| Author/Editor | Mathysen, D. G.; Hawlina, Marko | |
| Title | History and future of the European Board of Ophthalmology Diploma examination | |
| Type | članek | |
| Vol. and No. | Letnik 91, št. 6 | |
| Publication year | 2013 | |
| Volume | str. 589-593 | |
| ISSN | 1755-3768 - Acta ophthalmologica | |
| Language | eng | |
| Abstract | Purpose: The European Board of Ophthalmology Diploma (EBOD) examination has evolved over the last few years, especially with the introduction of negative marking (-0.5 points) for incorrect or blank answers (0 points for don't know option), which aimed to improve the quality and reliability of the examination. Methods: In 2010, negative marking at the written part of the EBOD examination has been introduced in an attempt to improve not only the reliability of the examination as entity but also the statistical performance parameters of the individual questions. As lower pass rates and discriminationof female candidates are feared by the general public when negative marking is concerned, these parameters have been explicitly investigated. Results: Introduction of negative marking has not only lead to improved reliability of the EBOD examination (increased Cronbach's alpha value: Ž0.80 without and Ž0.90 with negative marking), but also to improved statistical performance parameters of the individual questions. The pass rate of the EBOD examination has proven to remain at the same high level as withoutnegative marking (around 90%). Furthermore, although female candidates do seem to have different answering strategies (pŽ<Ž0.01, use of don't know option), no statistically significant difference has been found between total scores of male and female candidates (pŽ>Ž0.05). Conclusion: Introduction of negative marking at the written EBOD examination has proven to be beneficial, not only for the organizers (improvement of the statistical performance of theexamination and its questions), but also for candidates (better discrimination with borderline candidates). These results have been obtained without evidence of lower pass rates or discrimination of female candidates. | |
| Keywords | clinical validation of test items mathematical validation of test items multiple-choice questions statistical performance analysis |