An Evaluation of a Placement Assessment for an Adaptive Learning System
Saved in:
| Title: | An Evaluation of a Placement Assessment for an Adaptive Learning System |
|---|---|
| Language: | English |
| Authors: | Jeffrey Matayoshi, Eric Cosyn, Christopher Lechuga, Hasan Uzun |
| Source: | International Educational Data Mining Society. 2024. |
| Availability: | International Educational Data Mining Society. e-mail: admin@educationaldatamining.org; Web site: https://educationaldatamining.org/conferences/ |
| Peer Reviewed: | Y |
| Page Count: | 8 |
| Publication Date: | 2024 |
| Document Type: | Speeches/Meeting Papers Reports - Evaluative |
| Education Level: | Elementary Secondary Education |
| Descriptors: | Student Placement, Evaluation Methods, Elementary Secondary Education, Accuracy, Computer Assisted Testing, Adaptive Testing, Placement Tests, Knowledge Level, Test Validity, Mathematics Education |
| Abstract: | ALEKS is an adaptive learning and assessment system, with courses covering subjects such as math, chemistry, and statistics. In this work, we focus on the ALEKS math courses, which cover a wide range of content starting at second grade math and continuing through college-level precalculus. To help instructors and students navigate this content, the system recently introduced an adaptive placement assessment for its K-12 users in the U.S. This assessment evaluates a student's mathematical knowledge and recommends the most appropriate ALEKS course for that student. In what follows, we present several evaluations of this placement assessment. After first analyzing the performance of the assessment with standard classifier metrics, such as AUROC, we next look in more detail at the accuracy of the knowledge states--that is, we look at the accuracy of the assessment when classifying problem types as being known or not known by students. For our last analysis, we look at student outcomes in their ALEKS courses after taking the placement assessment. We then finish with a discussion of these results and their implications for the assessment. [For the complete proceedings, see ED675485.] |
| Abstractor: | As Provided |
| Entry Date: | 2025 |
| Accession Number: | ED675621 |
| Database: | ERIC |
| Abstract: | ALEKS is an adaptive learning and assessment system, with courses covering subjects such as math, chemistry, and statistics. In this work, we focus on the ALEKS math courses, which cover a wide range of content starting at second grade math and continuing through college-level precalculus. To help instructors and students navigate this content, the system recently introduced an adaptive placement assessment for its K-12 users in the U.S. This assessment evaluates a student's mathematical knowledge and recommends the most appropriate ALEKS course for that student. In what follows, we present several evaluations of this placement assessment. After first analyzing the performance of the assessment with standard classifier metrics, such as AUROC, we next look in more detail at the accuracy of the knowledge states--that is, we look at the accuracy of the assessment when classifying problem types as being known or not known by students. For our last analysis, we look at student outcomes in their ALEKS courses after taking the placement assessment. We then finish with a discussion of these results and their implications for the assessment. [For the complete proceedings, see ED675485.] |
|---|
