Writing Great Multiple-Choice Exams

Introduction

Multiple choice exams are ubiquitous in most colleges and universities. Despite being one of the most popular forms of assessment evidence, many of us are not trained in how to write great exams and, as a result, we unintentionally replicate exams we have encountered in the past, including their flaws.

“What Will Be on the Test?” 

Students often ask this question as the exam date draws nearer. As Parkes and Zimmaro (2016) write, when a student asks "what will be on the test," they are usually asking you for three distinct pieces of information: 

1. What will be on the test? (Content)
2. How will it be on the test? (Ways of knowing/what students will be asked to do)
3. How much of it will be on the test? (Proportion of the exam)

In order to write a great multiple choice exam, you need to be able to answer these questions clearly and intentionally. After all, if you're not sure how to answer these questions, how will your students know how to properly prepare for the exam? 

1. What will be on the test? The content of your exam should be explicitly connected to your course learning outcomes and the instructional material you used to teach toward those outcomes in your course. When writing an exam, ensure that each exam question ties to specific learning outcomes. 
2. How will it be on the test? This question should be answered in two ways.
   1. First, consider what kinds of questions the exam will entail (i.e., multiple choice, true-false, matching, fill in the blank, and/or short answer.)
   2. Second, consider the "ways of knowing" that you'll ask students to demonstrate with the exam. One of the most ubiquitous taxonomies for "ways of knowing" is Bloom's Taxonomy of ordered cognitive process dimensions. 
3. How much of it will be on the test? All instructors make considerations for the timing and volume of course topics/content. The considerations that shape how you allocate instructional time and activities should also factor into how you allocate exam time and proportion (i.e., the number of questions on a particular topic and how long students should spend on them relative to the total exam time). 

Writing Great Multiple Choice Questions

Multiple choice questions (also called exam "items") consists of: 

• The Stem: the statement or question that the student reads first, which prompts them to think about the answer,  
• The Key: the correct answer, and 
• Distractors: the other options which are not the correct answer.

Great Stems:

• Clear and concise: Lengthy and unclear questions can easily confuse even the best-prepared students, can prompt students to guess the right answer, and can create anxiety and frustration, which adds to students' cognitive load and diminishes their performance (Suskie 2018).
• Direct is best: Pose questions and/or statements in a straightforward way by logically ordering the answer to come after the question or at the end of the statement. Avoid "twisting" the stem by placing the answer at the beginning of the sentence because this creates excess cognitive load.
  • Example of a Direct Stem: Which measure of central tendency is most susceptible to outliers in the data? (Or: The measure of central tendency most susceptible to outliers in the data is _______.)
  • Example of a Twisted Stem: __________ is the measure of central tendency most susceptible to outliers in the data.
• Limit unnecessarily complex or technical language and prioritize language that is clear and accessible: Ambiguous terms and statements might lead students to answer incorrectly even when they know the correct answer (Brown, 2005). 
• Avoid negative phrasing: Stems that ask students to select the answer that is NOT correct are easily misread and are usually less reliable than direct stems (Haladyna 2004). 

Great Keys:

• Unambiguously correct: The key needs to be fully correct for the stem. It cannot be mostly correct, sometimes correct, or the best of bad options. The key must be the unequivocally correct answer. 
• Based on consensus from the field: If the key is arguable, meaning there is no clear consensus in the field that it is the unambiguously correct answer, it is not suitable for a multiple choice question. Instead, have students write a short answer/essay question to analyze, evaluate, and argue for/against the key.
• Rooted in course material: You should be able to locate the unambiguously correct answer in course material that ALL students had access to, such as reading, lecture materials, etc. 

