The Joy of Testing

Canadians intent on seeing reforms in classrooms here may want to look to England, where finely tuned
and exhaustively studied tests reveal everything from which ethnic groups aren't performing well on literacy to whether little Nigel has really mastered his latest algebra lesson

By Alanna Mitchell of The Globe and Mail
Saturday, May 22, 2004

When 1.8 million students sat down to write their standard curriculum tests in England this month, they set off a flurry of statistical analysis Canadians would find hard to imagine.
The test results will be combined to show how the education system is performing across the country and how each and every school fares against all the others.

As well, the stats will be sliced and diced to determine how boys do versus girls, how the rich do versus the poor, how one county does against the next, how city kids do compared with country kids and how members of various ethnic minorities perform. And once the gaps in the system have been identified, money
this is the part educators elsewhere truly envy is earmarked to help close them.

Contrast that with Canada, where the most comprehensive national statistics come from a random test of 50,000 to 60,000 students carried out each year. An increasing number of kids also sit down to provincial tests. Next week, for example, Grade 9 students across Ontario write standard math exams while those in Grades 3 and 6 wrap up provincial assessments in math, reading and writing.

But the results don't produce nearly as much detail, and Premier Dalton McGuinty has vowed to tear a page out of England's book to improve Ontario's system. This week's provincial budget made it clear that new money will go into education to boost achievement levels, a drive that may lead to more precise use of testing stats.

Like most provinces, Ontario generally sees testing not as a way to diagnose problems but as an incentive for teachers and students to bone up on curriculum. The national tests, such as they are, don't yield nearly as much information as the ones in England, where "they go to a level of detail that we don't in any jurisdiction," says Paul Cappon, director-general of the Canadian Council of Ministers of Education (CMEC).

English testing, conducted when students are 7, 11 and 14, can be used to produce specific reports. Each school can find out precisely how well its children are learning about, for example, geometry versus algebra, said Jackie Bawden, director of testing for the National Assessment Agency. If the school is falling behind in a specific subject, something can be done about it.

"What's really important to me is that the tests are used for informing the next stages of learning," Ms. Bawden said. "There is no point in doing assessment unless you are learning something from it and using the information."

It's the same idea for individual students. They receive personal reports on whether they have mastered certain parts of, say, multiplication, but are a bit weak on division. Again, the idea is to focus on whatever is not working well, and bring it up to expectations.

That is why Ken Boston, chief executive of England's Qualifications and Curriculum Authority, is passionate about conducting tests at set ages and making sure that everybody takes them. It's a way of ensuring that all students are treated equitably.

In much of North America, however, tests are seen as a way to "name, blame, shame." In the United States, for example, test results that fall below levels the federal government likes can cause funding to vanish, under President George W. Bush's new No Child Left Behind Act.

And the standard tests in Ontario were created shortly after former premier Mike Harris's initial education minister, John Snobelen, admitted that he wanted to "invent a crisis" to support reforms. That led critics to call the tests part of the ploy to raise anxiety. And in Alberta, the renewed focus on testing coincided with double-digit funding cuts to education and accusations that teachers were being overpaid.

England's tests also were designed to be shaming at first. Launched under former prime minister Margaret Thatcher, they are still derided by some teachers unions, which say they strip away teaching innovation and narrow the scope of what is taught. Just last year, the National Union of Teachers tried to organize a boycott of the tests, but failed.

Under Mrs. Thatcher, the government published the test results in "league tables," ranking schools as though they were soccer teams. If a school was crummy, parents were encouraged to vote with their feet and to transfer children to a better one.

The Thatcher government also instituted a system of grants so that children who could not normally afford private schools were able to attend one, in effect siphoning money away from the state system. As a result, bad state schools tended to get worse.

When Tony Blair took office in 1997, he made education reform the centrepiece of his domestic policy and became the first British PM to send his children to state schools.

He kept the national curriculum tests and the league tables, but he also ended the scholarships to private schools and injected billions into state schools. Some of the money is being spent across the system and some is devoted to schools whose students fare the worst in tests.

It is an approach Dr. Boston calls "strategic intervention," and his faith in it stems from a testing system he first set up in the Australian state of New South Wales. There, as in Canada, the results are neither trumpeted nor ranked nationally. Instead, each school produces a report for its own community.

However, the government decided to break out the results for aboriginal students and it soon became clear that they were doing much worse than the rest of their classmates. As a result, the state government set targets for aboriginal children, and dedicated research and resources to ensure that they received extra help.

