By Sahra-Josephine Hjorth
Today I am going to explore the crucial role of qualitative feedback in Student Learning, focusing mostly on qualitative feedback. I place emphasis on the importance of qualitative learning because this topic is often overlooked within an AI context. It has historically been easier to create AI-based systems that provide quantitative feedback. Like s numerical grade. However, advances in Generative AI, are providing grounds to reconsider the possibilities within the field of automated qualitative feedback. First, let’s take a look at what Qualitative automated feedback is:
Automated Qualitative Feedback is a technology-driven process that provides detailed and nuanced assessments of a student’s performance, focusing on specific aspects such as language proficiency, argument strength, tone, and other qualitative elements, to support their learning and skill development.
Automated Qualitative Feedback is a technology-driven process that provides detailed and nuanced assessments of a student’s performance, focusing on specific aspects such as language proficiency, argument strength, tone, and other qualitative elements, to support their learning and skill development.
So why is Qualitative Feedback so important, and can the process of delivering qualitative feedback be automated using Artificial Intelligence?
Qualitative Feedback Enhances Learning Outcomes
Picture this: You receive an ‘A’ on a math test. Does that tell you what areas you excelled in or where you might need improvement? Not quite. And this was something that bothered me quite a bit when I taught at University. We need to provide more insights into the student’s strengths and areas for improvement before a test, but certainly also after, leveraging qualitative and quantitative techniques. Researchers agree with this point, an example is Hattie and Timperley in their 2007 research paper “The Power of Feedback” argue that “Both qualitative and quantitative feedback provide students with a more comprehensive understanding of their performance,”. A comprehensive understanding allows students to pinpoint areas for improvement and take action accordingly. In essence, it’s not just about the grade; it’s about the insight that comes with it.
We need to provide more insights into the student’s strengths and areas for improvement before a test, but certainly also after, leveraging qualitative and quantitative techniques.
Qualitative Feedback Promoting Self-Regulated Learning
Qualitative feedback, often in the form of written comments or verbal communication, goes beyond the numbers. It encourages students to engage in self-regulated learning. As Nicol and Macfarlane-Dick describe in “Formative Assessment and Self-regulated Learning,” qualitative feedback helps students reflect on their work and make necessary adjustments. It’s about empowering students to be in control of their own learning journey.
Qualitative feedback Fosters Motivation
I always felt my students took their assignments and themselves more seriously, the more available you are to them, their needs, and ideas. We know that students are motivated by getting a good grade, but, Butler’s 1987 study, “Task-Involving and Ego-Involving Properties of Evaluation,” reveals an interesting twist, that the grade itself is not the only motivator, but that “the format of feedback messages can have a dramatic effect on motivation,” Quantitative feedback or the quest for a high grade may get students motivated at first, but qualitative feedback provides the context and guidance needed to maintain and increase motivation. It’s the fuel that keeps the learning engine running.
The combination of two feedback forms Enhance Assessment Validity
Combining both qualitative and quantitative feedback ensures the validity of assessments and provides the student with more holistic feedback. As Black and Wiliam discuss in “Assessment and Classroom Learning,” assessments designed for formative learning, which includes qualitative feedback, are more effective in promoting quality learning. This approach provides a holistic view of student performance, reducing the potential for bias and errors in assessment So the arguments for providing students with both qualitative and quantitative feedback are compelling. It’s not a matter of one versus the other but rather a harmonious blend of both that enriches the learning experience. By doing so, we empower students to understand their performance comprehensively, engage in self-regulated learning, stay motivated, develop essential communication skills, and enhance the validity of assessments. Feedback, when delivered in this balanced manner, truly becomes the compass guiding students toward their educational success. So, we need more tools that can help provide Automated Qualitative Feedback at scale, just as we have seen advances in Automated Grading.
