The Writing Analytics Tool: A Learning Engineering Approach to Designing AI-Supported Writing Instruction
Saved in:
| Title: | The Writing Analytics Tool: A Learning Engineering Approach to Designing AI-Supported Writing Instruction |
|---|---|
| Language: | English |
| Authors: | Danielle S. McNamara (ORCID |
| Source: | Grantee Submission. 2026Paper presented at the Learning Engineering Research Network Convening (LERN 2026). |
| Peer Reviewed: | Y |
| Page Count: | 9 |
| Publication Date: | 2026 |
| Sponsoring Agency: | National Center for Education Research (NCER) (ED/IES) |
| Contract Number: | R305A180261 |
| Document Type: | Speeches/Meeting Papers Reports - Research |
| Descriptors: | Artificial Intelligence, Writing Instruction, Computer Uses in Education, Learning Analytics, Computer Interfaces, Writing Research, Stakeholders |
| Abstract: | The Writing Analytics Toolkit (WAT) is an AI-driven platform designed to support writing instruction and research through transparent, theory-grounded writing analytics. Developed over six years through an IES-funded learning engineering initiative, WAT integrates natural language processing, machine learning, participatory design, and evidence-based instructional principles to serve students, teachers, and researchers within a single system. This paper describes WAT's architecture, analytics, and interfaces, and frames its development as a learning engineering case study. We document how stakeholder engagement, iterative design cycles, empirical validation, and scalability considerations shaped WAT's evolution. The work illustrates how learning engineering can guide the responsible design of AI-enabled educational tools that are pedagogically aligned, usable in authentic contexts, and extensible for research and innovation. [Additional funding provided by the Learning Engineering Institute at Arizona State University. This paper was published in: "Proceedings of the Learning Engineering Research Network Convening (LERN 2026)," 2026, pp. 329-336.] |
| Abstractor: | As Provided |
| IES Funded: | Yes |
| Entry Date: | 2026 |
| Access URL: | https://edtecharchives.org/conference_proceeding/2551/25363 |
| Accession Number: | ED678827 |
| Database: | ERIC |
| Abstract: | The Writing Analytics Toolkit (WAT) is an AI-driven platform designed to support writing instruction and research through transparent, theory-grounded writing analytics. Developed over six years through an IES-funded learning engineering initiative, WAT integrates natural language processing, machine learning, participatory design, and evidence-based instructional principles to serve students, teachers, and researchers within a single system. This paper describes WAT's architecture, analytics, and interfaces, and frames its development as a learning engineering case study. We document how stakeholder engagement, iterative design cycles, empirical validation, and scalability considerations shaped WAT's evolution. The work illustrates how learning engineering can guide the responsible design of AI-enabled educational tools that are pedagogically aligned, usable in authentic contexts, and extensible for research and innovation. [Additional funding provided by the Learning Engineering Institute at Arizona State University. This paper was published in: "Proceedings of the Learning Engineering Research Network Convening (LERN 2026)," 2026, pp. 329-336.] |
|---|
