Linking OPIc Levels to TOEIC Speaking Scores
- Author(s):
- Qu, Yanxuan; Schmidgall, Jonathan; Cid, Jaime; Chan, Eric
- Publication Year:
- 2019
- Report Number:
- RM-19-02
- Source:
- ETS Research Memorandum
- Document Type:
- Report
- Page Count:
- 19
- Subject/Key Words:
- English Proficiency Equipercentile Linking Oral Proficiency Interview-Computer (OPIc) Scale Alignment Score Comparison Score Mapping South Korea Speaking Proficiency Test Comparability Test Score Reliability TOEIC Speaking Test
Abstract
SUMMARY: An important aspect of test validity is the appropriateness of the interpretation and usage of test scores. Caution is needed when comparing scores on two tests, such as the OPIc (Korea) test and the TOEIC Speaking test, which have different content. Inappropriate comparison between scores of different tests may lead to unfair decision making. This study addressed the increasing public demand to compare scores on the OPIc test and the TOEIC Speaking test. • First, we examined why, despite similarities between the two tests, their scores cannot be used interchangeably. • Next, we investigated empirically the relationship of the two tests based on real data collected from Korean test takers who took both tests. A recommendation for the interpretation of the scores on the two tests is also provided. The study found that the highest OPIc (Korea) level is not comparable to the highest TOEIC Speaking score. A test taker with the highest OPIc (Korea) level is most likely not as proficient in speaking English as a test taker with the highest TOEIC Speaking score. Given the results, to prevent invalid and unfair comparison especially near the top of the two test scales, scores of the two tests should not be interpreted as equivalent or interchangeable. ABSTRACT: This study reports a comparison between scores on the Oral Proficiency Interview-Computer (OPIc) test and the TOEIC Speaking test for South Korean test takers. It begins by examining the extent to which the tests (a) measure speaking proficiency in a similar manner, (b) have similar test populations, (c) measure similar ranges of speaking proficiency, and (d) have comparable levels of reliability. Based on the comparability of the tests, several approaches to score mapping were investigated empirically using a sample of South Korean test takers (N = 1,019) who completed both tests within a specified time period. The analysis resulted in a table that can be used to compare scores on the OPIc (as currently used in South Korea) and the TOEIC Speaking test with the understanding that scores on the tests should not be considered interchangeable given important differences regarding aspects of their design and reported range of proficiency.