Visit our sister organization:

Visit our sister organization:

Visit our sister organization:

ASAP 2.0 Dataset

Download

Overview

As many educators know, grading essays by hand is hard, time-consuming and expensive. Automating the essay-scoring process could mean untold efficiencies for teachers and faster feedback for students.  Yet the first  Automated Student Assessment Prize (ASAP) competition to tackle grading student-written essays was held twelve years ago.

Student writing instruction and the potential for better essay scoring has leaped forward in that time. A newly-updated ASAP 2.0 dataset competition that concluded in July 2024 now makes it possible to better train models that can support overtaxed teachers in providing more timely feedback, especially in underserved communities.

More reliable techniques and automated writing evaluation systems (AWEs) could also allow essays to be introduced in testing, a key indicator of student learning that is currently commonly avoided due to grading challenges.

While previous efforts to develop open-source AWEs were limited by small datasets that were not nationally diverse or focused on common essay formats, the ASAP 2.0 competition hosted by the Learning Agency Lab and Vanderbilt University provided developers with a more expansive opportunity. The competition utilized the largest open-access writing dataset, incorporating  about 24,000 student-written argumentative essays, aligned to current standards for student-appropriate assessments. The ASAP 2.0 dataset also included samples across economic and location populations to mitigate the potential of algorithmic bias.
View algorithms, including competition-winning ones, that were trained on the dataset here.

Ultimately, ASAP 2.0 data includes the ability to ease existing hesitations teachers may have in assigning essays. Improved automated essay scoring could also aid students who often realize improved learning benefits and outcomes via essay writing.

This dataset is licensed under CC BY. This license enables reusers to distribute, remix, adapt, and build upon the material in any medium or format, and only so long as attribution is given to the creator.

Potential Uses

  • Train artificial intelligence/natural language processing algorithms for holistic scoring of student essays
  • Analyze the characteristics of strong and weak student writing
  • Perform subgroup analyses to understand population-level differences in student writing
© Copyright 2024. The Learning Agency LAB. All rights Reserved.