TL;DR:
- Stanford researchers have found that AI can enhance teachers’ performance and improve student learning and satisfaction.
- They implemented an AI tool called M-Powering Teachers to assist educators in improving their utilization of a teaching strategy called student uptake.
- M-Powering Teachers analyzes class transcripts and provides timely and specific feedback to instructors through a user-friendly app.
- Instructors who received AI feedback showed a 13% increase in uptake utilization compared to the control group.
- Students in the M-Powering Teachers group demonstrated improved learning outcomes and higher satisfaction with the course.
- The AI tool was also tested in a one-on-one mentoring program, where it increased mentor uptake by 10% and reduced their talking time by 5%.
- Future research will focus on evaluating M-Powering Teachers in K-12 in-person education, overcoming challenges related to audio quality and voice separation.
Main AI News:
Stanford researchers have made a groundbreaking discovery that has the potential to revolutionize education. According to their study, published in Educational Evaluation and Policy Analysis, machine learning AI can greatly enhance teachers’ performance and improve student learning and satisfaction. The researchers implemented the AI as a mentor for educators, specifically focusing on enhancing their utilization of a technique called student uptake, which involves acknowledging and building upon students’ contributions.
By revoicing, elaborating, or asking follow-up questions in response to student input, teachers amplify student voices and empower them in the learning process. As uptake has been correlated with positive outcomes in student learning and achievement, it is considered a fundamental teaching strategy. However, improving this skill has proven challenging.
The research team aimed to determine whether an AI tool could effectively assist teachers in enhancing their uptake abilities. Lead author Dora Demszky, an assistant professor at Stanford’s Graduate School of Education, stated, “We wanted to see whether an automated tool could support teachers’ professional development in a scalable and cost-effective way, and this is the first study to show that it does.”
Receiving timely and specific feedback is crucial for improving teaching performance, as previous research has demonstrated. Typically, feedback is provided through classroom observation, where experienced educators assess and offer actionable suggestions. The most commonly utilized observation tools in the US, such as the Framework for Teaching and CLASS, include components that measure uptake.
Unfortunately, teachers in the US have limited access to the valuable feedback that academic researchers have identified as essential. The majority of feedback comes from school administrators, who may have other responsibilities and may not provide high-quality feedback—resulting in a stressful and unsatisfactory experience for teachers.
To address this challenge, the AI program known as M-Powering Teachers was developed. M-Powering Teachers leverages natural language processing, a subset of AI that focuses on comprehending and analyzing human speech. By analyzing class transcripts, the AI identifies patterns in class discussions and translates them into actionable insights. The program measures uptake by monitoring how frequently teachers’ statements derive from student contributions, identifies questions that receive substantial responses, and analyzes the ratio of student-to-teacher speaking time.
The feedback generated by M-Powering Teachers is then provided to instructors through a user-friendly app within a few days after the class. The feedback employs positive, non-judgmental language and includes specific examples from the class transcript.
To evaluate the effectiveness of M-Powering Teachers, Demszky and her colleagues implemented the program during the spring 2021 session of Stanford’s Code in Place program—a free, online coding course that relies on volunteer coding instructors. These instructors typically possess limited or no formal training in education, making them ideal candidates for an accessible and scalable mentoring tool. The instructors were divided into two groups: one group received feedback from M-Powering Teachers, while the other did not.
The results were impressive. On average, instructors who received AI feedback demonstrated a 13% increase in their uptake utilization compared to the control group. Additionally, analysis of course surveys and completion rates of optional assignments revealed that students in the M-Powering Teachers group exhibited improved learning outcomes and higher satisfaction with the course.
To further validate the AI’s effectiveness in different educational settings, Demszky and study co-author Jing Liu, who is currently affiliated with the University of Maryland, tested M-Powering Teachers with instructors participating in the Polygence mentoring program. In this program, instructors work one-on-one with high school students. The results indicated that the AI tool increased mentor uptake by an average of 10% while reducing their talking time by 5%. The full data will be presented at an upcoming conference in July.
Demszky’s future research will focus on evaluating M-Powering Teachers in its most critical application—K-12 in-person education. However, this poses a challenge due to audio quality issues in classrooms and the difficulty of separating voices. Demszky explained, “Natural language processing can do so much once you have the transcripts, but you need good transcripts.” Overcoming this obstacle will be a vital step in realizing the full potential of M-Powering Teachers in traditional classroom environments.
Conlcusion:
The research conducted by Stanford researchers demonstrates the tremendous potential of AI in transforming education. The implementation of the AI tool, M-Powering Teachers, has shown significant improvements in teachers’ performance and student learning outcomes. By providing timely and specific feedback, AI enhances teachers’ utilization of effective teaching strategies.
This breakthrough has implications for the education market, as it offers a scalable and cost-effective solution to support teachers’ professional development. The positive outcomes observed in student learning and satisfaction indicate that AI can play a crucial role in improving educational experiences and outcomes.
