Leveraging AI to Reduce Student Anxiety and Enhance Learning

Discover how one educator harnessed the power of AI to create a more supportive and engaging learning environment in their Research Methods course. By integrating principles of Non-Violent Communication and leveraging custom AI models, they were able to reduce student anxiety, foster community, and promote deeper learning - all while efficiently managing their responsibilities as a researcher and educator.

Introduction

As educators, we cannot ignore the pervasive stress experienced by our students. I hear students always say “I am so stressed!” - in the classroom, outside the classroom, and on social media. Whether due to current environmental factors, social media use, or the lingering effects of isolation during the pandemic, it is our responsibility to address this issue head-on.

Recently, I participated in a six-week Non-Violent Communication (NVC) workshop focused on personal growth. NVC is an approach that emphasizes empathy, compassion, and clear communication to foster understanding and reduce conflict. I quickly realized that the principles of NVC could be applied to alleviate anxiety in the courses I teach.

As a researcher and educator with a full workload, including administrating grants, mentoring graduate students, writing papers and grants, and a full teaching load, I knew I needed an efficient way to integrate NVC into my teaching philosophy and course design. This is where AI came in - specifically, a custom GPT model from ChatGPT that I trained on my teaching philosophy and goals for reducing student anxiety.

Enhancing the Canvas Experience

The first step was to rework my Canvas pages, starting with the landing page, to set a welcoming tone and put students at ease from the outset. After a few iterations with ChatGPT, I arrived at this inviting message:

Hello, and welcome to this course! Welcome to our Research Methods in Psychology course. Let’s approach our studies with dedication and a sense of curiosity, learning from the course materials and each other, in an atmosphere of mutual respect. Before we start, ensure to review the course manual here [Download here], familiarize yourself with Canvas, and secure your textbook. Looking forward to a rewarding semester together!

Next, I tasked the AI with rewriting the overview of graded assignments, aiming to reframe them as opportunities for growth rather than high-stakes assessments:

Graded Assignments Our journey in this course involves various forms of engagement and learning. Rest assured, every assignment is a stepping stone towards a deeper understanding rather than merely a means for grading. Each task is designed to help you explore and grow rather than simply judge your performance.

4 Exams (lowest score dropped) - Remember, an exam is not just a test; it’s an opportunity to reinforce what you’ve learned. It’s not about perfection, but progress, constituting 20% of your final grade.

Participation - Your input matters and is valuable, making up 5% of the final grade.

Reading Assignment and Summary Paper - This is your chance to engage with the material and express your understanding, which is 5% of the final grade.

Assignments Related to Research Proposal - These tasks are opportunities for in-depth exploration, accounting for 40% of the final grade.

Research Proposal Presentation - A moment to share your learning journey, worth 10% of the final grade.

Computational Assignment (SPSS based) - This component, at 5% of the final grade, invites you to apply your learned skills in a practical context.

Mandatory Research Credits (3 credits) - Participating in research enriches your educational experience and is valued at 15% of the final grade.

Each date and assignment will be posted on Canvas in advance to support your mindful planning and preparation.

The AI-generated versions strike a markedly different tone compared to traditional, matter-of-fact course overviews. By emphasizing learning, growth, and the inherent value of each student’s contributions, the messaging sets the stage for a more supportive and less stress-inducing learning environment.

Transforming experience with SPSS assignments through Discussions

One of the most anxiety-provoking aspects of my Research Methods course is the use of SPSS, a statistical software package. To address this, I leveraged AI to create engaging discussion prompts that encourage students to reflect on their learning process, share insights and struggles, and support each other.

The first discussion, titled “Embarking on the SPSS Journey: Preparing for Our Statistical Analysis Assignment” and timed to precede the SPSS assignment, invites students to consider their prior experience with SPSS, set learning goals, and surface any concerns. The AI-generated prompt is carefully crafted to create a safe space for sharing while also seeding a growth mindset:

Discussion Topic: Welcome to the preliminary discussion for our upcoming SPSS assignment. This space is for us to collectively prepare for the task ahead, familiarize ourselves with the expectations, and build a supportive learning community.

Before you dive into the data analysis, take a moment to reflect and share:

  1. Your prior experience with SPSS: Have you worked with it before, and if so, in what capacity?
  2. Your expectations for this assignment: What do you aim to learn or achieve through this process?
  3. One question or concern you have as we approach this statistical exploration.

