Home
FUTURE STUDENTS ask.UWindsor SUPPORT UWINDSOR

Assessment Design

Teaching and Learning Resources

Assessment Design

Principles of Effective Assessment Design

Assessment, at its core, is the practice of evaluating student learning. It can be summative, looking to assess whether or not students have achieved the relevant learning outcomes at an appropriate standard, or formative, itself a part of the learning process.

Effective assessment will have the following characteristics:

  • Aligned: assessment tasks will target the relevant learning outcomes of a course and will be consistent with teaching methods, as well as course expectations and rubrics.
  • Valid: the assessment measures what it purports to.
  • Reliable: assessments are graded consistently.
  • Fair: effective assessment will avoid unnecessary busywork or complications, and distribute workload evenly and consistently for students.
  • Create opportunities for feedback: Feedback is one of the most powerful components of teaching and is a critical tool for supporting student learning.
  • Authentic: assessment that challenges students with complex and real-world problems and questions is more likely to motivate and engage students.

Strategies for Effective Assessment

Scaffolded Assignments

Dividing large projects and assignments into smaller chunks allows opportunities for students to engage with more complex problems and questions, as well as for instructors to provide feedback to help students develop higher-level skills and complete higher quality final products.

Some Examples of Scaffolded Assignments

The following table identifies some complex projects broken down into possible steps.

Genre (Final Submission)
Possible Step 1
Possible Step 2
Possible Step 3
Literature Review
Library search of possible sources and identification of inclusion / exclusion criteria
Annotated bibliography
Compare and contrast annotated articles
Design project
Proposal
Interim report
Gaps, implications and limitations analysis
Critical Analysis
Summary of perspectives
Concept map
Synthesis of perspectives
Portfolio
Sample artifacts with Reflection
Sample artifacts with Reflection
Sample artifacts with Reflection
Research Report (Group)
Proposal (Hypothesis; context development)
Group process assessment
Results and Analysis
Recommendation / Proposal
Problem statement
Solution identification
Cost/benefit analysis

General Tips for Effective Scaffolding

  • Begin with complex higher order problems or tasks, with no one easy or right answer.
  • Clarify the role that students will play, an authentic audience for the final product, and a specific genre (or choice of formats!), to accompany the task.
  • Where possible, build in variety in the steps in terms of process and format (and ease for relying on AI for completion).
  • Provide formative feedback on early stages that guides students to effective completion of later stages.
  • Emphasize inclusion of specific details, contexts, considerations.

Process-Based Assignments

Focusing on learning processes is a strategic way to embed both formative and summative assessment into assignments, and to ensure that assessment is not just about judging students, but also directing and fostering their learning. Through process-based assessment students become more aware of the varied skills and abilities required to be successful in a discipline, and to become more strategic in identifying their strengths, weaknesses, and pathways to improvement. Some examples of process-based assessment:

  • Brainstorming
  • Concept-Mapping
  • Outlining
  • Reflection Assignments
  • Editing and Revision

Impacts of Generative Artificial Intelligence (GenAI) on Assessment

What is GenAI?

“Generative AI is technology that creates human-like content – including text, images, video and computer code – by identifying pattens in large quantities of training data, and then creates original material that has similar characteristics.” (From University of Windsor, Bylaws 54 & 55: Sample Course Syllabus Statements on the Use of Generative Artificial Intelligence.)

Is Use of Artificial Intelligence a Breach of Academic Integrity?

Artificial Intelligence (AI) carries significant potential for supporting learning. While there’s no doubt that, in some cases, use of AI would be a breach of academic integrity, in other situations, AI may help increase learning and knowledge acquisition, while still adhering to the core principles of fair, cooperative, and honest inquiry. The line between integrity and misconduct can be quite fuzzy, difficult to draw, and highly dependent on context.

Consider:

  • Using AI to build a study tool that mines digital textbooks or course LMS sites for information, then generating flashcards or simulations that students can use to test themselves on the material.
  • Using AI to generate actionable feedback on assignments.
  • Using AI to generate or help organize ideas.
  • Using built-in AI tools for fixing grammar, spelling, and ensuring readability of a written assignment.
  • Using AI to extract patterns from very large datasets.

What motivates the use of Artificial Intelligence (AI)?

GenAI has many useful functions, and many more are likely to come that will help make tasks easier and more efficient.

  • Ability to gather information that is (rightly or wrongly) perceived to be unbiased and accurate
  • Idea generation
  • Efficient or easy completion of tasks
  • “Cognitive offloading” (Perkins, 2023)
  • Ability to engage conversationally or obtain feedback for tutoring or testing oneself
  • Perceived ability to obtain higher grades, where focus is on grades, not learning
  • It’s the new thing that everyone is talking about

While all of these uses can be entirely legitimate, there are times when this use will overstep the goals, expectations, and rules for assessments, and would constitute a breach of academic integrity. Interestingly, when compared to the standard reasons that students cheat at university there is some interesting overlap. For more details, please see our page on Academic Integrity.

