Stages of Test Development By Lily Novita - 69090007 Make a full and clear statement of the testing ‘problem’. Write complete specifications for the test. Write and moderate items. Trial the items informally on native speakers and reject or modify problematic ones as necessary. Trial the test on a group of non-native speakers similar to those for whom the test is intended. Analyse the results of the trial and make any necessary changes. Calibrate scales. Validate. Write handbooks for test takers, test users and staff. Train any necessary staff (interviewers, raters, etc.). 1. Stating the Problem The essential initial step in any testing is to make oneself perfectly clear what one wants to know and for what purpose What kind of test it is constructed for? What is the precise purpose? What abilities are to be tested? How detailed must the results be? How accurate must the results be? How important is backwash? What constraints are set by unavailability of expertise, facilities, time ? (for construction, administration and scoring) 2. Writing specifications for the test Content Operations Types of text Addresses of texts Length of text(s) Topics Readability Structural range Vocabulary Range Dialect, accent, style Speed of processing 2. Writing specifications for the test Structure , timing, medium/channel and techniques Test structure Number of items Medium / channel Timing Techniques 2. Writing specifications for the test Criterial levels of performance Accuracy Appropriacy Range Flexibility Size 2. Writing specifications for the test Scoring procedures Subjectivity Achievement of high reliability & validity in scoring Rating scale to be used? No. of people rating each piece of work? Solutions on disagreements between raters 3. Writing and moderating items Sampling Writing items Moderating items 4. Informal trialling of items on native speakers Moderation of grammar test 1. Is the English grammatically correct? 2. Is the English natural and acceptable? 3. Is the English in accordance with the specifications? 4. Does the item test what it is supposed to test, as specified? 5. The correct response cannot be obtained without the appropriate knowledge of grammar (other than random sampling) 6. Is the item economical? 7. (a) Multiple choice – is there just one correct response? (b) Gap filling – are there just one or two correct responses? 8. Multiple choice : Are all the distractors likely to distract? 9. Is the key complete and correct? Yes No 5. Trialling of the test on a group of non-native speakers similar to those for whom the test is intended trials are designed to help ensure that the items function appropriately and are not confusing for the students. this is accomplished by embedding field test items in the operational test, to ensure that the items are taken by a representative group of motivated students under standard conditions. 6. Analysis of the results of the trial – making of any necessary changes 2 kinds of analysis should be carried out : Statistical analysis : reveals qualities (reliability) as a whole and individual items – how difficult they are , how well they discriminate between stronger and weaker candidates. Qualitative analysis : responses are examined to discover misinterpretations, unanticipated but possibly correct answers and indicators of other faulty items. 7. Calibration of scales It means collecting samples of performance which cover the full range of the scales. A calibration test is a procedure in which an instrument, tool, or device is tested to confirm that it conforms with the standard. Calibration is very important, as it ensures that objects are working properly. There are a number of reasons to conduct a calibration test, ranging from concerns that something is not working as it should to preparations for an event in which very precise calibration is desired, and there are a number of ways to perform a calibration. 8. Validation Essential validation – for high stakes or published tests Small-scale validation – for low stakes used within an institution 9. Writing handbooks for test takers, test users and staffs (contents) The rationale for the test; An account of how the test was developed and validated A description of the test Sample items Advice on preparing for taking the test An explanation of how test scores are to be interpreted Training materials Details of test administration 10. Training Staff All staffs who will be involved in the test process should be trained : interviewers, raters, scorers, computer operators, and invigilators.