ELT News Think Tank

This Month's Think Tank Panel

Marc Helgesen

Peter Viney

Curtis Kelly

Chris Hunt

Panelists: Marc | Peter | Curtis | Chris
Date: November 2003
Discuss this topic on our Message Board.

Topic: "How important is grading and testing?"

Peter Viney

Can I start by quoting children's author Philip Pullman on "the brutal and unceasing emphasis on testing and marking" that is destroying the British education system? I know so many good teachers who are desperate to get out of the profession because of the mountain of bureaucracy which is based on grading. "More money for education" sadly means more civil servants, more bureaucrats and an even bigger mountain of forms to complete and tests to administer. Secondary education's loss is often ELT's gain.

Philip Pullman (who is tipped as the next Tolkien when the films come out) was a teacher before he was an author. He says:

"I recently read through the sections on reading in key stages 1 to 3 of the national literacy strategy, and I was very struck by something about the verbs. I wrote them all down. They included:

reinforce, predict, check, discuss, identify, categorize, evaluate, distinguish, summarise, infer, analyse, locate... and so on: 71 different verbs, by my count, for the activities that come under the heading of 'reading.'

And the word 'enjoy' didn't appear once."

(Philip Pullman: 'Lost The Plot' in The Guardian Education, September 30th 2003)

So from reading to language teaching. My long-time co-author, the late Bernie Hartley had strong views on testing and language teaching. He believed that testing had an invariably negative effect in language classes, because languages are unlike most other subjects in the curriculum. In most subjects, you gain knowledge through the medium of language. Language is the medium, not the aim. But in language teaching the aim should be stimulating language production, not imparting knowledge about language. The related areas in the curriculum are the initial teaching of reading and writing skills.

“We were always reluctant to test at the early stages, because testing creates an attitude to language - that it consists of facts to impart and test.”

So how do you grade except by performance? Is it fair to take an oral approach, then test students in writing?

In the seventies, we used to administer complex ARELS Oral exams, which were conducted in a Language Laboratory and the tapes were then sent away to be marked. The exam was one of the hardest things I've ever had to administer. You had to control a thirty-booth lab, and the teacher had to get everything right first time. But I feel it was an excellent measure of student language performance. I tried to use some of the ideas in listening exercises later.

There was always one where the student was the 'third person' in a dialogue between two angry and authentically fast speakers. At intervals, they'd ask the student, "So what do you think?" or "Don't you agree?" there'd be a 'ting' on the tape and the student had to respond. There were very open activities like telling a picture story. However, years of experience in labs indicated to me that about 10% of the population freeze when confronted with the technology, and some people can neither understand audio exercises nor operate the controls. The lab exam ameliorated this somewhat, because you switched the lab to total teacher control over mics and tape, but nevertheless some people simply could not speak to a machine.

Unlike many British test writers, I am a multiple choice fan. My own kids took British A-levels, but my oldest son wanted to study in the USA, so took American SATS and SATs II as well. He reckoned they were not only more testing, but a better measure than the traditional A-level essays. However, I'd add that SATs II in English is about a hundred years out of date, and that many of the 'correct' grammatical answers are only 'correct' to an anally-retentive prescriptive grammarian who's read no works on grammar written in the last fifty years.

We were always reluctant to test at the early stages, because testing creates an attitude to language - that it consists of facts to impart and test. We had to bow to the demands of teachers for grades though.

People tell me that constant short tests are motivating. They tell me they were motivated by it themselves and their own very bright offspring are motivated by it. I always say, "And were you in the top 10%?" The answer is always "yes." It may be fun to fight over whether you're second or third in a class of fifty. It's boring to contest 24th and 25th place. It's downright depressing to contest 47th and 48th. It's not an even playing field either. Next time, and next time and next time people won't have shifted many positions in the class ranking.

I'm also convinced that foreign language ability is not related directly to IQ. I think it's quite a separate intelligence among multiple intelligences. I've taught very intelligent people who are awful linguists and other people who are not obviously 'bright' who can pick up languages easily and cheerfully. One of the fastest progressing groups I ever taught was on a Catering and Tourism Course and consisted mainly of waiters, with a few hairdressers thrown in. Actually, I find that very academic people often have extra problems because they can't express their intellect at the early stages, and are unwilling to converse about menus, shopping and other seeming trivia. They're also used to being in the top 10% and are horrified to find themselves in a different situation. Some have compared early language class tests to the driving test, where high intellect is of little help and may be a hindrance.

If you do have to test students, you don't have to rank them. Test feedback can be vague and imparted privately - but the top 10% will want to know. Do you remember coming out of exams at school or university? There were always a couple of excited hyperactive individuals who actually wanted to TALK about the thing. They always claimed to have done terribly. They always ended up coming top.

When we designed tests, we designed 'high-score' tests which Bernie called motivation tests, a word that publishers singularly failed to adopt. They enable the teacher to grade a class, but for most classes a very low mark will be 50% and a high mark will be 95%.

Panelists: Marc | Peter | Curtis | Chris

Discuss this topic on our Message Board

Peter Viney, Freelance ELT Author

Co-author of New American Streamline & Grapevine. Peter's Web site

<<Back Number | Top | Recent Issue>>