that will be the key to economic survival in the 21st century. The arts, by providing ways to think differently about the world, are a way to promote such creativity. And so their defense in the curriculum can be based on economic and cognitive psychological reasons, which add to other reasons for defending the arts as both a natural expressions of our humanity, and for their occasional attainment of indescribable beauty.
The curriculum for the poor
The California study also makes clear that the arts are rationed: They are taught primarily to the wealthy and not the poor.
Figure 3. Percent of California students receiving instruction in various areas of the arts, by poverty level of the school they attend (Woodworth, Gallagher & Guha, 2007)
Figure 3 presents these data. Close to twice as many students in schools that serve the wealthy (low poverty schools) receive instruction in the arts as do the students in schools that serve the poor. This an example of the apartheid system of schooling to which Kozol (2005) refers.
Wealthier students, if they are lucky, will be exposed to a wider range of the arts and humanities in their high schools because the breadth of the curriculum offerings in the high achieving schools has not needed to be cut back. Students in these schools are usually passing their state tests, their schools usually make adequate yearly progress, and their parents have the political power and resources to maintain a broader curriculum. These wealthier students, even were they to miss some exposure to the arts and humanities in the public schools, have parents who pay to provide them with extra curriculum activities (music lessons, drama club, sports), and they are much more likely to encounter the arts and humanities in their colleges. But poorer public school students may not be exposed to the ways of thinking embedded in the arts and humanities at all, and since their college attendance rates are low and getting lower at the most prestigious institutions of higher education (Gerald and Haycock, 2006), poorer students may never get adequate education in the arts and humanities.
Although teachers' voices are often dismissed, surveys of teachers reveal how the NCLB high-stakes testing culture affects the content of their courses. In Colorado teachers say (Taylor, Sheppard, Kinner & Rosenthal, 2003):
Another Colorado teacher says:
FAIR USE NOTICE
This site contains copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available in our efforts to advance understanding of education issues vital to a democracy. We believe this constitutes a 'fair use' of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17 U.S.C. Section 107, the material on this site is distributed without profit to those who have expressed a prior interest in receiving the included information for research and educational purposes. For more information click here. If you wish to use copyrighted material from this site for purposes of your own that go beyond 'fair use', you must obtain permission from the copyright owner.