Studies in America appear to confirm British fears about the adverse effect of testing regimes on the quality of education. David Budge reports.
AMERICAN research has confirmed what many British teachers have suspected but could not prove - increased emphasis on test scores can reduce the quality of school
education, even when results appear to be improving.
About 25 US states have adopted "high-stakes" testing in recent years. Students who do badly in the tests cannot move up to the next grade or graduate from high school. Teachers and schools are also rewarded or penalised on the basis of test results.
The tests' supporters argue that their introduction in states such as Texas and North Carolina has improved student performance and made schools much more accountable. As in Britain, school performance tables are now published in many newspapers.
But there is growing evidence that the test-driven approach to education reform has been
counter-productive in the US. Researchers contend that high-stakes testing has narrowed the school curriculum in poorer districts, stultified teaching practices and encouraged teachers and administrators to cheat. Some also maintain that apparent improvements in children's performance are actually due to the score inflation that stems from teaching to the test.
The findings will trigger alarm bells in Britain where national tests and the literacy strategy are said to be squeezing the time available for more creative work (see last week's TES).
One recent US study on high-stakes testing found that there is no clear relationship between the tests and learning gains. Indeed, analysis of 13-year-olds' performance suggests that states without high-stakes tests have made more educational progress.
The study's authors, Monty Neil of the National Centre for Fair and Open Testing and
Keith Gayler, of Harvard Graduate School of Education, reached this conclusion after scrutinising children's performance in the National Assessment of Education Progress, which records reading and maths scores.
Their study raises most doubts about Texas's testing programme. It points out that although Houston schools had improved their scores in the tests they had been drilled for - the Texas Assessment of Academic Skills - the city's students did less well in Stanford 9, a commercial test.
Neil and Gayler also believe that high-stakes testing in Texas may be further disadvantaging minority groups. "On the 1998 NAEP reading test, the score gap in Texas between whites and African-Americans has widened, the opposite of what has happened on the state's own test.
"It would seem that the results which have been most touted about the Texas experiment, the improvement among minority groups, is an artefact of teaching to the test."
A second study by Linda McNeil of Rice University and Angela Valenzuela of the University of Texas at Austin provides further evidence that the state's testing regime may be damaging minority children's education.
"White, middle-class children continue to receive an education appropriate for their age and grade level, while poor and minority children are devoting class time to practice test materials," they said.
McNeil and Vlenzuela also found that the TAAS system distorts educational spending, diverting "scarce instructional dollars" away from laboratory supplies and books towards test-preparation materials. One largely Hispanic, low-performing high school with a severe shortage of textbooks spent $20,000 (almost its entire teaching supplies budget) on a set of test-revision packs.
However, a few researchers have supported high-stakes testing. "Linda McNeil assumes there's some wonderful alternative out there," said Martin Conroy, a professor of education at Stanford University. "But what was happening before for those kids was probably even worse."
The papers by McNeil and Valenzuela and Neil and Gayler were commissioned by the Civil Rights Project at Harvard University. They will appear in 'Raising Standards or Raising Barriers? Inequality and High Stakes Testing in Public Education', to be published by the Century Foundation
TEXAS CURRICULUM MASSACRE?
TEXAS provides a prime example of how high-stakes testing can skew the curriculum in disadvantaged areas.
The pressure to do well in the Texas Assessment of Academic Skills leads teachers to spend several hours a week drilling pupils for the tests. Social studies classes are frequently suspended, according to researchers Linda McNeil and Angela Valenzuela (see report above). And even science and maths teaching is cut back.
In some schools the curriculum for nine-year-olds has been virtually replaced by daily practice of the five-
paragraph "persuasive essay" that figures in the TAAS test. Each paragraph begins with a "topic sentence",
followed by three "supporting sentences", and a "concluding sentence", which recaps the topic sentence. Adhering to this strict format is regarded as more important than the development of ideas.
One teacher told McNeil and Valenzuela that her children were merely "writing for someone sitting in a cubicle in a bureaucracy, counting sentences and indentations".
HIGH STAKES GAME IN NORTH CAROLINA
NORTH Carolina is one state that doles out rewards and punishments to teachers on the basis of their school's
performance in tests.
Teachers in successful schools can receive a $1,500 (pound;1,000) bonus but "failing" schools face sanctions. Teachers can be required to sit competency tests and
principals and staff who are "not willing to improve their practice" are sacked.
As in Texas, an inordinate amount of classroom time is also devoted to test revision. One University of North
Carolina survey of 236 elementary teachers in five school districts showed that the vast majority (80 per cent) spent at least a day a week practising for end-of-year tests. Even more worryingly, 28 per cent of the teachers admitted they set aside more than three days a week for such work.
'Running with scissors: the impact of high stakes
assessment on teachers and students', by M Gail Jones et al, University of North Carolina, Chapel Hill. * Education researchers who wish to disseminate their findings in The TES should send summaries of no more than 750 words to David Budge, Research Editor, The TES, Admiral House, 66-68 East Smithfield, London E1W 1BX. Tel 020 7782 3276. E-mail email@example.com