Performance differences for in-class and online administration of low-stakes research-based assessments

TL;DR

This study compares online, low-stakes research-based assessments administered outside of class using the LASSO platform with traditional in-class paper tests, finding they produce equivalent data for evaluating student outcomes.

Contribution

It provides empirical evidence that online, low-stakes RBAs via LASSO are comparable to traditional paper tests, supporting broader adoption of online assessment methods.

Findings

LASSO platform yields equivalent data to paper tests

Online assessments are effective for low-stakes RBAs

Supports wider use of online assessments in education

Abstract

Research-based assessments (RBAs), such as the Force Concept Inventory, have played central roles in many course transformations from traditional lecture-based instruction to research-based teaching methods. In order to support instructors in assessing their courses, the online Learning About STEM Student Outcomes (LASSO) platform simplifies administering, scoring, and interpreting RBAs. Reducing the barriers to using RBAs will support more instructors in objectively assessing the efficacy of their courses and, subsequently, transforming their courses to improve student outcomes. The purpose of this study was to investigate the extent to which RBAs administered online and outside of class with the LASSO platform provided equivalent data to traditional paper and pencil tests administered in class. Research indicates that these two modes of administering assessments provide equivalent data for graded exams that are administered in class. However, little research has focused on ungraded (low-stakes) exams that are administered outside of class. We used an experimental design to investigate the differences between these two test modes. Results indicated that the LASSO platform provided equivalent data to paper and pencil tests.