NOTE: The article below is mirrored from the JALT Testing & Evaluation SIG website.
PDF SHIKEN: JALT Testing & Evaluation SIG Newsletter Vol. 5 No. 1. April 2001. (p. 11 - 12) [ISSN 1881-5537] PDF

Insights in Language Testing: An Interview with Randy Thrasher

by Tim Newfields

Photo of Prof. Randy Thrasher, c. 2001
Randy Thrasher received his B.A. from Duke University in history in 1959, then a M.A. in English language from Michigan University in 1964 and a Ph.D. in Linguistics from the same school in 1974. He has taught at Okinawa Christian Junior College (1959 - 1962), Kwansei Gakuin University (1965 - 1980) and since 1980 has been at International Christian University. He is a founding member and currently vice-president of the Japan Language Testing Association. He worked under Jack Upshur in the Michigan English Language Institute's Testing Office and has published numerous books, tests, and teaching materials. This interview was conducted electronically in March-April 2001.

Q: Could you describe your current research interests?

A: I'm presently working on developing computer delivered English for Special Purposes tests and looking at issues such as possible test taker differences between pencil and paper and computer based tests, ways of delivering productive tests via computers, and score reporting methods that give a fairer picture of test taker proficiency. I am also interested in using Sperber and Wilson's Relevance Theory to develop a more suitable model of communication that can be used in language teaching and testing.

Q: Have you noticed any test taker differences between pencil and paper test and computer based tests?

A: We are just beginning to collect the data and it's really too early to make any claims, but we have noticed that the ability estimates stabilize more quickly for grammar items than they do for vocabulary items in the computer adaptive format. After about 20 items the grammar ability estimates stabilize, but it requires more than 30 vocabulary items before the same level of stability is realized.

Q: You mentioned Sperber and Wilson's Relevance Theory. Could you briefly outline what is and suggest how it might be important for language teachers?

A: Language teachers have been engaged in what we call the Communicative Approach for over 20 years now but we don't have an adequate theory of communication. I believe that Relevance Theory might provide the lacking theoretical underpinning of what we are trying to do in the classroom. Sperber and Wilson argue that what is encoded (the words we utter) is only a small part of communication. The larger part is the inferential device that fleshes out the bare bones linguistic utterance. The inferential device must both assign references to pronouns and specify the exact meaning of words such as 'now', 'soon', and 'already' as well as utilize background knowledge to recover the intended message. It is this second aspect of the inferential device that we language teachers need to think about. In preparing for my course on American English, I found the following in a letter from Emily Dickinson to her friend –
Is not an absent friend as mysterious bulb in the ground,
and is not a bulb the most captivating floral form.
Must it not have enthralled the Bible,
if we may infer from its selection?
"The lily of the field."
I never pass one without being chagrined for Solomon.
When you think what is involved in making sense of that last sentence you begin to realize the importance of Sperber and Wilson's point.

[ p. 11 ]

Q: What changes have you noticed in the language testing field since you first became involved?

A: I've been at it for 40 years now so it is practically a different field. When I started Lado had just published the first book devoted to language testing and the Michigan Test of English Language Proficiency was 'the' measure of language ability for college entrance. I've seen the development of the TOEFL® (actually wrote vocabulary items for the first form), lived through the cloze and noise testing craze, and am now looking for the light at the end of the task based testing tunnel.

Q: You are active in the Japan Language Testing Association (JLTA). Could you mention what projects the JLTA is currently planning?

A: One of JLTA's goals is to help language teachers better understand testing and to upgrade their test writing and analysis skills. So we are planning a summer workshop for junior and senior high school English teachers. And we are trying to develop a Code of Practice for language testing . . .

Q: Could you mention more about that?

A: Both ILTA and JLTA are working to prepare a Code of Practice based on the ILTA Code of Ethics. Some ILTA members have wondered if a universal language testing Code of Practice is possible, so JLTA has agreed to try to draft a code that would be appropriate for Japan and an ILTA committee headed by Lyle Bachman are trying to draft a similar code. We hope that the fruits of the labor of the two committees will show whether a universal code of practice is in fact possible.

Q: Why do you feel a Code of Practice is necessary?

A: Such a code is essential if we are to become responsible language testers. A code of practice would put pressure on everyone from the Ministry of Education and the National Center for University Entrance Examinations [Daigaku Nyuushi Sentaa] to junior high school teachers putting together their entrance exam to examine what they are doing and be open with the test takers and other stakeholders.

Q: What changes would you like to see in language testing in Japan in the future?

A: The greatest need is to become more professional. We do a huge amount of testing, but usually never analyze the results or run item analyses. Few people ever publish the standard error of measurement of their tests or conduct validity studies. This is why a testing code of practice is needed. We need to hold the various testing bodies and those of us who are test writing teachers accountable for the quality of the tests that are produced.

References and Further Reading

Japan Language Testing Association. (2007). The JLTA Code of Good Testing Practice. Retrieved on July 28, 2007 from

International Language Testing Association. (2000). Code of Ethics for ILTA. Retrieved March 10, 2009 from

Sperber, D. & Wilson, D. (1995). Relevance: Communication and cognition. (2nd ed.) Oxford: Blackwell.

Sperber, D. & Wilson, D. (1997). Remarks on Relevance Theory and the social sciences. Multilingua 16. 145-151. Retrieved March 10, 2009 from

Wilson, D. & Sperber, D. (2004). Relevance Theory. In G. Ward and L. Horn (Eds.) Handbook of Pragmatics. Oxford: Blackwell, 607-632. Retrieved March 10, 2009 from

Newsletter: Topic IndexAuthor IndexTitle IndexDate Index
TEVAL SIG: Main Page Background Links Network Join

[ p. 12 ]