Log-file Analysis in International Large-Scale Assessments (LOGAN) (completed)

Table of numbers and graphs showing results from assessments

How do we extract and analyze log data from computer-based assessments? The LOGAN project has the answer. Photo: Luke Chesser/Unsplash

About the project

A new type of data is emerging from digital-based assessments through computer-generated log files. Gathering the traces of users’ interactions with the test software, these time-stamped files (a.k.a. “log files”) can provide a wealthy source of information beyond the traditional correct/incorrect answers to the test items. This project aims to develop statistical tools for the analysis of log-file data from international surveys such as the Programme of International Student Assessment (PISA) and the Programme for International Assessment of Adult Competencies (PIAAC). Through the development of an open-source R package called LOGAN, this project is an evolving portfolio of research targeting the extraction and analysis of meaningful features from these files.

Background

Despite the availability of public log-file data from computer-based tests, there is still a lack of software that adequately deals with this kind of data. The LOGAN package was created to fill this gap, looking into the particularities of international large-scale assessments data.