TOEFL tips preparing students for the computer based test

Test format The computer-based TOEFL test features many of the question types used on the paper-based test and adds new question types that can be offered only on computer. Listening section The Listening section measures the candidate’s ability to understand English as it is spoken in North America. This section includes various stimuli — such as dialogues, short conversations, academic discussions, and excerpts from lectures (also called minilectures) — and poses questions that test comprehension of main ideas, order of a process, supporting ideas, and important details, as well as the ability to make inferences and categorize topics and objects. All stimuli are about academic topics and situations. After each stimulus is played, examinees both see and hear each question before they see the answer choices. This encourages examinees to listen for main ideas. Five types of questions are found in the Listening section: ● ● ● ● ● traditional multiple-choice questions with four answer choices questions that require examinees to select a visual or part of a visual questions with more than one answer (e.g., two answers out of four choices) questions that require examinees to order events or steps in a process questions that require examinees to match objects or text to categories File Copyright ᭧ 1998, ETS Listening - Practice Questions & Review 14 of 27 The professor briefly explains a process. Summarize the process by putting the events in order. Click on a sentence. Then click on the space where it belongs. Use each sentence only once. Fossilization takes place. An insect is trapped. Time Resin is produced by a tree. Resin becomes hard. 1. Resin is produced by a tree. 2. An insect is trapped. 3. Resin becomes hard. 4. Fossilization takes place. Volume ? Help Answer Confirm Next Visuals accompany the listening stimuli to provide context and illustrate the topic. These visuals are a new feature of the computer-based test that allow for varied and interactive stimuli that more closely approximate actual listening experiences, such as classroom lectures and campus conversations. The new question types require test takers to demonstrate their English proficiency by interacting with the information presented to them. Examinees must use integrated language skills to manipulate graphics, categorize information, and order elements in a process, tasks that are similar to those required of students in the classroom. The actual time needed to complete the Listening section is 40 to 60 minutes, but only 15 to 25 minutes of that time is for answering questions. The rest is for listening to the stimuli. The clock runs while examinees answer questions but not while they listen to the stimuli. The listening stimuli and questions are presented only once, as in the paper-based test. However, the computer-based test is different from the paper-based test 10 in the following ways: examinees see the visual as they listen, and they both see and hear the question before the answer choices appear. Also, the computer-based Listening section is self-paced. That is, examinees can choose to go on to the next question when they are ready. This is different from the paper-based test, in which the pace of the Listening section is determined by the pace of the prerecorded tape. Copyright ᭧ 1998, ETS Listening - Practice Questions & Review File Volume Time ? Help Answer Confirm Next The Listening section consists of two parts: A and B. In Part A, examinees hear dialogues between two people. In the dialogues, each speaker takes one to two turns. Each dialogue is followed by one multiplechoice question with four answer choices. In Part B, examinees hear short conversations in which both of the speakers take more than one turn. Each conversation lasts up to one minute and is followed by two to three multiple-choice questions with four answer choices. Examinees should click on the best answer on the basis of what is stated or implied by the speakers. In Part B, examinees also hear academic discussions among two or more speakers, and minilectures about a variety of topics. Test takers do not need special background knowledge to answer the questions correctly; all the information needed to answer the questions is contained in the listening material presented. The discussions are up to two minutes in length; the minilectures are up to two and one-half minutes long. Both are followed by several questions. For most questions, examinees will click on the best of four possible answers, but some of the new types of questions listed on page 10 are included as well. Chart B: Listening Section Format Part Type of Stimulus Number of Stimuli Number of Questions A Dialogues 11-17 1 each B Short Conversations (up to 1 minute) 2-3 2-3 each B Minilectures (up to 2.5 minutes) and Academic Discussions (up to 2 minutes) 4-6 3-6 each 11 Structure section The Structure section measures the candidate’s ability to recognize language that is appropriate for standard written English. The language tested is formal, rather than conversational. The same two question types used on the paper-based TOEFL test also appear on the computer-based test: ● questions in which examinees must complete a sentence using one of four choices provided ● questions in which they must identify one of four underlined words or phrases that is incorrect Both question types are mixed randomly throughout the section rather than separated into two subsections as in the paper-based test. Reading section The Reading section measures the candidate’s ability to understand short passages similar in topic and style to academic texts used in North American colleges and universities. This section tests comprehension of main ideas, factual information, inferences, and vocabulary (direct meaning, synonym, and antonym). It also assesses the ability to understand the rhetorical organization of a passage. Examinees read four to five passages of 250 to 350 words on academic subjects and answer 11 questions about each passage. As in the Listening section, test takers do not need any special background knowledge to answer the questions in the Reading section correctly; all the information needed to answer the questions is contained in the passages. Examinees must read