Abstract
Eighty-three Test of English as a Foreign Language (TOEFL) writing prompts administered via computer-based testing between July 1998 and August 2000 were examined for differences attributable to the response mode (handwriting or word processing) chosen by examinees. Differences were examined statistically using polytomous logistic regression. A variable measuring English-language ability (ELA) was developed from the multiple-choice components of the TOEFL and used as a matching variable. Although there was little observed difference in mean writing scores, when examinees were matched on ELA, small differences were observed in effect sizes consistently favoring the handwriting response mode. This difference favoring the handwriting response mode occurred for all of the writing prompts analyzed, suggesting a general effect for response mode. Differences for individual writing prompts were small, however.
Original language | English |
---|---|
Pages (from-to) | 577-595 |
Number of pages | 19 |
Journal | Educational and Psychological Measurement |
Volume | 65 |
Issue number | 4 |
DOIs | |
Publication status | Published - 2005 Aug |
Keywords
- Comparability
- Computer-based writing assessment
- Essay prompts
- Fairness
- Logistic regression
- Proportional odds ratio model
- Response mode
ASJC Scopus subject areas
- Education
- Developmental and Educational Psychology
- Applied Psychology
- Applied Mathematics