Multiple-choice items are used in large-scale assessments of mathematical achievement for secondary students in many countries. Research findings can be implemented to improve the quality of the items and hence increase the amount of information gathered about student learning from each item. One way to achieve this is to create items for which partial credit can be given when students select particular incorrect options. To improve the items in this way requires a critical analysis of how the items contribute to the measure of student achievement as well as extensive knowledge of the test construct.