评价与测量基本课件201152testusefulness.pptVIP

V) 2 Reliability The reliability of a test is its consistency : tape measure that stays the same length all the time as opposed to a piece of elastic. Same results should be obtained wherever the tape measure is used. Reliability Comparison with different tests: i.e. whether the students takes one version of the test e.g. EPT, or another, the result should be the same. Factors that Affect Language Test Scores THREE ASPECTS OF RELIABILITY 1) Circumstances in which the test is taken 2) The way in which it is marked 3) the uniformity of assessment it makes.  EXTRINSIC sources of error: i) Examiner variability ii) Variability of testing conditions. INTRINSIC sources of error: i) Lack of stability ii) Lack of equivalence. How To obtain estimates of equivalence reliability Equivalence reliability: A measuring device is equivalent to another measuring device if both give the same results when applied to the same object. 1). Equivalent-forms Method: Parallel versions correlate the two sets of scores. Administer both versions to the same group of individuals 2). Split-half Method: To split the test into halves that are most equivalent, the usual procedure is to score the even-numbered and the odd-numbered items separately. If 100 items, scores calculated separately for 2 sets of 50 items and results correlated. 3) Test-retest Method: Administer the same test twice to the same group of individuals. The resulting test scores are correlated, and this correlation coefficient provides a measure of stability; that is, it indicates how stable the test results are over the given period of time. Variance estimates: M = the mean; n = number of items in the test; s = Standard Deviation; r = reliability estimate. Kuder-Richardson formula 21: r= 1 - [M (n-M)] / n s squared Score f x Equivalence Reliability for a certified achievement test should reach 0.7. Factors Influencing Reliabil