Examples of use-cases for Automated Qualitative Feedback:
Taking automated feedback to the next level involves the ability to provide qualitative feedback on various aspects of a student’s work, whether it’s a speech, an essay, or any other form of communication. Here are some examples of qualitative feedback that can be automated by AI. Within the next two years, we will see these capabilities within most learning platforms and apps:
1. Speech Analysis: AI can analyze spoken language for factors like clarity, pronunciation, and fluency. It can provide feedback on articulation and suggest improvements for better communication.
2. Punctuation and Grammar: Automated feedback can identify and highlight errors in punctuation, grammar, and syntax, helping students improve the correctness of their written and spoken language. Here we already see Grammerly as a clear industry leader.
3. Persuasion and Argumentation: AI can assess the strength of arguments by analyzing the use of evidence, logic, and rhetoric. It can provide feedback on the persuasiveness of an argument and suggest ways to enhance it. We are innovating within this area at present here at CanopyLAB.
4. Tone of Voice: In the context of speech, AI can evaluate the tone of voice, considering factors like volume, pitch, and modulation. It can provide feedback on the appropriateness of tone for the given context.
5. Clarity and Coherence: Automated feedback can assess how well ideas are organized and connected in written or spoken communication. It can point out areas where the text or speech may lack clarity or coherence. This is another area we are building in at present.
Automated feedback can assess how well ideas are organized and connected in written or spoken communication. It can point out areas where the text or speech may lack clarity or coherence.
6. Language Style: AI can analyze the style of language used in writing, such as formal, informal, or academic. It can provide feedback on whether the chosen style aligns with the intended audience and purpose.
7. Use of Evidence: In academic writing or persuasive speeches, AI can evaluate how well evidence is integrated. It can suggest improvements for better incorporation of data and references. Any company that works within NLP should have a fairly easy time innovating within this space.
8. Engagement and Relevance: AI can determine the level of engagement and relevance of content to the topic or audience. It can offer feedback on how to make the content more engaging and pertinent.
9. Cultural Sensitivity: In a global context, AI can provide feedback on cultural sensitivity, ensuring that written or spoken content respects diverse perspectives and does not inadvertently offend. This is a bit controversial of course, because is it really an AI system’s job to censor the student’s perspective? Also, some ongoing conflicts and topics do not have clear-cut answers, so it is essential to not build a system that sides with one party over another.
10. Delivery Skills: For oral presentations, AI can evaluate factors like eye contact, body language, and pace. It can offer feedback on the effectiveness of the speaker’s delivery. That way your roommate, lover, or spouse does not have to listen to you practice your big speech 50 times before exam day.
For oral presentations, AI can evaluate factors like eye contact, body language, and pace.
AI-powered qualitative feedback has the potential to offer a holistic assessment of a student’s performance, providing insights into both the technical aspects and the nuances of effective communication. This level of detailed feedback can be a valuable tool for educators and learners, helping them continually refine their skills and capabilities. I am personally incredibly excited to see the edtech industry adopt more Automated Qualitative Feedback!
AI-powered qualitative feedback has the potential to offer a holistic assessment of a student’s performance, providing insights into both the technical aspects and the nuances of effective communication.
References
1. Hattie, J., & Timperley, H. (2007). “The Power of Feedback.” Review of Educational Research, 77(1), 81-112.
2. Nicol, D. J., & Macfarlane-Dick, D. (2006). “Formative assessment and self-regulated learning: A model and seven principles of good feedback practice.” Studies in Higher Education, 31(2), 199-218.
3. Butler, R. (1987). “Task-Involving and Ego-Involving Properties of Evaluation: Effects of Different Feedback Conditions on Motivational Perceptions, Interest, and Performance.” Journal of Educational Psychology, 79(4), 474-482.
4. Topping, K. (1998). “Peer Assessment between Students in Colleges and Universities.” Review of Educational Research, 68(3), 249-276.
5. Black, P., & Wiliam, D. (1998). “Assessment and Classroom Learning.” Assessment in Education: Principles, Policy & Practice, 5(1), 7-74.
