Final published version
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Chapter
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Chapter
}
TY - CHAP
T1 - Using Multi-Faceted Rasch Analysis to Examine Raters, Prompts, and Rubrics in an Office-Hour Role-Play Task
AU - Ren, Haoshan (Sally)
AU - Looney, Stephen
AU - Cushing, Sara
PY - 2023/7/19
Y1 - 2023/7/19
N2 - Locally-administered placement tests for International Teaching Assistants (ITAs) impact the quality of ITAs’ graduate studies and the quality of undergraduate education. However, in-house testing programs face challenges in terms of funding, time, and expertise to ensure the reliability and address the validity of their ITA tests. This chapter examines comparability of prompts used in an office-hour roleplay task on an in-house ITA test. The role-play task requires test-takers to respond to one of four office-hour scenarios where the test-taker plays the role of the teacher and one of the two raters plays the role of the undergraduate student. We used Many-Faceted Rasch Measurement (MFRM) to investigate prompt difficulty, rubric consistency, and selected interactions. All raters were self-consistent, but we found significant differences in raters’ relative severities. Further examination reveals that, although raters were instructed to select one of the four prompts randomly, some raters selected certain prompts to the exclusion of others. The rater x prompt interaction shows that raters’ internal consistency is related to selection, while variation in rater severity is an independent effect. This chapter provides insights into practices that connect test analysis, test quality, and test administration in a local context.
AB - Locally-administered placement tests for International Teaching Assistants (ITAs) impact the quality of ITAs’ graduate studies and the quality of undergraduate education. However, in-house testing programs face challenges in terms of funding, time, and expertise to ensure the reliability and address the validity of their ITA tests. This chapter examines comparability of prompts used in an office-hour roleplay task on an in-house ITA test. The role-play task requires test-takers to respond to one of four office-hour scenarios where the test-taker plays the role of the teacher and one of the two raters plays the role of the undergraduate student. We used Many-Faceted Rasch Measurement (MFRM) to investigate prompt difficulty, rubric consistency, and selected interactions. All raters were self-consistent, but we found significant differences in raters’ relative severities. Further examination reveals that, although raters were instructed to select one of the four prompts randomly, some raters selected certain prompts to the exclusion of others. The rater x prompt interaction shows that raters’ internal consistency is related to selection, while variation in rater severity is an independent effect. This chapter provides insights into practices that connect test analysis, test quality, and test administration in a local context.
U2 - 10.1007/978-3-031-33541-9_2
DO - 10.1007/978-3-031-33541-9_2
M3 - Chapter
SN - 9783031335402
SN - 9783031335433
T3 - Educational Linguistics
SP - 13
EP - 33
BT - Local Language Testing
A2 - Yan, Xun
A2 - Dimova, Slobdanka
A2 - Ginther, April
PB - Springer
CY - Cham
ER -