Shiken: JALT Testing & Evaluation SIG Newsletter. Vol. 6 No. 3. Sep. 2002. (p. 11 - 13) [ISSN 1881-5537]
An Interview with Michihiro Hirai
by Tim Newfields
Michihiro Hirai served as director of the Hitachi Institute of Foreign Languages from March 1998 to June 2002,
after working as a computer engineer for Hitachi, Ltd. for 33 years. He received a master's degree in engineering
from the University of Pennsylvania in 1973 and gave a presentation on the correlations between productive and
receptive skill test scores at the recent language testing conference in Kyoto. In September a book summarizing
his views on English education will be published. This interview was conducted electronically in August 2002.
How did you become involved in language testing?
I have two different perspectives on language testing: as an English learner and as an English educator.
After completing my studies in the U.S. in 1973, I wanted to get some proof of my English ability,
so I took the most prestigious English qualification tests available at the time:
the Eiken First Grade exam
and the Ministry of Transportation-Approved Interpreter/Guide Test. Soon after passing these,
I was assigned to export mainframe computers worldwide. Serving as an intermediary between the marketing
and engineering sections of my company, I used English almost every day. However, it was not until turning
49 that a fervent interest in English qualification tests developed. At that time there were over 30
English examinations in Japan, reflecting both a popular craze and the needs of English training in industry.
During the decade that followed I took more than 30 English qualification tests and succeeded in eventually
passing nearly all of them. So, I have acquainted myself with the English testing scene in Japan fairly well by now.
In the meantime I was transferred to my company's internal language training center and started overseeing the
training of 400-500 employees a year. One thing I noticed was that the TOEIC® scores were often not a reliable
indicator of productive English skills. This prompted me to quantitatively analyze the correlations between
productive and receptive skill test scores as outlined in my article.
What language tests seem particularly well-suited to assess the English proficiency of Japanese engaged in business?
I'm an advocate of multi-dimensional language assessment for business environments. I prefer to use a radar
chart with five axes: grammar/vocabulary, reading, listening, speaking, and writing. This is because a survey
I conducted at Hitachi indicates that productive skills (speaking and writing) need boosting more than receptive
skills (reading and listening). Multi-dimensional assessment can be achieved by employing two or three different
tests, or alternatively by employing a multi-skills test with an appropriate breakdown of the sub-test scores
into four or five skill areas.
Of the English qualification tests commercially available in Japan which are designed to assess business language proficiency,
I personally recommend a combination of a Type A (which measures receptive skills), Type B (which measures speaking),
and Type C (which measures writing) in which Type A consists of a TOEIC or BULATS Standard,
Type B consists of a BULATS Speaking or SITE (Standard Interview Test of English),
and Type C consists of a BULATS Writing or Babel's Plain Business Writing.
Incidentally, Hitachi, Ltd., has recently adopted BULATS Writing as part of its personnel assessment system at my recommendation in addition to the TOEIC and a company-internal interview test.
For engineers and researchers, who tend to be left out when we talk about "business" and yet who constitute a sizable portion of business population, the two Type C exams mentioned above could be replaced with the Technical English Proficiency Test (TEP), which was developed jointly by Waseda University and the University of Michigan.
As an alternative multi-skills test, a BETA can be recommended.
Our readers may not be familiar with the BETA and SITE. Could you describe these tests in more detail?
The BETA (Businessmen's English Test and Appraisal) test was created by John Upshur and Randy Thrasher in 1976 for the
International Language Centre (ILC), a language school based in Tokyo. The BETA Test consists of two steps.
It is a comprehensive general business English test, and I personally think highly of it. It is very well designed,
although it has not enjoyed a big commercial success. ILC has been administering BETA practically on an on-demand basis.
Though the BETA was listed in the 1993 edition of Sanshuu-sha's Booklet of English Language Proficiency Exams,
it is not listed in the 2002 edition.
The SITE (Standard Interview Test of English), created in 1998, focuses on oral communication skills through a
one-on-one interview. It evaluates not only linguistic ability of the test-taker, but also non-verbal aspects
such as gestures and facial expressions. The interview, which is videotaped and evaluated by two independent raters,
consists of two parts: a "free talking" phase with typical questions and answers for about 9 minutes and a role play
phase for about 6 minutes. The SITE test is an offshoot of some research conducted under the auspices of the former Ministry of Education.
You mentioned that testing in Japan can be considered both as a business and to some degree a craze. Could you expand on those comments?
