Please use this identifier to cite or link to this item:
Rater bias in assessing the pragmatics of KFL learners using facets analysis
|Title:||Rater bias in assessing the pragmatics of KFL learners using facets analysis|
|Authors:||Youn, Soo Jung|
|Advisor:||Brown, James D.|
|Abstract:||As interest in research on second language pragmatics increases, some pragmatics research has been done on Korean as a foreign language (KFL) learners. This research has focused on pedagogical aspects of Korean pragmatics and interlanguage pragmatics. However, very little research has been done on the pragmatics assessment of KFL learners, in terms of discussing appropriate test types and whether certain factors affect raters’ assessments of KFL learners’ pragmatics performance. The focus of this study is on investigatin whether various factors, including test types, speech acts, groups of candidate, and test items, affect three raters’ assessments of the pragmatic competence of KFL learners. For these purposes, this study analyzes interactions between test results and raters using the computer program FACETS (Linacre, 1996): the interactions between rater bias and test types, rater bias and speech acts, rater bias and item difficulty, and rater bias and examinee levels. This study uses three different pragmatics tests adapted from Hudson, Detmer, and Brown’s (1995) pragmatics prototype tests: Open-written Discourse Completion Task, Language Lab, and Role-play. Within each of these three test types are three speech acts: refusal, apology, and request. The results of this research indicate that all three raters showed different degrees of severity in their ratings, depending on the test type and speech act. Additionally, each rater showed unique bias patterns within the interactions. I will discuss how certain speech acts and test types affect rater assessments, what kinds of systematic bias patterns the raters show across various factors, and what these research findings mean for KFL classrooms.|
|Appears in Collections:||SLS Papers|
Please contact email@example.com if you need this content in an alternative format.