Skip to content
ascd logo

Log in to Witsby: ASCD’s Next-Generation Professional Learning and Credentialing Platform
March 1, 1995
Vol. 52
No. 6

Exam-Based Reform Stifles Student Writing in the U.K.

As many Americans look to high-stakes exams as the road to school reform, the British experience shows that that road may lead away from, rather than toward, our destination.

In British secondary schools, high-stakes national examinations, based on high standards for teaching and learning, have been in place since the beginning of this century (Freedman 1994). In fact, as Madaus and Kellaghan (1991) point out, “Proposed American reforms to institute such examinations seem to have been inspired by, if not actually modeled on, the British system.”
In 1986, I joined a British colleague, Alex McLeod of the University of London Institute of Education, to begin observing the British system of secondary school exams close-up. What we saw was depressing. In classes preparing for the English language and literature exams, the high stakes, and the teachers' and students' perceptions of the kind of writing that was appropriate for the exams, undermined students' work and attitudes in both obvious and subtle ways. While it isn't easy to transport findings from one cultural context to another, we in the United States can still learn from the British experience.

U.S. Enthusiasm

As U.S. policymakers and educators contemplate national exams, hopes are high that the right kind will lead to improved instruction. America 2000 (1991), the blueprint for educational reform produced by the National Governors Association and spearheaded by then Governor Clinton, calls for voluntary, national, high-stakes achievement examinations in core subject areas for grades 4, 8, and 12, based on high national standards and goals. The scores eventually would be given to potential employers or used for college admission (see also Cheney 1991, National Council on Education Standards and Testing 1992, Simmons and Resnick 1993, Tucker 1992).
Consistent with the current federal agenda, Warren Simmons and Lauren Resnick (1993) suggested creating “world-class content and performance standards, a performance-based examination system that embodies those standards, together with rubrics and procedures for scoring students' work reliably and fairly.” “Our goal,” they said, “is to build a revitalized education system using assessment as a tool for transforming instruction and learning.”
The question is how best to achieve our goals. During our observations in 1987 and 1988, Britain's national examinations were, in every way, what we in the United States are striving for in our most ideal examination system. Yet the evidence suggests that a system of high-stakes examinations—even well-designed performance-based examinations—provides a flawed foundation on which to build a national educational reform movement. In fact, exams have the potential to move us away from, rather than toward, the point we all want to reach.
A stronger foundation for reform likely will come through working with teachers and school administrators to rethink and then reshape both the curriculum and the organization of the school. Testing, then, would follow from, not lead, the reform effort.

The Study

Beginning in 1986, we distributed a questionnaire to 695 teachers and 702 of their secondary students in both the United States and Great Britain (England, Scotland, and Wales). We also observed eight English language arts classes—four in England and four in the United States—at grade levels equivalent to 6–9.
The case study classes were paired by grade level—one in the United States and one in England. They then exchanged their writing for an entire academic year. The American and British teacher in each pair coordinated their curriculums, so that their students were doing roughly the same kinds of writing at the same time. Although the students did send personal letters back and forth, the main focus was on major and substantive pieces—autobiographies, books about school and community life, opinion essays and essays about literature. Research teams in both countries observed the classrooms and interviewed students and teachers about the ongoing instructional process.
In selecting the teachers to survey and the classes to study, we sought to choose those that represented the most thoughtful current practice in their respective countries. The American schools involved in the exchanges were in largely urban, working-class settings in the San Francisco Bay area. In England, the schools were also in working-class areas and, unlike their American counterparts, the classes included many students whose native language was not English.

The British System

Two of the four British classrooms were 9th grade equivalent and were in the first year of the two-year examination course for English language and English literature. (They were paired with two 9th grade remedial classes in California.) In England during the 9th and 10th grade years, the entire curriculum, in every subject, consists of an examination syllabus that prepares students for a pivotal set of exams, the General Certificate of Secondary Education (the GCSE). Students usually take these exams in six or more subjects. (Besides English language and literature, they may take, for example, mathematics, geography, drama, Spanish, French, Russian, physics, chemistry, biology, and so on.) They are normally taught by the same teacher for both years of each course.
These exams determine the students' futures. Those with poor results (about 70 percent) generally leave school. Those with good results continue in school to prepare for university or other postsecondary education, or leave with a fair chance of employment.
Unless they are enrolled in a specific vocational program, students who continue in school take an additional two-year examination course to prepare them to take yet another set of exams. These, the A levels, determine university admissions. Only about 20 percent of these students, or less than 6 percent of the nation's 18-year-olds, actually complete the A level course and take the examinations (Statistics of Schools 1989).

