Assessment Design Recommendations
Reminder: These ideas are not in any way intended to be prescriptive 鈥 these are simply suggestions from our peers for ideas that have shown promise or worked for them. We acknowledge that there is no quick fix, and that upholding academic integrity takes effort.
Considerations for designing assessments
- Use course learning outcomes to inform your assessments. What is it that students should know by the end of your course and how might they demonstrate this knowledge?
- Focus on higher levels of learning. Rather than testing students only on foundational/basic knowledge, consider having them apply concepts to new situations, integrate across concepts, demonstrate the ability to compare and contrast, and so on.
- Questions that require students to analyze, evaluate, and create are inherently more difficult to look up (Google, generative A.I.) online and easier to spot when multiple students provide the same response.
- For further reading, see Waterloo鈥檚 Centre for Teaching Excellence article on .
- For further reading, see , particularly in relation to course and assessment design considerations.
听
- Incorporate choice when selecting the topic of an assessment, or randomly assign different topics to different students; this method can help to reduce the likelihood of unauthorized collaboration amongst students.
Question format
- Create multiple-choice questions that rely on knowledge application rather than memorization.
- Taking a concept and modifying it or applying it to an unfamiliar setting that wasn鈥檛 explicitly covered in lectures or readings can make it more challenging to look up the answer and gets at a deeper level of understanding from students.
- For example: if testing about colour vision, rather than asking students to identify which type of cone receptor responds to long wavelengths, you could instead ask what might happen to colour vision if you were to remove one of the types of cones.
听
- For example: if testing about colour vision, rather than asking students to identify which type of cone receptor responds to long wavelengths, you could instead ask what might happen to colour vision if you were to remove one of the types of cones.
- Taking a concept and modifying it or applying it to an unfamiliar setting that wasn鈥檛 explicitly covered in lectures or readings can make it more challenging to look up the answer and gets at a deeper level of understanding from students.
- Give students examples of research designs, schematics, maps, diagrams, or other visuals and ask them to identify variables (e.g., what is/are the dependent variable(s) in this example) or components. This can be a great way to test understanding and isn鈥檛 something that can be easily looked up online.
- Consider using a multi-select question format to ask a combination of true-false statements.
- This allows for more nuance than a typical multiple-choice question that only allows for one true or one false statement (e.g., 鈥淲hich of the following statement is TRUE?鈥). Using a multi-select format allows you to ask, 鈥淥ut of the five (5) statements below, select all TRUE statements鈥.
- For , you can choose to grade 鈥渁ll or nothing,鈥 鈥渞ight minus wrong,鈥 or 鈥渃orrect answers only.鈥 To locate the correct answer to this question on the internet or in a textbook, students would need to look up every single option to determine the correct set of responses.
听
- Consider using 鈥渁nalogy鈥 questions to test students鈥 understanding of the relationship between concepts.
- For example, if testing the concept of transduction (converting sensory stimuli into neural signals), a question might be:
Photoreceptors are to vision, as ________ are to ________.
a) hair cells; audition
b) hair cells; olfaction
c) papilla; taste buds
d) cilia; hair cells
(Students need to understand that photoreceptors are the cells responsible for transduction in vision and which of the answers has a pairing of the cells responsible for transduction and the corresponding sense.)
Open-book or take-home exams
Typically, open-book (or take-home exams) are assessments that rely on high-order thinking such as analysis, synthesis, and evaluation (rather than recall or recognition), and in which students are generally permitted to consult reference/resource materials. However, the exact definition and corresponding rules and regulations around these types of exams can vary widely.
- Provide clear and explicit instructions of what is permitted or disallowed. Definitions of open-book and take-home exams vary across faculties, departments, and course instructors, so it is therefore essential that all course instructors provide clear guidelines to students about acceptable resources, references, supports, etc.
- Considerations: as you think about what is permitted and or disallowed in your course, it is worthwhile to be explicit on your syllabus and in your communications with your students your responses to each of these questions:
- Are students permitted to access materials posted to their Brightspace course space when completing the exam?
- Can students discuss their ideas/responses with each other prior to exam submission?
- What constitutes 鈥渦nauthorized collaboration鈥?
- Are there sites/resources students can/cannot turn to for support for the completion of their exam (e.g., tutors, writing support, 鈥渉omework鈥 websites, generative A.I., etc.)
- Considerations: as you think about what is permitted and or disallowed in your course, it is worthwhile to be explicit on your syllabus and in your communications with your students your responses to each of these questions:
Alternative assessment options
The idea of an 鈥渁lternative鈥 assessment is to replace traditional (often multiple-choice style) tests with assessments that are more resistant to integrity infractions. Typically, this is because alternative assessments rely on personal connections and interpretations of specific course content, encourage creativity and innovation from students, and/or require students to demonstrate their process from beginning to end. It must be acknowledged that some of these options are not scalable for large classes, but they may prove useful in some courses (either as replacements for traditional tests or in addition to tests to spread out the composition of final course grades).
There are many options for alternative assessments; below is a sampling of ideas but more suggestions can be found in the links at the bottom of this section.