PAR Framework Releases Full Data Definitions

The Predictive Analytics Reporting (PAR) Framework project has publicly released the full definitions for the data included in the PAR database to encourage distribution of the definitions in the higher education research community.

The PAR Framework is a longitudinal, multi-institutional data-mining project managed by the WICHE Cooperative for Educational Technologies (WCET). According to information on the organization's site, "the PAR Framework makes it possible to create meaningful benchmark reports, conduct predictive analyses, and explore best practices in student retention on a massive collection of student records in a quest to better understand the variables affecting student loss."

The data includes more than 60 institutional, transcript, and student-level variables. "These variables will be used as building blocks in various strategic combinations to craft meaningful outcome measures and actionable predictors of student risk which further highlights the importance of common definitions," said Beth Davis, PAR Framework project director, in a prepared statement.

The PAR Framework has published the data definitions using the Data Cookbook, a collaborative data dictionary and data management tool built for higher education by IData. The definitions are published under a Creative Commons license with attribution, non-commerical, share-alike conditions.

Until now, the data fields and definitions used in the PAR Framework have been available only to the organization's 16 institutional partners, who have submitted more than 1.7 million anonymized and institutionally de-identified student records and 8.1 million course-level records to the dataset.

"The online Data Cookbook forum for publicly sharing our data model creates opportunities for alignment and collaboration with other projects and other datasets meaningful in higher education," said Mike Sharkey, director of academic analytics at the Apollo Group (parent company of the University of Phoenix), in a prepared statement.

The PAR Framework plans to continue to refine its data set to align with the recently released Common Education Data Standards (CEDS) version 3 and other higher education data sets.

Further information about the PAR Framework's data definitions can be found on the Data Cookbook site.

About the Author

Leila Meyer is a technology writer based in British Columbia. She can be reached at [email protected].

Featured

  • computer with a red warning icon on its screen, surrounded by digital grids, glowing neural network patterns, and a holographic brain

    Report Highlights Security Risks of Open Source AI

    In these days of rampant ransomware and other cybersecurity exploits, security is paramount to both proprietary and open source AI approaches — and here the open source movement might be susceptible to some inherent drawbacks, such as use of possibly insecure code from unknown sources.

  • pattern of interconnected glowing nodes and lines forming a neural network structure

    Meta AI Releases Open Source Machine Learning Library to Tackle Dataset Management Challenges

    Meta AI has announced LeanUniverse, an open source machine learning (ML) library designed to address the growing challenges of managing datasets in large-scale machine learning projects.

  • modern college building with circuit and brain motifs

    Anthropic Launches Claude for Education

    Anthropic has announced a version of its Claude AI assistant tailored for higher education institutions. Claude for Education "gives academic institutions secure, reliable AI access for their entire community," the company said, to enable colleges and universities to develop and implement AI-enabled approaches across teaching, learning, and administration.

  • glowing brain, connected circuits, and abstract representations of a book and graduation cap on a light gray gradient background

    Snowflake Launches Program to Upskill 100,000 People in Data and AI

    Cloud data platform Snowflake is embarking on an effort to train and certify more than 100,000 users on its AI Data Cloud by 2027. The One Million Minds + One Platform program will provide Snowflake-delivered courses, training materials, and free access to Snowflake software, at no cost to learners.