Testing by Portfolio

In 1987, England's examination system had just been revised. It was the first year that all secondary students took the same courses and exams; before then, students had been separated into two exam tracks: one for the university bound (O level) and the other for the non-university bound.
Also at that time, schools were given the option of substituting a portfolio of writing for the English language and literature exams. The examining board did require that students write at least one of their portfolio items on an impromptu subject under controlled or testing conditions. But the student and teacher could select the rest of the material to represent the student's best efforts and to show a range of knowledge and skill.
In exam portfolios we collected, we found answers to questions about literature and imaginative writing based on literature; expository essays about literature or current events; and sometimes original fiction, an autobiography, or other personal writing. In every sense, this was what we in the United States call performance assessment.
As for scoring, first a committee of teachers at the student's school, and then a committee from neighboring schools, graded the portfolio as a whole, under the direction of teachers who coordinated school marking with standards set by a national board. The portfolios were then sent to the national board to be checked for consistency with national standards. When committee members spotted inconsistencies, they re-scored the portfolios.
Unlike the U.S. classrooms we studied, and unlike the two groups of younger British students who were not preparing for exams, the exam classes had to adhere to requirements that inhibited the teachers' abilities to build a coherent curriculum. They also restricted the amount and kind of writing the students did. The teachers of these exam classes, Philippa Furlong and Gillian Hargrove, found that, contrary to their theory about how students learn to write, they inadvertently began to take away their students' responsibility for the subject matter when their students wrote for the exam audience.
  1. Allen and Rebecca meet 10 years later. Maybe you can change the balance of power between the two of them. What are they doing now? How do they remember that summer? Does Rebecca feel guilty about cutting Allen? Does Allen feel hurt or angry? How do they feel about segregation? How have they changed and developed as people?
  2. Write an alternative ending to the one in the book, or a further chapter. Maybe Allen rejects Rebecca, maybe the parents try to intervene and Allen and Rebecca try to resist them. Whatever you choose, your new ending has to be consistent with the characters as they appear in the book.

Performing for the Tester

When I began working with Alex McLeod and British teachers to plan the exchanges, we all assumed that performance examinations requiring a variety of types of writing would be entirely compatible with the goals of the exchange and the same writing could be used for each. We were wrong. In effect, the pressure of the examinations took over.
Here's how one student, Surge, explained his predicament: Everything has to be perfect, and you get so much coursework.... It's just so confusing, and you get really frustrated sometimes. And like the teachers will either end up getting in arguments with the class or the class will end up getting in an argument with the teacher, but I think it's only because what some teachers don't understand is that we get so much coursework, like from English, math, history and all that, and it's got to be in on a certain date, and it's just hard to bring it all in at once.
Furlong explained her students' inhibitions in writing for the examiner: The kids didn't feel that confident about really baring their souls in an assignment that was going to go off to an examiner even though they knew that I was one of the examiners. They knew that beyond me there was an unknown quantity.
One student of Furlong's, Andi, confirmed her assessment in a slightly chilling way: They [the examiners] are not people at all ... when you are marking a GCSE paper, then it's your future, which is serious.
In the end, Furlong's students decided that most exam writing was just for the examiner and would be of no interest to anyone else, including the students in the United States. Her students' writing for the exams was rushed, and extended pieces that were common in the early years disappeared.
By contrast, students worked hard to connect with the exchange audience. As one young woman said, students' writing has to be as interesting as they can make it: You had to sort of build up a personality for yourself so they could imagine what you were like... And you had to make it slightly longer. Across all eight classes involved in the writing exchanges, we found that students were most committed to their writing when they worked with their teacher to choose the subject. They were motivated by their own decision making and also by being part of a community working together. When writing for the exam, however, this happened rarely. In addition, there was a spirit of sharing when the students wrote for the exchange. Overall, the American students who were paired with the exam classes seemed to be more engaged in their writing and also produced a greater range of writing types.