Great Distractors:

• Plausible: That is, distractors should be options that are plausible answers. A distractor that is obviously non-plausible has no assessment value.
• Reflective of common misconceptions: Plausible distractors have assessment value when rooted in common misconceptions or mistakes. If a student selects a distractor, you can provide feedback about the misconception likely at work in their learning. 
• No "Gimmes" and no "Gotchas": "Gimmes" consist of non-plausible distractors or exceedingly obvious keys that have no assessment value. "Gotchas" are trick questions or unnecessarily complex questions designed to "catch" students. As Parkes and Zimmaro (2016) write: "gimmes and gotchas are problematic because they signal... an us-vs.-them dynamic which draws attention to the test and the grade and away from the learning" (8).

Promoting Learning with Exam Blueprints

"Blueprinting encourages intentionality in assessment design, promotes alignment across teaching, learning, and measurement, and provides a defensible foundation for evaluating whether an assessment meaningfully supports instructional goals (Hutson, et al, 2026).

Exam Blueprints are tools used to design and communicate about multiple choice exams. A well-designed exam blueprint will be able to answer 3 important questions: 

1. What will be on the test? (Content)
2. How will it be on the test? (Ways of knowing/what students will be asked to do)
3. How much of it will be on the test? (Proportion of the exam)

An exam blueprint is a table that presents clear answers to these questions. The columns of the table specify particular topics or content areas that will be featured in the exam. The rows specify the types of questions that will appear in the exam and the kind of thinking/knowing that will be assessed in the question(s). The cells of the table specify the proportion allocated to the question type and thinking level. 

Here's an example exam blueprint for a fictional Sociology course, based on a 50-question exam: 

Ideally, design your exam blueprint before designing your instructional activities. Developing an exam blueprint will help you ensure alignment with your learning objectives and help you plan and prioritize necessary teaching topics and activities.

Once you have an exam blueprint, share it with students ahead of time. Show your students how to read the blueprint and engage them in a class discussion about high-impact study practices that will be particularly helpful for the exam, based on the topics, quantities, and ways of knowing that will be featured in the questions.

Conclusion 

As you can see from all of these considerations, writing great multiple choice exams requires intention, care, and ongoing efforts to keep your exam, learning outcomes, and instructional activities in alignment. 

As much as possible, adopt a student-centered, assessment-oriented mindset toward multiple choice exams. 

Keep your exam student-centered by designing the exam and writing questions in ways that will get the best from your students, and prepare them for success with the exam by aligning your instructional approaches with what you will ask students to do/know on the exam.

Keep your exam assessment-oriented by prioritizing assessment, not only evaluation. Ensure every question is directly connected to your learning outcomes and write your exam in a way that will generate valid and reliable evidence of student learning. Importantly, use the results of the exams to generate actionable information for you and meaningful feedback for your students. 

• References
• Brown, J. D. (2005). Testing in language programs: A comprehensive guide to English language assessment (New edition). New York: McGraw-Hill.
  Clegg, V. L, & Cashin, W. E. (1986). Improving Multiple-Choice Tests. IDEA Paper No. 16. Retrieved from: www.theideacenter.org.
  Haladyna, T. (2004). Developing and Validating Multiple-choice Test Items. Ebook: Ebrary, Inc.
  Hutson, B., Schmauder, R., Daughterty, K., Eagle, D., McCarthy, K., McCauley, S., & Taylor, J. N. (2026). Framework for Use of Multiple-Choice Questions. Intersection: A Journal at the Intersection of Assessment and Learning, 7(3). https://doi.org/https://doi.org/10.61669/001c.165518.
  Nilson, L. B. (2016). Teaching at its best: A research-based resource for college instructors (4th ed.). San Francisco, CA: Jossey-Bass.
  Suskie, L. (2018). Assessing student learning: A common sense guide. (3rd ed.). San Francisco, CA: Jossey-Bass.
  Zimmaro, D. M. (2016). Writing good multiple-choice exams. Measurement and Evaluation Center. University of Texas at Austin. Retrieved from https://facultyinnovate.utexas.edu/sites/default/files/writing-good-multiple-choice-exams-fic-120116.pdf