Introduction

More than 30 years ago the traditional TOEIC LR test was conceived in Japan and created by the Educational Testing service (ETS), an institution for developing and administering tests to measure the proficiency of English as a foreign language. ETS was asked to create a "test system which accurately measures communication skills in English, give standards as to where the skills are in terms of expertise and occupation, and where they are in relation to objectives and goals" (Mitsuhashi, 2010). Additionally, the test content was to be able to measure a wide range of abilities from beginner to advanced level. The results were to "allow the test taker to measure his or her own skills, while at the same time allowing companies or schools to utilize the scores for making many important decisions" (Mitsuhashi, 2010). ETS was also asked to design a test that would be "capable of permitting test taking by large numbers of people at once" and to be "capable of being given relatively affordable and conveniently" (Mitsuhashi, 2010).

The test created under these conditions consists of two separate sections, listening and reading. The test,which takes 2 hours, has two sections each with 100 questions for a total of 200 questions. The score for each section ranges from 4 to 495, with totals ranging from 10 to 990.The TOEIC LR test was created at a time when corporations still had not much need for employees with high levels of speaking and writing skills. Because of this, ETS decided not to directly test speaking and writing abilities. ETS did carry out validity tests to examine the correlation between the TOEIC Reading/Listening score and a separate speaking/writing test. They found a correlation rate of 0.83, indicating that the LR TOEIC test scores could be used to predict a test taker’s speaking and writing abilities (Mitsuhashi, 2010).This then raises the question that if the LR TOEIC test is an accurate test of a test taker's speaking and writing abilities why did the ETS introduce the TOEIC Speaking and Writing tests (TOEIC SW tests), in January 2007? One reason is that although the correlation between a test taker's TOEIC listening and speaking score, and reading and writing score, is quite high (0.83) it is not a perfect match (1.0), so it could not be concluded that that for some test takers the difference between their listening and reading abilities, and their speaking and wiring abilities may be statistically significant.

In addition, there was an increasing demand by companies and universities for a more direct and accurate measurement of speaking and writing skills due to the increased use of English in the workplace and because of the advancement of technology such as e-mail and video conferencing.

One of the main differences between the TOEIC LR and TOEIC SW tests is that the LR test is normreferenced which indicates the position of a test taker relative to other test takers. In contrast, the SW test is a criterion-based test, in which the test taker is evaluated on how well a set of certain standards have been met. Therefore, the two tests are qualitatively different. The IIBC (The Institute for International Business Communication) released information regarding the statistical relationships between the TOEIC LR and the TOEIC SW. This information is shown in Table 1.

Table 1 Statistical relationships between the TOEIC LR and the TOEIC SW tests
Level (LW) L R S W
5 (860-990) 430-495 430-495 150-200 160-200
4 (730-860) 365-430 365-430 130-150 140-160
3 (470-730) 235-365 235-365 100-130 110-140
2 (220-470) 110-235 110-235 70-100 70-100
1 (5-220) 5-110 5-110 0-70 0-70
(http://www.toeic.or.jp/toeic_en/about/)

Because the TOEIC LR has been around so long, it is well known and respected by employers and teachers as a measure of a test taker's proficiency level. The type of questions, the scoring system and proficiency levels are widely understood and there is a multitude of books giving advice on how to achieve better results.
The same cannot be said for the TOEIC SW test. For this reason, what follows is a detailed description of the test itself, how it is administered and how the scores relate to a test taker’s proficiency level.