Strategies instructors can use to promote Academic Integrity in a world of GenAI

While it may be tempting to approach the use of GenAI from a surveillance and punitive lens, this approach is not recommended. Despite advances in AI detection tools, advances in the quality of AI outputs are typically at least one step ahead. False positives and false negatives are common, and there are no tools that can accurately distinguish between human and virtual text. Instead, we recommend taking a proactive formative approach, enhancing classroom culture and assignment design to engage in whole person learning, and supporting student skill development and learning through scaffolded and process-based assessments.

Whole-Person Learning

Learning involves much more than simply the acquisition of knowledge, and ability to learn can be affected by a whole host of factors including emotional, mental, and physical health; current life exigencies; and a sense of community or belonging. Students and instructors alike bring their whole selves into the classroom, and course, assessment, and lesson planning that acknowledges this can help mitigate both visible and invisible barriers to learning.

Some general considerations for whole-person learning include, first and foremost, recognition of the importance of relationality to the classroom. Developing strong student-teacher relationships, as well as peer-to-peer relationships in the classroom can foster a sense of belonging, a collaborative community of inquiry, and engagement with the course materials. Providing opportunities for self-directed learning and student autonomy can both boost skills and enhance the perceived relevance of the subject matter to student goals. Feedback that focuses on students as authors/researchers, rather than simply on product content or lower-level skill acquisition highlights for students their contributions to the academic discourse. This recognition of “life happens” and we’re all working together enhances motivation, a sense of purpose, and fosters a climate of academic integrity.

Some Further Examples of Whole Person Learning

  • Getting to know students, by name where possible, and their interests and motives beyond the classroom.
  • Providing choice in assignments; flexible formats; allowances for “life happens”.
  • Creating opportunities for students to identify their own goals for learning, and to reflect on their achievement of those goals.
  • Emphasizing decolonizing approaches to academic integrity as involving mutual recognition, respect, sharing and responsibility (Poitras Pratt & Gladue, 2022).

Scaffolding and Process-Based Assignments

While GenAI might be used to generate responses to almost any written assessment (including providing scripts, outlines, and background information for video, audio, or presentation-based assignments), responses tend to lack specific, concrete, or personal details, and so requiring the inclusion of those details can mitigate the advantages for students to offload these cognitive skills onto other tools. Additional strategies:

  • By focusing more on the rationales and reflections behind choices and tasks, there can still be learning, even when some shortcuts are taken.
  • When breaking up assignments into smaller components, there will always be some that are easier to generate with AI than others. However, simply dividing a larger assignment into discrete steps can helps make it easier and more advantageous for the student to invest their own thinking and planning into each step, as it can otherwise be quite difficult to create coherence and connection between different tasks.

Some Examples

The following table identifies some specific types of process-based assignments, the specific challenges that might be encountered in a world of AI, and a few strategies for creating effective prompts to further support learning.

Activity/Skill
Possible aims
Challenges
Effective prompts
Brainstorming
Foster idea-generation and divergent perspectives
AI often recommended for brainstorming; has access to massive amounts of data; but also biased and limited to “most common”
Draw on group activities for brainstorming to encourage further diversity
Reflection
  • Encourage self-reflection/ analysis of personal strengths and weaknesses
  • Encourage reflection on learning, achievement and areas to grow
  • Encourage reflection on course material, state of knowledge in the field
Superficial responses are easily mimicked
  • Require specific detail that includes course material and classroom conversations
  • Ensure prompts are specific to the type of reflection, including this in the marking scheme
Summary
  • Develop skills in realizing relevance
  • Develop writing skills
Abstracts and summaries are easily generated; another very well-known convenience of AI
Try in-class/in person minute papers or quick summaries with debrief for topics that can be easily copied
Question / Hypothesis Generation
Build effective research skills
Questions can easily be generated with little to no thought
Have students develop a rationale for the question that embeds relevant course material
Data Analysis
Foster high-level quantitative and numeracy skills
AI is well known for its ability to analyze large datasets but relying on automated processes will leave students lacking in critical skills
  • Use in-class, lab or tutorial time for a more proctored environment
  • Provide unique or different datasets to different student groups
  • Generate datasets unique to each class, and consider providing different datasets to different students/groups
  • Situate prompts in specific problems likely to be of interest to students.
Evaluation and Synthesis
Develop higher order critical thinking skills that are typically the aim of higher education
Despite claims that AI is not especially effective at creating effective, informed evaluations, it can easily pull commonly used rationales and string them together in plausible arguments
  • Build in reflections on process for these types of assignments, encouraging students to think about what they might do differently in the future, what they have learned, etc.
  • Consider integrating AI literacy into the assignment
Writing
Communication skills are a key graduate attribute and degree level expectation, and effective writing is a cornerstone supporting these skills
AI is well-known to produce outputs that can pass exams and assignments at the university level
Build in peer review or presentation opportunities, where students will also need to speak on their topics
Citation
Participation in disciplinary conventions; understanding of academic discourse
Easily falsified; technicalities of precise citation less important with easy copy/paste and bibliography generation tools
Add to the citation task a requirement for students to provide a rationale about citation choice and placement
Curiosity
Learning is built on a foundation of curiosity; involving both humility around not knowing and caring to know
AI offers convenience that sometimes makes it easier to offload cognitive work and remain incurious
  • Identify areas or provide choices that allow students to pursue their own natural curiosity
  • Emphasize implications and connections that allow students to see the relevance of the problem

