From: Validity argument for the VELC Test® score interpretations and uses
Inference | Interpretive argument (assumptions) | Validity argument (warrants) | Backings collected |
---|---|---|---|
Scoring | ・The VELC Test® items work on test-takers to make placement decisions. | ・Because most of the items were working for placement purposes, inference from the observation to the observed score was valid. | ・Item discrimination values ・Item difficulty estimates |
Generalization | ・The test score is generalizable to other observations so as to reduce the measurement error. | ・Because the test score was reliable and generalizable with a small amount of measurement error, inference from the observed score to the universe score was valid. | ・Cronbach alpha coefficients ・Standard error of measurement ・Generalizability coefficients |
Extrapolation | ・The test score indicates what test-takers can do with their English proficiency. | ・Because the test score indicated what test-takers could do with their English, inference from the universe score to the target score was valid. | ・Person ability estimates in relation to can-do item difficulty estimates |
Decision | ・The test score is appropirate for making placement decisions and useful for test-takers’ further learning. | ・Because the test score can be used to place certain strata and the test score report can be useful for further learning, inference from the target score to the use was fairly valid, yet the test consequence need to be investigated further. | ・Rasch separation index ・experts’ judgment |