From Bad to Worse

Besides the pressures of the required content, high stakes, and heavy workload, the British examination restricted the teachers in how they could help their students. By the time the 1994 exam syllabus was developed, the Northern Examinations and Assessment Board felt the need to be explicit about how much help teachers could provide: Advice remains on a general level, only becoming specific to exemplify general comments, and the onus is left on the candidate to incorporate the teacher's general advice by making specific alterations and thus submitting a final draft. This level of advice is acceptable, whereas proofreading, where the teacher points out a detailed series of errors, omissions, and amendments for the candidate to correct in the subsequent draft, is inadmissible.
Such policies have inhibited British teachers from the kind of explicit teaching needed to reach the neediest students, a serious drawback, considering the lower-scoring students make up some 65–70 percent of those who take the exams. As Hargrove notes, the exam requirements are especially problematic for teaching bilingual writers, who may learn from specific corrections and by having a series of errors pointed out in detail.
Today, the British examination system has grown even more conservative and many teachers feel it is deteriorating. The emphasis in English has shifted from portfolios or coursework to “terminal” exams. The British government feared that with portfolios only, it was difficult to know what students could do independently. Further, the whole exam system is in constant motion, a kind of political football that goes hither and yon in response to government policies.
Ultimately, the British teachers thought issues such as good curriculums, good teaching, and high performance standards were quite unrelated to the national tests or even to the national standard-setting efforts. Meanwhile, the U.S. teachers left the experience convinced that any kind of high-stakes examinations with associated curriculums would harm their students' writing development.

Teaching, not Tests

What happened in the British exam classrooms suggests that the path to curriculum reform through examinations, though tempting, remains elusive. Exams are popular among policymakers because they provide one of the few levers on the curriculum that policymakers can control. And there are some potentially positive side effects, especially the substantive professional debates about standards, and the professional development borne of creating and scoring performance-based exams.
When national exams take control of something as personal as writing, however, a distant examiner, rather than the teacher and students, end up owning the writing. Worse yet, when the exam is used to sort students in ways that determine access to higher education and to valued job opportunities, the high stakes put formidable pressures on the classroom and the exam is particularly apt to have a negative effect on the curriculum.
Darling-Hammond (1994) argues that any exams used to sort people discourage educational equity and fail to promote school reform. She maintains that portfolios and performance assessments won't help unless we also change the ways in which we use assessments: ...from sorting mechanisms to diagnostic supports; from external monitors of performance to locally generated tools for inquiring deeply into teaching and learning; and from purveyors of sanctions for those already underserved to levers for equalizing resources....
Similarly, George Madaus (1993) concludes that the nation cannot test its way out of its educational problems...it is the teachers, not tests or assessments, that must be the cornerstone of reform efforts. Many maintain that reformers must begin by working collaboratively with schools and the communities they serve, involving teachers, administrators, students, and parents—for example, James Comer's School Development Program, Henry Levin's Accelerated Schools, and Ted Sizer's Coalition of Essential Schools.
Admittedly the reach of these efforts remains limited. None attends equally well to every aspect of the problem, and none can begin to solve the larger social problems that plague schools serving high percentages of students from conditions of poverty. Still, these reformers are taking sensible first steps. They offer a welcome counterpoint to those who advocate reform via exam.

America 2000: An Education Strategy: Sourcebook (ED/0S91-13). (1991). Washington, D.C.: U.S. Department of Education.

Cheney, L. (1991). National Tests: What Other Countries Expect their Students to Know. Washington, D.C.: National Endowment for the Humanities.

Darling-Hammond, L. (1994). “Performance-Based Assessment and Educational Equity.” Harvard Educational Review 64, 1: 5–30.

Freedman, S. (1994). Exchanging Writing, Exchanging Cultures: Lessons in School Reform from the United States and Great Britain. Cambridge, Mass.: Harvard University Press.

Madaus, G. (1993). “A National Testing System: Manna from Above? An Historical/Technological Perspective.” Educational Assessment 1, 1: 9–26.

Madaus, G., and T. Kellaghan. (1991). Examination Systems in the European Community: Implications for a National Examination System in the United States. Paper prepared for the Science, Education, and Transportation Program, Office of Technology Assessment, U.S. Congress, Washington, D.C.

National Council on Education Standards and Testing. (1992). Raising Standards for American Education: A Report to Congress, the Secretary of Education, the National Education Goals Panel, and the American People. Washington, D.C.: NCEST.

Simmons, W., and L. Resnick. (February 1993). “Assessment as the Catalyst of School Reform.” Educational Leadership 50, 5: 11–15.

Statistics of Schools, January 1988. (1989). London: Department of Education and Science.

Tucker, M. (Fall 1992). “Toward National Standards: Measuring Up.” America's Agenda: 21–22.

Sarah Warshauer Freedman has been a contributor to Educational Leadership.

Learn More

ASCD is a community dedicated to educators' professional growth and well-being.

Let us help you put your vision into action.
From our issue
Product cover image 195020.jpg
Aiming for Higher Standards
Go To Publication