Date
Publisher
The Learning Agency Lab
Teachers must be on AI writing tools to impact student outcomes, and our research highlights the importance of involving teachers at various stages of the AI development process from data collection to evaluation of performance in order for these algorithms to have a significant impact. Our research also showed that machines alone are not enough, and AWFTs are best viewed as tools to support teachers rather than a cure-all. High-quality algorithms are able to demonstrate human-comparable accuracy in evaluating student writing, showing machines can be as effective as humans in evaluating student work and indicating the potential of algorithms to provide valuable feedback.
Why it matters
To impact student outcomes, assisted writing feedback tools cannot be treated as a cure-all. These algorithms require teacher involvement from data collection to performance evaluation to have a significant impact.
Key findings
- High-quality algorithms evaluated student writing with accuracy comparable to humans, demonstrating the potential of AI to provide valuable feedback.
- Assisted writing tools were not enough on their own and functioned best to support teachers rather than serve as complete fixes.
- Involving teachers throughout algorithm development, from data collection to performance evaluation, was critical for tools to significantly impact student outcomes.
Summary drafted with AI from the paper’s abstract and may contain errors. See the full paper for details.
What is the application?
Who age?