The Japanese are a people who respect authority and value qualifications/certificates that give authority,
probably because we live in a very competitive society. There are a large number of qualification tests administered
in a variety of fields (not only languages) in Japan today. So, Japan is inherently a good market for testing businesses.
Now in English alone we have about 40 qualification tests - it's hard to track down the exact number.
From the test provider's perspective, however, it is not very easy to be profitable in the English testing business because
business volume counts. ETS and/or the Chauncy Group should be making huge profits because of their sheer volumes and because the computerized versions of the TOEFL and TOEIC cost very little to administer.
Another biggie is the Eiken, which drew about 2.97 million test takers (all the levels combined) in 2001.
Like in any other business, you must achieve a "critical mass" and find ways to cut cost through computerization in order to
be profitable. It is in these aspects that American test providers such as the ETS excel and thus dominate the testing market.
I believe British tests such as Cambridge and Oxford are superior in quality (thoroughness and coverage/scope), but unfortunately
British test providers are not as aggressive and market-oriented as their American counterparts.
Can you think of any reasons why the Eiken exams have become so prevalent in Japan?
The Eiken test was created many years ago (in 1963, I believe), and was probably one of the very few English qualification
tests at the time, alongside the Interpreter/Guide national exam. It was out of a good intention of the Ministry of Education
(well, everything coming from the government is supposed to be good <g> ), meaning it was aimed at providing an authoritative
yardstick of English proficiency. It was designed, I believe, to test and certify the candidate's level of mastery of school English,
i.e., the English taught in Japanese schools. Baptized and indirectly supported by the Government, no wonder it became very prevalent.
Another factor that has contributed to its wide-spread acceptance is the lack of alternative tests of comparable authority,
awareness, and momentum, at least until recently. (I am pointing to the recent trend away from the Eiken in favor of the TOEIC in
industry. This may trigger a change of tide in higher level educational institutes sooner or later.)
A third factor is that a growing number of universities and colleges in Japan are adopting Eiken qualifications as one of the
requisites for acceptance or exempting holders of higher Eiken levels from the English test in the entrance exam.
Are there any trends in the language testing field in Japan which particularly concern you?
There are two trends which might seem contradictory in some ways, but are curiously related.
First, there is an astonishing degree of diversity in the tests available. Compared with the handful of
English tests that we had in Japan in the 1970s, now there are about 40 English qualification tests anybody can take.
Test taking is a big industry in Japan and it is surprising how many tests there are.
The second trend may seem antithetical to this: there is an increasing dominance of the TOEIC in many fields.
To my regret, this diversity of tests available - which is a welcome phenomena - is not well understood or exploited by
Japanese consumers. Perhaps many Japanese do not feel comfortable with too much diversity. At any rate, today the TOEIC
is dominating the English testing market. The TOEIC is a good test, but only for the domains it covers (receptive reading
and listening skills) and scope (general office level English rather than actual business English). There is nothing wrong
about the TOEIC as long as one clearly understands its limitations. Unfortunately, most Japanese do not. What concerns me
most is the simplistic apotheosis of the TOEIC, which naively equates English proficiency with TOEIC scores.
If you had the power to change one thing about the way testing is conducted in Japan, what would that be?
I would like to enforce multi-dimensional testing. It is ridiculous and even harmful to rely on only one yardstick
such as TOEIC with a limited coverage and scope (receptive skills only; the language limited in general office environments),
while what we need most in Japanese industries is productive skills in actual business situations.
This autumn you are publishing a book which summarizes many of your beliefs about English teaching in Japan.
Could you tell us a little more about that book?
Through several decades of experience as an English learner and four years as an
English educator, I have not only accumulated a number of tips but also crystallized some ideas about English learning and teaching.
Since there are not many books published combining these two perspectives, I thought it worthwhile to sum up and share my ideas with
the general public. The book, titled [Maximally Efficient Approaches to Studying English
for Businesspersons] will come out shortly. It starts with an introductory chapter highlighting some of the differences between school
English and real-life English, followed by my views on English language assessment and how Japanese businesspersons should study English.
The bulk of the book is devoted to the learning aspects and goes skill by skill.
Hirai, M. (Sep. 2002). Bijinesu paasonn no tame no eigo choo-kouritsu benkyou-hou. [Maximally Efficient Approaches to Studying English for Businesspersons]. Tokyo: Nihon Jitsugyo Shuppan. [ISBN: 4-534-03467-9]
Sanshuu-sha Henshuu-bu. (2002). Eigo no shikaku wo toru magajin bukku 2002. [Booklet of English Language Proficiency Exams -
2002 Edition]. Tokyo: Author.