Chapter 3 Constructing Tests

Chapter 3 Constructing Tests

• 1.
  EII604 Chapter 3 –Part C Designing Classroom Language Test
• 2.
  Assessment of studentlearning provides evidence so that sound educational decisions can be made. This evidence should help us to evaluate (or judge the merit of) a teaching program or we may use the evidence to make statements about student competence or to make decisions about the next aspect of teaching for particular students. -UNESCO
• 3.
  Stages of TestConstruction STAGE 1 Determining Objectives STAGE 2 Planning Test Specifications STAGE 3 Writing the Test STAGE 5 Preparation STAGE 4 Pre-testing
• 4.
  BEFORE WE START Justa reminder: Don’t forget your “R and V”: Reliability • consistency (free from extraneous sources of error) Validity • how well a test measures what it is supposed to measure
• 5.
  STAGE 1: ESTABLISHOBJECTIVES STEP OBJECTIVE WHAT TO CONSIDER 1 The objective of the test What should the students know and what should they be able to do? 3 The need for the test How will the outcome help students, teacher, or institution better their performance or program? 3 Alternative testing forms Is this test really necessary? Can the outcome be determined by an alternative form? 4 The test population Who will take the test? 5 The style of administration Will the test be taken on the computer or by paper? Where will the test be administered? Can it be administered at any time? 6 Test administrator Who is authorized to give this test? Is there training required to administer the test? 7 The content The content of the test should cover performance, not linguistic domain (grammar, lexis, etc.) 8 Test format What kind of tasks will the test include? Will the questions be multiple choice, in the form of an essay question or a combination of item formats?
• 6.
  STAGE 2: PLANNINGTEST SPECIFICATIONS STEP OBJECTIVE WHAT TO CONSIDER 1 The content The content that the test should cover. This will depend on your curriculum. E.g. If you are working with an integrated curriculum, you will need to design a test that measures performance, not linguistic domain (grammar, lexis, etc.) 2 Items, types and tasks Does the prompt (question) require students to listen or read? Does the response require the student to speak or write? (see Brown p. 51, Figure 3.1). What kind of tasks will the test include? Will the questions be multiple choice, in the form of an essay question or a combination of item formats? 3 Timing How much time does the test require in order to assure reliability? Consider how long it takes for a student to read, interpret and think about questions. This is especially important for students with English as a second language – they may have to reread several times and redraft answers. General guideline: You should allow 1 minute for each multiple choice question (MCQ) and up to 2 minutes for more challenging MCQs; for short answer questions (SAQ), allow even up to 3 minutes. 4 Performance What performance is the test measuring? Is this consistent with the content? 5 Scoring What skill(s) are you focused on testing principally? Will the sum of the tasks of the focused skill(s) account for a larger percentage or points than the rest of the test? How will you distribute points throughout the test questions? (see Brown, pp 61- 64) 6 Reporting How will the test be reported to students, parents and administrators?
• 7.
  STAGE 3: WRITINGTHE TEST STEP OBJECTIVE WHAT TO CONSIDER 1 Face Validity How will students perceive the test? When they look at it initially, will they feel it is appropriate and to the point? Does it test what they were told it would test? 2 Authenticity Does the test include topics familiar to the student? Are the language and content authentic? 3 Validity Does the test measure what it is supposed to measure as far as content and level of difficulty? 4 Timing Do the test questions fit the allocated time allowed for the test administration? 5 Instructions Are the instructions for each section absolutely clear for the students? Is there an example item for each section? 6 Distractors In the case of MCQs, are the distractors sufficiently challenging as to not make the correct answer easy to identify. Are the distractors clearly wrong and not designed to confuse the student? 7 Format Is the format of the test organized? Is each section clearly identifiable? Are the instructions and sample question easy to read? Is it easy for the student to identify and read the answer questions (and options)?
• 8.
  STAGE 4: PRE-TESTING STEPOBJECTIVE WHAT TO CONSIDER 1 Test revision Has a colleague looked over the test and given you feedback? If possible, did you test out all or parts of the test on students similar to the target group? 2 Timing Were you able to time the test? 3 Components Are all components of the test (e.g. audio, video, text) available, clear and effective? 4 Format Was it easy to navigate through the test? Was the test easy to read? 5 Content Did it cover the content it was supposed to test? 6 Face validity Did it test what it was intended to test?
• 9.
  STAGE 5: PREPARATION STEPOBJECTIVE WHAT TO CONSIDER 1 Test Sufficient and legible copies of the test and answer sheets have been made and assembled. 2 Test resources Correct components are prepared and available. If the test requires a CD player, one is available. 3 Location Classroom for testing is assigned and available. There is proper lighting and ventilation. The distances between desks are appropriate (usually 1.25 meters from one center to another). Locate sockets where CD player will play. Is it in an ideal location in the room? 4 Test administrator / supervisors Test administrators and supervisors are available for the session.