through or scroll to the end of each passage before receiving questions on that passage. The 70 to 90 minutes allotted for this section include the time spent reading the passages and answering the questions. Chart C: Reading Section Format Length of Passage 250-350 words Number of Passages 4-5 Number of Questions 11 per passage There are several question types in this section, most of them multiple-choice. New question types include ● questions that require examinees to click on a word, phrase, sentence, or paragraph to answer ● questions that ask examinees to “insert a sentence” where it fits best The new question types require examinees to demonstrate their comprehension of a passage by performing the following tasks: ● highlighting information in a passage to demonstrate their comprehension of main ideas and supporting details ● identifying paraphrases and inferences to demonstrate their ability to analyze meaning ● inserting an extra sentence into the appropriate part of a passage to demonstrate a more cognitively complex aspect of comprehension: the ability to understand the organization of a passage, including elements of coherence and cohesion. For example, examinees may be required to understand how transitional words or pronouns are used to establish cohesion. Or examinees may be required to identify the coherent, logical progression of ideas. ● identifying words or phrases with similar meanings, demonstrating an ability to comprehend discrete language segments and use context clues 12 Copyright ᭧ 1998, ETS Reading Comprehension - Practice Question & Review File 6 of 20 Questions 1 to 6 More Available But this was clearly not just a matter of the physical advantages of early humans of erect posture and having the hands free to carry something else. Fetching branches for a fire implies that the individuals concerned thought about what they were doing, and knew why they were doing it. Keeping a fire going implies foresight and care. Wood had to be gathered, and perhaps even stored during wet periods. Such activities did not come naturally to early humans; they required learning and discipline. Especially when humans began to collect fuel over larger distances, they devoted part of their energy to maintaining something outside themselves, something beyond their own immediate needs. This is not to say they were acting “unselfishly.” Tending the fire was a form of “deferred gratification” or putting off the satisfaction of immediate need in planning for future needs, like that which was later to become an essential ingredient in agriculture and livestock-raising. The following sentence can be added to paragraph 2. On the contrary, in caring for the fire they were also caring for themselves. Where should it best fit in the paragraph? Click on the square ( ) to add the sentence to the paragraph. Paragraph 2 is marked with an arrow ( ). ? Help Time Previous Next Writing section The Writing section measures the candidate’s ability to write an essay in English on a single assigned topic, including the ability to generate, organize, and develop ideas; to support those ideas with examples or evidence; and to compose a response in standard written English. This section directly measures the test taker’s ability to produce language and complements the information obtained from the Structure section. The essay is required on the computer-based test, not optional, as the Test of Written English is with the paper-based test. The essay topic is chosen by the computer from a pool of topics published in the Bulletin and on the TOEFL Web site. Examinees must compose the essay in the 30-minute time limit and can either handwrite the essay or type it on the computer. (The word-processing program used is Notepad-based and does not include tab, spell-check, or grammar-check functions.) Examinees are given scratch paper on which they can organize their thoughts. However, only essays written on the official answer sheet or typed into the computer are scored. (See Section C for the rating scale used to grade the essay.) Copyright ᭧ 1998, ETS Writing Practice Topics TOEFL - Writing File Read the question below and type your response in the box. Do you agree or disagree with the following statement? Teachers should make learning enjoyable and fun for their students. Use reasons and specific examples to support your opinion. Cut Paste Undo ? Time Help Answer Confirm Next 13 Step 3: Learn how computer-based testing works and is scored. Computer-adaptive testing The computer-adaptive sections of the TOEFL test, Listening and Structure, are tailored to the individual test taker. Each examinee receives a set of questions that meets the test design and is generally appropriate for his or her proficiency level. Test design determines the total number of questions and the variety of question types, as well as the subject matter presented (i.e., variety of topics for listening stimuli and reading passages). The computer-adaptive sections start with questions of moderate difficulty. As test takers answer each question, the computer scores the question and uses that information, as well as the responses to previous questions, to determine which question is presented next. As long as candidates respond correctly, the computer typically selects a next question of greater or equal difficulty. In contrast, if they answer a question incorrectly, the computer typically selects a question of lesser or equal difficulty. The computer is programmed to choose questions that meet the test design as it continuously adjusts to find questions of appropriate difficulty for test takers of all performance levels. This means that different test takers will be given different, but comparable, questions. However, examinees should not feel alarmed if they receive slightly easier questions than expected; it may mean that the computer is meeting the test design. All examinees must answer a predetermined number of questions from each of the specified content categories, and some of these categories naturally involve less difficult material. Thus, even individuals who receive high scores are typically required to answer some relatively easy questions. However, with computer-adaptive testing, candidates receive fewer questions that are either too easy or too difficult for