Introduction: Multiple choice exams are most common tests in medical education. So some teacher- made tests have different problems. The aim of this study was to evaluate quantitative and qualitative aspects of MCQ in nursing school in North of Iran in 2009.
Methods: In this descriptive study, 523 multiple choice questions of community health, pediatric psycology nursing groups including 12 courses at second academic semester of 2009 were studied.
The structure of questions were assessed by Millman checklist. Bloom taxonomy of the questions considered as well. Difficulty, discrimination, and other quantitative indicators were evaluated by software too.
Results: From 523 questions, 94.3% had a good structure. Most structural problems related to linear order of items (57.4%), lack of major information in question stem (32.7%), repeated words in the items (32.1%) and lack of coordination among items respect to length and vocabulary stracuture (29.8 %). 96.4% of questions were from "knowledge and understanding" taxonomy. 79.8% of questions were about learning abligation. The average score of test difficulty was from 0.49 to 0.8. Generally, 21.1% of questions were appropriate, 43.7% were very easy questions and 14.5% were too hard. Almost all of the tests had low discrimination power from 0.14 to 0.32. Only two courses had optimum discrimination power.
Conclusion: Although the structure of multiple choice tests were relatively good, but the distribution of Taxonomy, difficulty and discrimination power of the tests should be corrected organizing educational workshops for faculty members also evaluating their activities by Educational Development Center of the Universities can be effective.