Testing Business English: an interview with Barry O'Sullivan

Shiken: JALT Testing & Evaluation SIG Newsletter
Vol. 10 No. 1. Mar. 2006. (p. 20 - 21) [ISSN 1881-5537]

Testing Business English: an interview with Barry O'Sullivan

by Daniel Dunkley (Aichi Gakuin University)

Photo of Barry O'Sullivan, c. 2005

Barry O'Sullivan has a PhD in language testing from the University of Reading, U.K, and is particularly interested in issues related to performance testing, test validation and test-data management and analysis. He has lectured for many years on various aspects of language testing, and is currently acting director of the Centre for Language Assessment Research (CLARe) at Roehampton University, London. He has previous experience teaching in Ireland, England, Peru and at Okayama University, Japan.

Barry's publications have appeared in a number of international journals and he has presented at international conferences around the world. His book Issues in Business English Testing came out in January 2006 and his next book will appear in Autumn 2006. Barry currently works with several government ministries, universities and test developers in Europe, Asia, the Middle East and Central America.

This interview was conducted in person in the U.K. in March 2005. It was made possible by a research grant from Aichi Gakuin University.

Q: Let's start with a very basic question: What do you mean by 'Business English'? Is it really one entity?

A: This is a very good question: in fact it is the basic question in English for Specific Purposes. Now, some people will argue that it doesn't even exist because it doesn't have sufficiently clear boundaries. They'll say that if you try to test language within a specific context you get "leaking" of other skills from the context into the language. So it's very difficult to actually define ESP.

Let's take an example. Since the middle ages people have come to the U.K. from the continent, particularly Germany, because of their interest in British cloth. German merchants would send their apprentice cloth-makers to the U.K to learn their trade and also to learn the language of that trade. We also know that international trade was common in the ancient world, so we can presume that language was a relevant factor back then.

Q: So from that point of view, Business English does exist. But how is the language of Business English really different from general everyday English?

A: From a lot of work done in the 1960s, and into the 70s and 80s the notion of "business language" or "specific purpose language" as precise language, to use Dan Douglas's phrase, can be supported. There is an argument that in specific purpose areas such as business there are usages which are 'specific' from the syntactic and lexical perspectives as well as from the perspective of genre.

The difficulty in the theories of business language testing and Special Purpose testing has been in the misunderstanding that a SP test should only test the language of that specific domain. Now I don't believe that's possible, or that anybody does it. It may be possible to test or teach the language for a particular domain with very clear boundaries, but only in a tiny minority of areas. For example, the language of air-traffic controllers is a very controlled use of language.

Now, in regular specific purpose domains, particularly business, language is placed within a specific core-related directly to that domain, and that core is placed within the general domain of language use. So any test is going to have to look at both: the core and the general domain. You need to accept these three things: the need to look at both the core and the specific domain, the notion of specificity of tests (in which a test can be very highly specific or very general) and finally the idea that a test can be somewhere on that continuum.

If you agree with these three points, you have a theoretical perspective of specific purpose testing works. So for example, in my book I look at a number of Businesses language tests in various languages, and discover that tests vary from the quite specific to the more or less general. They all call themselves 'Business English tests' or 'Business Language tests', but there's a wide range of specificity both within each test, and certainly between tests. So I've suggested we should be looking at individual items and tasks from the perspective of general to specific continuum, as well as a number of different perspectives. And that's where I talk about the framework which Weir proposed in 2005, which calls for "context validity." We can use this in combination with the notion of test specificity to define a test and place it on a continuum. So for me there's a very clear theoretical argument for defending the teaching and testing of business English.

[ p. 20 ]

Q: Why do you feel the TOEIC^® has become such a widespread business English test?

A: I would love to know the answer to that one! I believe it is so popular partly because it's cheap and cheerful, partly because it's well marketed, partly because of the limited understanding of the people who use it, and partly because there is no 'simple' or 'cheap' alternative. As a tester, I have to admit to having quite negative feelings about that test.

Q: What areas of research are you working on now?

A: Now, I am still very interested in performance assessment – though I am beginning to broaden my horizons. I'm very interested in testing languages other than English, and have even worked recently on tests of other subject areas (science and mathematics). In the area of English language testing I'm focusing more and more on the testing of listening. This is a big move for me as I have been preaching for years that I hate testing listening as it's too complex. Maybe the challenge has proved too enticing!

Q: Let's turn to your book, Issues in Business English Testing. You've criticized other testing books for having too much "density of content." How is your book different?

A: I hope that I've written it in a way that's as accessible as possible. This is because it's written for an audience that I hope will take it away and absorb some of the ideas. What I have done is to look at a brief history of Business language testing, and then taken an overview of current practice, in other words what is actually happening. It's based on the Cambridge examinations, and features a case study of the revision of the BEC (Business English Certificate) exam. I show how their principled approach led to change. But probably most importantly, when I talk about the theoretical side, I try to do it in a language that I feel people will understand.

Q: What trends do you notice in the field of Business English testing?

A: I see the most interesting trends taking us further away from the standardized norm-referenced format. Dan Douglas, argues that this format is unsuitable for specific purpose tests and I totally agree. I also see some very good work with integrated skills tasks, particularly with some of the City & Guilds Pittman tests. However, in some ways the picture is pretty bleak. Many tests out there are totally unsupported by any evidence of their true value, though this is not only true of business language tests I should stress.

Works Cited

Douglas, D. (2000). Assessing Languages for Specific Purposes. Cambridge: Cambridge University Press.

O'Sullivan, B. (2005). Issues in Testing Business English. Cambridge: Cambridge University Press.

Weir, C. (2005). Language testing and validation: An evidence-based approach. London: Palgrave.

NEWSLETTER: Topic Index

Date Index
TEVAL SIG: Main Page

Links

Join

http://jalt.org/test/osul_dun.htm

[ p. 21 ]