Beginning in Summer, 1998, students taking the Test of English as a Foreign Language (TOEFL) in many countries will leave their No 2 pencils behind and use a computer instead. This change is part of an evolutionary effort to create a new and better generation of English proficiency tests.
The transition of TOEFL to computer is part of a project called TOEFL 2000 which began in 1993. It reflects the beliefs of the TOEFL Policy Council and Educational Testing Service (ETS®) that the computer offers new opportunities for better English proficiency assessment that is more responsive to test takers and score users. Use of performance based questions will also provide schools with better information about an international student's ability to understand and use English.
The computer based TOEFL will be administered in four sections Listening, Structure, Reading and Writing. Most sections will have new or improved question types by using computer technology.
The Listening section will continue to measure the test taker's ability to understand English as it is spoken in North America, including frequently used vocabulary, expressions and grammar. With the computer based version, test takers will now listen to dialogues, talks and group discussions through personal headphones while they see context-setting visuals on the computer screen.
The Structure section will measure the ability to recognise language that is appropriate for standard written English. The Reading section will still use passages to measure the ability to understand non-technical reading material, but new tasks that require the test taker to become more closely involved with the text have been developed. The Writing section will measure the person's ability to generate, organise and support ideas using standard written English in an essay. In order not to disadvantage people who lack keyboard skills, test takers may handwrite or type the essay. The essay rating will now be combined with the Structure section score to create a compiled Structure/Writing scaled score. The essay rating will constitute one-half of the Structure/Writing scaled score.
Another improvement is that the Listening and Structure sections will be computer adaptive. Test takers receive questions that are appropriate for their performance level. The section begins with a question of average difficulty. If the test taker answers the question correctly, the next question will be one that fulfils the test design and will typically be of the same or higher difficulty level. If the test taker answers incorrectly, the next question will he one that fulfils the test design and will typically be of the same or lower difficulty level. Thus, all subsequent questions presented are based in part on the test taker's performance on previous questions and in part on the test design.
Computer-based TOEFL will be offered worldwide at Sylvan Technology Centers, specified university test centers, ETS field offices and other locations worldwide. Testing will be available year-round at more than 300 test centers around the world. Test takers will make appointments by calling either their local test center or Regional Registration Center. The appointments can be made within a few days of testing. However, test takers should consider admission deadlines and call early to maximise chances of getting preferred test dates at the most convenient center.
The entire testing experience will also improve as test takers will sit in private carrels where they will use volume-controlled headphones. Score reporting will also be faster as students will see partial scores on the screen at the test center and official score reports will be sent usually within two weeks of testing.
Because the content and format of the TOEFL test have changed, scores on the computer-based test will be reported on a new score scale. This new score scale has been designed to distinguish scores received on the computer-based TOEFL from those received on the paper-based test. For the computer-based test the examinee will receive four scaled scores:
Listening (0 to 30), Structure/Writing (0 to 30), Reading (0 to 30), and a total score (000 to 300). The three section scores and a total score will be reported on the score report. The essay rating will be integrated in the Structure/Writing score; in addition, the score on the essay will be reported as an independent rating on this form. To assist score users in setting new score standards on the new scale, work has begun to produce a concordance study and table that will relate scores on the computer-based test with those on the paper based exam. Results of the concordance study and table will be available in the Spring, 1998.
The TOEFL program has taken steps to assure that an individual's test performance is not influenced by a lack of computer experience. A tutorial designed especially for non-native speakers of English, has been developed to teach the skills needed to take the computer based TOEFL. The results from a study conducted by ETS indicate the tutorial is effective because there was no practical difference between the performance of test takers who were familiar and unfamiliar with the computer. Additional information on this computer familiarity study will be available in early 1998.
Although the United States, Canada, Latin America, Europe, the Middle East, Africa, Australia and New Zealand, and eight areas in Asia (Brunei, Indonesia, Nepal, Malaysia, Mongolia, the Philippines, Singapore, and Sri Lanka) will make the transition to computer in 1998, the paper based test will continue to be administered in the other areas in Asia. However, once computer-based TOEFL is introduced in a country, the paper-based program will be eliminated ETS plans to complete the transition to computer by the year 2001.