What happened? According to Dr. Boston, the Australians realized that the tests were the first tool they had ever had that actually raised education levels among aboriginal children.

So now, even though profiling student achievement by ethnicity seems controversial to Canadians, it is generally accepted in England as a way of targeting resources effectively. According to a report on the subject published last year, "it allows us to be honest about the extent of the problem and to target support where it is most needed."

For example, last year's test results told the Department for Education and Skills that students of Chinese and Indian heritage do better than average, whereas black (be they of Caribbean or African descent), Pakistani and Bangladeshi kids do worse.

So the government has set up ethnic-minority achievement grants worth 155-million a year to be spent on extra teaching and resources for these children. As well, it is examining what practices work best at schools where, the statistics tell them, visible-minority children are doing just as well as everybody else.

In addition to all that, the challenge of making sure that all children from non-white backgrounds do well at school has become central to the government's education reforms.

Of course, the system isn't universally endorsed. In fact, it is still hotly opposed by some who contend that it strips the curriculum of depth, makes robots out of teachers, fails to honour complexity, diminishes dignity and, in the end, tests nothing relevant.

But standard testing is catching on in many industrial nations, and most Britons now agree that the principle behind it is worthwhile.

Dr. Cappon says Canadian education also is working to close the gap for aboriginal students, and CMEC research about six years ago sparked change by showing that francophone students outside Quebec fared worse than either francophones in Quebec or anglophones throughout the country.

But for now, at least, testing remains extremely limited. Provincial results are broken down by sex, sometimes by whether English is the first language, and by whether the student has special learning needs. But no checking is done to see whether students who are poor, new immigrants or members of a visible minority learn as efficiently everyone else.

In fact, Canada is one of few countries that does not sort education stats according to culture, ethnicity or even, Dr. Cappon noted, a person's health. A medical doctor, he worked on a project 10 years ago to determine how HIV affects different groups in society and could find no data according to ethnocultural community.

But divvying up the stats by race is not on the Canadian radar screen. "There is a lot of pride in Canada about how inclusive our systems are," he said. "Here, there's a level of confidence they've got it right."

Alanna Mitchell is a senior features writer at The Globe and Mail.

What low scores can lead to

They are descendants of immigrants from the Caribbean who settled in England as far back as the 1940s, and as a group they are doing remarkably poorly at school. Worse, in fact, than members of any other visible minority.

The basic story: The children start out strongly, only to falter badly at the age of 11 and do even worse by 14. At 16, fewer than a third produce a good showing, compared with more than half of white, Indian and Chinese students.

All this information has come from an analysis of curriculum-testing results, examined according to ethnic background. And it has led to sweeping reforms at a few key schools now being held up as models.

Most of the success at these schools revolves around shifting attitudes toward black pupils of Caribbean descent. Now, they are expected to do well; no longer is it assumed that they'll perform below average.

Racial awareness, says a report entitled Achievement of Black Caribbean Pupils: Three Successful Primary Schools, is the first step on the road to success. A system must "develop the confidence and sophistication of schools in approaching ethnic diversity," with staff, pupils and parents talking openly about any barriers to doing well at school.

Rooting out discrimination must be "central to the school's basic systems and approaches and made relevant to all the staff," the report concludes.

A shining example of this approach is Sudbourne Primary School in London's Lambeth district. Roughly one-quarter of its students are of Caribbean heritage, and the school also has more than its share of low-income students and students whose first language is not English.

Nevertheless, the national curriculum tests show that all of them now do much better than the national average and outside students are clamouring to get into Sudbourne.

It's a far cry from the situation in 1982, when Susan Scarsbrook became head teacher. Enrolment was dropping, and parents, especially those from the black community, were furious. Books used in the classroom didn't show ethnic minorities, and parents felt they had little connection with the school and weren't being listened to. Mrs. Scarsbrook took the criticisms to heart. She began consulting parents and together they figured out that regular communication between home and school was vital. As well, cultural diversity began to be talked about as a benefit and a source of pride. Discrimination was no longer even considered, let alone tolerated.

The school's statement of intent now states clearly that staff members "are opposed to racism in any form . . . because it is wrong, offensive and illegal." The students seem to be taking it to heart. "It's fair because the teachers will always listen to both sides of the argument, both points of view, and get to the bottom of things," one student told inspectors from the Office for Standards in Education. "They are not sexist or racist."