This pre-assignment dialogue is designed to surface any collective wisdom we have in our class and address common questions or concerns early on. It’s also an opportunity for us to gauge our starting points, set our learning intentions, and ensure that everyone begins the assignment with a shared understanding and readiness.

To ensure a rich and collaborative discussion, please post your insights and questions by the specified due date. Engage with your peers by responding to at least one other post. Your initial contributions will only become visible to others once you’ve made them, encouraging original thought. Utilize the ‘like’ function to acknowledge posts that you find insightful or helpful.

Active participation is crucial. By contributing to this discussion, you will gain points in the computational assignment category. Conversely, failure to participate will impact your points adversely. Allocate your efforts wisely and with intention. Your thoughtful engagement here is the first step in mastering the analytical skills that lie ahead.

The second discussion, titled “SPSS Analysis In-Progress: Reflecting and Sharing Insights” and occurring during the assignment window, serves as a forum for reflection, community-building, and collaborative problem-solving:

Discussion Topic: As we navigate through the complexities of our SPSS statistical analysis assignment, this space serves as a reflective repository for our experiences and a beacon for collective wisdom.

As you dive into the data, consider the following prompts for reflection and discussion:

Challenges Encountered: What obstacles have you faced in manipulating and analyzing the data so far, and how have you approached these challenges?

Conceptual Insights: Share any “aha” moments or deeper understandings of the statistical concepts at play.

Technical Tips and Tricks: Have you discovered any SPSS features or shortcuts that could benefit your classmates?

Mindful Learning Moments: How are you ensuring a balanced approach to this assignment, considering the time and effort required?

Collaborative Problem-Solving: If you see a peer encountering a difficulty that you’ve overcome, offer your solution and support to help them progress.

Your participation in sharing reflections, responding to your peers, and offering support is not just encouraged—it’s a vital part of the learning process for everyone. As this assignment is iterative and ongoing, check back often to share your progress and learn from the experiences of others. This is an opportunity for growth, understanding, and community building. Let’s make the most of it!

Participate by posting your insights and questions by the due date, and respond to at least one peer to maintain a dynamic exchange. Remember, your posts become visible after your initial input, supporting unique contributions. Use the ‘like’ feature to recognize valuable posts. Remember, active engagement here will contribute to your computational assignment score, while non-participation will detract from it. Invest your efforts with deliberation. Your involvement is key to developing the analytical skills essential for this course.

The thoughtful prompts and structured opportunities for peer interaction yielded astounding results. Students demonstrated meta-learning, articulating insights like:

1) I have have some prior experience with SPSS because I used it in my Statistics in Psychology class. I made graphs and charts based off of data. However, the step by step process was very specific and we only used SPSS on the McConnell computers rather than downloading them onto our laptops.

2) I expect to learn and become more familiar with this helpful tool that I can use in the future concerning this field. I also expect to come across some difficulty at first, but I’m looking forward to having further access to SPSS through my laptop in future courses.

3) I am slightly concerned I may get confused towards the directions or any possible technical difficulties regarding the SPSS installation on my laptop.

Another student commented:

I have worked with SPSS before a few times in the context of my psych stats class. That was in my fall semester of last year, so it has been a little while and I am a bit rusty with the program.

Through the SPSS assignment, I am hoping to learn how to better use the program and potentially use the data we generate in our group’s final proposal. It will be interesting to see what we find in the results and if they support our hypothesis. It will be very beneficial to the final proposal if we are able to use the program correctly.

One of my biggest concerns that I have as we approach SPSS is how complex the system itself may be. I don’t remember a lot about the system from last year but I do remember it could be difficult if you missed a step to generate the proper data. I would like to know the specific main functions of the program and what they mean in the context of the data that is being created so I can understand what forgetting or adding something may do to what is being shown.”

And another student posted:

I have used SPSS before very briefly in my psych stats class, but we never went in depth with it and only ever did very simple things with step-by-step instructions. I feel like my professor for psych stats let me down in this sense that now I have very little knowledge and experience with using SPSS.

I hope to learn more about how to use different aspects of SPSS and become more comfortable with using SPSS.

I am really worried that I am not competent enough in using SPSS to do this assignment properly.

