PAR Framework (LAEP Inventory)

Cloud created by:

Rebecca Ferguson
11 February 2016

Synopsis

The Predictive Analytics Reporting (PAR) Framework is a non-profit provider of analytics-as-a-service to a range of types of higher education institution in the USA (two- and four-year courses of study; public and private; traditional and non‐traditional institutions)

 

It undertakes benchmarking, prediction and work to understand the signs of risk versus progress to completion. In addition to prediction, an aim of PAR is to support the identification of good practice in student retention through data analysis, shared models and benchmarking across institutions.

 

The PAR Framework motivations are two‐fold: a) that there is a cost saving in having a central analytics service with highly skilled staff, covering multiple aspects of expertise from data science to policy and higher education practice; b) cross‐institutional benchmark studies provide valuable information on effective strategies to promote achievement, engagement and progress, which a single-institution analytics activity would be unable to reveal.

Classification

Inventory type:

candidate for mainstreaming

Keywords:

predictive analytics, analytics as a service

Context of Practice

Learning:

post-compulsory

Geographical:

national: USA

Pedagogic:

The PAR Framework is not explicit in supporting any pedagogic practice over another.

Practical Matters

Tools used:

PAR does not disclose the tools it uses, but they are presumed to be generic data-analytic tools. Exchange of data and results between the member institutions and PAR uses files produced according to a set of Common Data Definitions, which are published under a Creative Commons licence.

Design and implementation:

PAR is a membership organisation in which each member institution contributes its data to a central database and receives the results of student-level analysis on its own data; PAR maintains a team including data scientists and researchers. Benchmark data is available to all member institutions. Governance is member‐led.

 

Each member institution is required to follow its normal institutional approval process for human subject research (ethics committee or institutional review board) and the PAR team all have certification in human subject research.

Maturity and Evidence of Utility

PAR is essentially already mainstream in that numerous higher education institutions in the United States are member institutions, but we have classified it as a candidate for mainstreaming as the model has yet to be replicated in other geographical regions.

 

PAR is now an independent non-profit organisation but it has evolved over a number of years, having been a service managed by the WICHE Cooperative for Educational Technologies (WCET) non‐profit organisation until late 2014, with funding from the Bill and Melinda Gates Foundation, at which point 16 institutions were part of the collaborative venture. Previously, PAR had been a smaller-scale pilot project. As of November 2015, PAR declared that more than 33 campuses participate in the collaborative.

 

There is an absence of independent academic research on the effectiveness of PAR.

Further Information

Overview of the PAR Framework: http://www.parframework.org/about-par/overview/

 

Extra content

Embedded Content

Contribute

Contribute to the discussion

Please log in to post a comment. Register here if you haven't signed up yet.