Automated Feedback Coach: Student AI tutor to write better feedback now available in more regions and tools

A tool to promote feedback skills and personalized learning

An important part of the feedback process is letting students know how to deliver quality and constructive feedback. In fact, students’ feedback often falls into the either too short, too positive, or too negative spectrum. This is due to a lack of guidance and also an unwillingness to complete the activity. It is also challenging and time-consuming for instructors to follow, and provide instruction for each student in their feedback delivery process. That's why we developed the Automated Feedback Coach to assist higher education educators in guiding students to deliver better feedback.

With this plug-in, instructors can choose to activate the feature to automatically generate real-time suggestions when students’ feedback is: too short, too general, sounds like a personal attack, overly positive or negative, or overlapped with other given feedback. This AI plug-in also identifies and praises students for quality feedback comments. In Spring 2023, we further improved the Automated Feedback Coach using the advanced Large Language Models provided by Azure OpenAI Service to process students’ input and provide feedback. Azure’s OpenAI Service was chosen to deliver this feature because of its enterprise-level security, compliance, and regional availability. Compared to version 1.0, it is much faster, more scalable, and able to provide more specific, actionable feedback based on the review context.

Since its launch in the Group Member Evaluation tool to limited partners in the US and Europe, the Automated Feedback Coach has garnered increasing interest among our users, with many of them inquiring about its availability in other languages and FeedbackFruits solutions, specifically the Peer Review tool.

That’s why in our latest release, the Automated Feedback Coach has now become available in the Peer Review tool and can provide students with feedback suggestions in Dutch and French, besides English.

Automated Feedback Coach supports institutions in creating inclusive learning experiences

Since its first launch in 2021, over 28.000 teachers and students across North America, Europe, and Australia have used the Automated Feedback Coach version 1.0 in 567 activities. Our private beta-enhanced version of the Automated Feedback Coach has been used by 50 learning designers and teachers from 23 institutions around the world.

By providing real-time, personalized suggestions to guide students in writing feedback to others, Automated Feedback Coach supports a learning environment that nurtures feedback skills, critical thinking, and self-regulation. Based on interviews with many of our users, we found that they enjoyed the tool’s potential to promote a more personalized, fair, and inclusive learning experience. Furthermore, the feature’s inclination to suggest respect, understanding, and offering help to the team members was described as in line with the DEI efforts to promote openness and inclusivity among diverse learners.

‍

‍

Other software, such as plagiarism detectors, simply detect and pinpoint the mistakes in students’ writing, and also offer automatic adjustments of the sentences. However, these tools only provide short-term solutions to student learning. The Automated Feedback Coach, on the other hand, plays the role of a personal coach who offers suggestions that help students adjust and improve their feedback delivery during the peer and group assessment activity. Most importantly, this process motivates the students to develop critical self-reflection and collaborate better with each other. As the tool lets students decide on which feedback suggestions to follow, they are motivated them to take ownership of their own learning. This, accordingly boosts a sense of self-regulation and active learning. As for instructors, the tool saves them plenty of time from having to go through each student feedback and allow for focus on facilitating the entire student cohort.

The easy-to-use, user-friendly interface of Automated Feedback Coach also significantly reduces the learning curve for both instructors and students, allowing for quick and easy mastery of the technology. Hannah Silber, a lecturer at the University of Wisconsin School of Engineering piloted this feature with 20 students from 2 of her summer courses. She reported the feature to be very easy to set up and use.

“None of the students had trouble using it, which is pretty big. As an instructor, I like how easy it was to use because it means that in our department, it will be easier for other instructors to want to give it a try in their courses as well.”

So how easy and convenient it is to use Automated Feedback Coach to enhance students’ feedback quality? Below you can find how the tool works.

Easy and smooth activation

Within Group Member Evaluation and Peer Review, instructors can easily enable Automated Feedback Coach by toggling on the option “Automated Feedback Coach” in the “Guiding students” drop-down bar of the Given Reviews module.

‍

Instant, personalized suggestions on students’ feedback

After the feature has been activated, students enrolled in that activity will see "Automated feedback coach" in their feedback-giving window. Once the student starts writing the feedback, the AI coach will process their reviews and generate suggestions to help students adjust their reviews. The AI-generated feedback suggestions in Automated Feedback Coach aim to:

• Be helpful and actionable
• Ensure that students’ feedback addresses the given criteria
• Emphasize and give compliments on quality feedback comments
• Prevent students from giving feedback that is: random or unrelated; too short or general; sounds like a personal attack; is overly positive or negative; overlaps with previous feedback
• Highlight areas of improvement when giving feedback

Furthermore, the suggestions automatically update as students adjust their reviews.

‍

‍

Students can receive AI feedback in the review dialog and the sidebar in the media viewer within Peer Review and Group Member Evaluation.

‍

According to Hannah, her students enjoyed using Automated Feedback Coach, remarking that:

“The coach helps them give more detailed feedback and that the coach gave them good tips on what to improve on and assisted in giving detailed responses about specific things: whether it be my teammates’ attention to details, punctuation, organization, I think they are actually enjoying getting more specific feedback about themselves.”

‍

Promote an active and personalized learning environment

With the goal of cultivating a student-centered learning experience, we designed Automated Feedback Coach so that students can take ownership of their feedback process, by choosing whether to enable the feature, their preferred language of feedback suggestions, and reflect on their experience with the tool via a Satismeter survey.

‍

Secure data privacy and protection

We make sure that all students' data is processed in compliance with FeedbackFruits' general Privacy policy.

When using Automated Feedback Coach, all students' data is processed in the following four stages:

1. The data from students including their written feedback, prompts, and criteria go through the Azure Open AI Service to be processed and sent to a model. This process is called tokenization.
2. Once the data are tokenized and arrive at the model, the model then creates a generation based on the custom instructions (created by FeedbackFruits) and an input (your “tokens”).
3. After the model completes its generation, the instructions, input, and generation are all sent through the Azure OpenAI Service’s Content Moderation System and Abuse Monitoring System to ensure both the input and the model’s response are appropriate, do not violate any terms of service, and not harmful to you. In the Abuse Monitoring System, Azure OpenAI Service may store your input data as well as the model generation for up to 30 days.
4. Finally, the model’s generation is returned to FeedbackFruits in a standard JSON format from Azure through their API. This only contains the completion data as well as some relevant metadata including the reason for stopping and if there was an error.

In the above process, students’ input is only processed by the language models hosted at Azure to generate automated feedback. The input is never used to train, retrain, or improve any of the Azure OpenAI Service’s Models.

Moreover, we use the Generative AI models as a “plug-and-play” service; therefore not retrain or fine-tune the models for a specific use case outside of providing custom instructions to the model. In other words, we do not use any student data to retrain, train, or improve our models.

Read more about our data privacy policy in this article.

Trying out the latest Automated Feedback Coach

If you are interested in trying out this plug-in for your feedback activities, the FeedbackFruits team is here to help! Schedule a meeting with us and let us help you create quality learning experiences that promote a personalized and self-directed learning environment.

We are working to make Automated Feedback Coach is available in more regions, especiallt Canada and Asia-Pacific.

In case you need more details on the Automated Feedback Coach, here are some readings prepared by our team to help you get started:

• About Automated Feedback Coach
• 1-page brochure