them. The computer is constantly probing to confirm just how high each examinee can perform; this means that the test is appropriately challenging to each test taker. More technically speaking, in a computer-adaptive test, the computer is programmed to estimate an examinee’s ability and choose items that will provide the most information to refine the ability estimate. At the beginning of the test or section, the assumed ability is at the middle of the score range. The first question is at a moderate level of difficulty, meaning that there is a high probability that examinees from middle to high ability will answer it correctly and a lower probability that examinees at a low level of ability will answer it correctly. Naturally, at the beginning of a section, almost any question will add significantly to the information about the examinee. Later in the section, more distinction between ability levels is possible. For instance, in a 20-item section, suppose that after 10 items an examinee has answered four out of five above-average difficulty questions correctly. The computer will assume the examinee’s ability level to be in the range of medium to high; no useful information can be gained by administering easy questions to this candidate. Conversely, if, after 10 questions an examinee answered four out of five questions of average difficulty incorrectly, the computer assumes that the person is in the medium to low ability level and chooses easier questions to find the candidate’s level of ability more accurately. In the computer-adaptive sections, only one question is presented at a time. Because the computer scores each question before selecting the next one, examinees must answer each question when it is presented. This means test takers cannot skip questions, and once they have confirmed an answer, they cannot return to that question. However, examinees are able to change their answer to a question as often they like until they have made their final choice, confirmed their answer, and moved on to the next question. Occasionally examinees may give an incorrect answer due to careless error or may answer correctly with a lucky guess. However, the adaptive nature of the test generally will lead them back to questions of appropriate difficulty. After they answer a question incorrectly, examinees still have the opportunity to receive more difficult questions if they answer subsequent questions correctly. The computer tracks the pattern of examinee responses and obtains substantial evidence before confirming an estimate of ability. 14 How the computer-adaptive sections are scored The scores on the computer-adaptive Listening and Structure sections depend on the number of questions examinees answer in the time allotted, the difficulty of the questions given, and the candidates’ performance on those questions. (Pretest questions may be randomly inserted, but they do not count toward a score.1) More credit is given for correctly answering a difficult question than for correctly answering an easy question. If two test takers have the same number of correct responses, generally the person who answers more of the difficult questions correctly will receive the higher score. Similarly, for two test takers who correctly answer questions of equivalent difficulty on average, the person who proceeds slowly through the test and leaves several questions unanswered at the end will receive a lower score than the one who answered every question. The scoring of a computer-adaptive section is cumulative. If the last question in a section is relatively easy and is answered incorrectly, it does not mean the examinee will receive a low score. Because adaptive tests are tailored to the ability level of the examinee, it is expected that all examinees ultimately will answer some questions incorrectly. The scoring process takes this into account and, consequently, it is possible to earn a very high score even if several questions are answered incorrectly. Computerized nonadaptive testing The Reading section is not computer-adaptive. This section contains passages accompanied by several questions covering the full range of difficulty. Because the selection of these passages and questions is not based on cumulative performance, examinees are allowed to skip items or go back to previous questions. As with the adaptive sections, the Reading section scores are based on the number of questions examinees answer in the time allotted, as well as the difficulty level of and examinees’ performance on the questions given. The fundamental difference between the Reading section and the adaptive Listening and Structure sections is in how the questions are selected. In the Reading section, questions are selected to fit the test design; they are not tailored to examinee ability level. Step 4: Use test-taking strategies that will allow them to do their best. Students are strongly urged to review the following information as they prepare for the test. General test-taking strategies Long before the day of the test, examinees should: ● Learn about the test. It’s always best to know as much as possible about what to expect before arriving at the test center. Test takers can use the free Bulletin or the TOEFL Web site to get started. These are valuable resources for learning about the types of questions, test directions, and procedures to expect on test day. They describe the number of questions in each section, the time allotted for each section, and the total test time. ● Use TOEFL test preparation materials for more practice. At minimum students should do some sample questions from each section of the test to become familiar with them. The TOEFL Sampler and POWERPREP software (available in the year 2000) will help candidates practice for the computer-based test. Most of the questions on the computer-based TOEFL test are similar to those in the paper-based test. Therefore, examinees can also use exercises from paper-based TOEFL test preparation materials to prepare for the computer-based test. See page 38 for information about how to obtain official TOEFL preparation materials. 1 The administration of pretest questions ensures that each question provides reliable information about an examinee’s ability, a critical part of the test-building process for future TOEFL tests. 15