Importantly, students also shared strategies for managing the learning process:

1) The main challenged I encountered was trying to fully understand the instructions throughout completing the assignment. I often had to troubleshoot and figure the platform out myself. More specifically, I found it difficult reporting the output findings in correct words and format. Overall I overcame these obstacles by using the example as a reference and troubleshooting using the provided sources as a guide.

2) My biggest “Aha” moment was when I began understanding the platform and finally figured out how to install the platform onto my laptop. A deeper understanding I came to realize was when I created the box plot presenting the data with greater insight.

3) A tip I recommend is to mainly look at the data from the “data view” tab rather than the variable tab. This is because I originally looked at the “variable view” tab and it made customizing data much more confusing within the first task instructions.

4) I ensured a balanced approach throughout this assignment by giving myself plenty of time to troubleshoot and reference multiple different sources provided to help me further understand both the statistics and the platform SPSS altogether. I also made sure to take time for breaks in between the assignment to keep a fresh mind, helping me understand the instructions with a different perspective each time.

5) A common difficulty I’ve realized is their is some minor confusion within the first half of the assignment (first data table). When forming the “Descriptive Statistics” data table, the instruction provide two different guides on how to achieve this. I recommend only reviewing the “Descriptive Stats for Many Numeric Variables (Descriptives)” link because this is the main reference that helped me actually create the table itself. It took my a long time to figure out how to do this because I spent a large amount of time only referencing the first (Compare Means) link—it didn’t form the table I was looking for when referring to the sample document.

Another student posted:

Going into this assignment I was quite nervous because I felt that I lacked sufficient knowledge of how to use SPSS but once I sat down and really looked into what the assignment entailed and I dug into the handouts and instruction guides I found that it wasn’t actually too bad. I found challenges when trying to create the boxplot and run the ANOVA because I had never worked with SPSS to do either of these before but I overcame these challenges by following the guidelines included in the course syllabus. When I was doing this assignment I realized how this type of data analysis would be useful for a research project like the proposed ones we are doing in class. I figured that this assignment would require a lot of time and effort and that is why, in order to have a balanced approach, I started this assignment early, set clear expectations for myself, and set aside time specifically for working on this assignment. If you find yourself struggling with this assignment, I would say, use the step by step instructions listed in the course syllabus. That is what I used and I found it very helpful.

The discussions fostered a sense of community and mutual support, with students offering encouragement and tips to their peers:

“… you are not alone in your hesitations with any sort of math assignment! Ever since I was a child I have dreaded doing anything that has to do with math. Luckily for us, from what I remember from my minimal experience with SPSS, the software doesn’t require much mathematical thought and is designed to make statistical analysis easier for users.”

“I agree, I expect it to be challenging at first, but will get a better understanding of the program with practice.”

“I used SPSS last year in stats too but was instructed on each assignment! Im confused and worried as well but i’m so excited to learn more!”

Positive Student Outcomes

The impact of the AI-enhanced, anxiety-reducing strategies was evident in end-of-course evaluations. Students reported feeling comfortable, cared for, and focused on learning rather than just grades:

“It is a comfortable learning environment so if help is needed, ask.”

“This class is required so I would recommend it, but I also loved this class and would have taken it as an elective if I didn’t need to.”

“He makes sure all his students feel good in this class and he is always open to helping you for your specific needs.”

“He was very fair and honest with all of us, and generous in how he taught us by not grading us in a harsh manner where we had done something “wrong”, but more in a sense where we just had something to learn.”

“I thought that he prepared us very well for the next few years of school and beyond. It felt like he cared more about us learning than about us getting graded which is a very welcomed change.”

“Actually shows you what to do rather than speaks at you.”

Conclusion

This experience demonstrates the immense potential of AI to help educators efficiently weave their teaching philosophies and pedagogical approaches into every facet of their courses. By training an AI model on the principles of NVC and my goals for reducing student anxiety, I was able to create a learning environment that prioritizes empathy, growth, and community-building.

The AI-generated content and discussion prompts enabled me to provide a personalized, supportive experience for my students at a scale that would have been challenging to achieve on my own, given the competing demands on my time. The results speak for themselves - reduced anxiety, enhanced engagement, and deeper learning.

As we continue to explore the applications of AI in education, it’s clear that it has the potential to be a game-changer. By leveraging these powerful tools to amplify our pedagogical approaches and create more humane, supportive learning environments, we can help our students thrive academically and emotionally. The future of education is bright, and AI will undoubtedly play a key role in shaping it.