The purpose of this study was to identify minimum test scores (cut scores) on each of the four
sections/tests of the TOEIC test corresponding to the different STANAG proficiency levels. The TOEIC
test is a measure of general English-language skills and was not designed to be a specific measure
of the STANAG levels. Therefore, the first phase of this study was to conduct an alignment process,
whereby the experts on the panel identified which particular STANAG levels were addressed by the
TOEIC test. Two rounds of alignment judgments occurred, with at least 10 of the 15 experts (67%)
indicating the following alignments: TOEIC Listening section (Levels 1 and 2), TOEIC Reading section
(Levels 2 and 3), TOEIC Writing test (Levels 1, 2, and 3), and TOEIC Speaking test (Levels 1 and 2).
Standards were recommended for those levels identified by the alignment process. A modified
Angoff standard-setting approach with a holistic component was applied to the selected-response
questions (TOEIC Listening and Reading test), and a variation of a performance sample approach
was applied to the productive-response questions (TOEIC Writing and Speaking tests). Three rounds
of judgments, with feedback and discussion occurred; the feedback included data on how test takers
performed on the questions and the percentage of test takers who would have been classified into
each of the proficiency levels. Table 10 presents the final cut score recommendations, both in raw
score and scaled score metrics.
The purpose of this study was to identify minimum test scores (cut scores) on each of the foursections/tests of the TOEIC test corresponding to the different STANAG proficiency levels. The TOEICtest is a measure of general English-language skills and was not designed to be a specific measureof the STANAG levels. Therefore, the first phase of this study was to conduct an alignment process,whereby the experts on the panel identified which particular STANAG levels were addressed by theTOEIC test. Two rounds of alignment judgments occurred, with at least 10 of the 15 experts (67%)indicating the following alignments: TOEIC Listening section (Levels 1 and 2), TOEIC Reading section(Levels 2 and 3), TOEIC Writing test (Levels 1, 2, and 3), and TOEIC Speaking test (Levels 1 and 2).Standards were recommended for those levels identified by the alignment process. A modifiedAngoff standard-setting approach with a holistic component was applied to the selected-responsequestions (TOEIC Listening and Reading test), and a variation of a performance sample approachwas applied to the productive-response questions (TOEIC Writing and Speaking tests). Three roundsof judgments, with feedback and discussion occurred; the feedback included data on how test takersperformed on the questions and the percentage of test takers who would have been classified intoeach of the proficiency levels. Table 10 presents the final cut score recommendations, both in rawscore and scaled score metrics.
การแปล กรุณารอสักครู่..