Building GenAI Literacy in the Classroom

As AI is becoming more commonplace and integrated into industry and everyday life, AI literacy is increasingly important. Its value in providing effective, efficient tools to support a variety of tasks is mitigated by the challenge that it carries bias, makes mistakes, and itself has no integrity, no orientation towards truth, and no sense of fairness or honesty.

Because of these limits, it is important for instructors and students alike to understand how it works, and to have the skills to evaluate any AI outputs and recommendations. In the same way that a mathematician must understand the concepts of calculus or hyperbolic geometry when using a calculator for computation, using AI effectively requires an understanding of underlying concepts and how the algorithm employs them.

Some Examples of Integrating AI Literacy into the Classroom

Modelling: Various means can be used to model AI literacy, including generating AI output, sharing it with students, and demonstrating processes of evaluation by articulating the strengths and weaknesses of that output. Another strategy might be discussing case examples where AI output has been used both successfully and unsuccessfully.

Assignments that require AI use: Many industries require knowledgeable use of AI, and so assessments that help students learn how to use it effectively can help them develop skills that will serve them well in the workplace. Examples might include permitting students to use AI as a companion for specific stages of an assessment; or perhaps drawing on AI for feedback or self-testing tools.

Caution: If requiring students to employ AI, it is important to consider the degree to which you may also be requiring students to hand over their intellectual property to the AI machine, as well as to provide free labour and effort to support its training. As AI does not have integrity, and is arguably itself a plagiarist, there are ethical dilemmas surrounding its use, and the expectation that others should use it – and because of this, students may not wish to participate.

Provide AI-generated materials for validation and critique: Instead of asking students to engage actively with AI, another approach might be to generate your own materials, and then have students work to validate and critique those materials. These types of fact-checking exercises can be constructed as formative assessments or as in-class activities and can help students see both the benefits and limitations of AI generated claims.

Brainstorming Tool: AI can be effectively used for brainstorming, jumpstarting ideas when confronted with a blank page, or perhaps even more potently, for encouraging students to think beyond superficial and commonplace answers to vexing problems (e.g. Karout & Harouni, 2023).

GenAI and Academic Integrity References

Baidoo-Anu, D., & Owusu A. L., (2023). Education in the Era of Generative Artificial Intelligence (AI): Understanding the Potential Benefits of ChatGPT in Promoting Teaching and Learning. Journal of AI. 7(1), 52-62.

Karout, D. & Harouni, H. (2023, June 14). ChatGPT is Unoriginal-- and Exactly What Humans Need. Wired. https://www.wired.com/story/chatgpt-education-originality

Kumar, R., Eaton, S.E., Mindzak, M., Morrison, R. (2024). Academic Integrity and Artificial Intelligence: An Overview. In: Eaton, S.E. (ed.) Second Handbook of Academic Integrity. Springer International Handbooks of Education. Springer, Cham. https://doi.org/10.1007/978-3-031-54144-5_153

Perkins, M. (2023). Academic Integrity considerations of AI Large Language Models in the post-pandemic era: ChatGPT and beyond. Journal of University Teaching & Learning Practice, 20(2). https://doi.org/10.53761/1.20.02.07

Poitras Pratt, Y., & Gladue, K. (2022). Re-defining academic integrity: Embracing Indigenous truths. In Eaton, S.E. and Christensen-Hughes, J. (Eds.) Academic integrity in Canada: An enduring and essential challenge (pp. 103-123). Cham: Springer International Publishing.

© Created by Allyson Skene, 2024

Additional Assessment Resources

Case-Based Learning

Problem-Based Learning

Inquiry-Based Learning