Self-Assessment in a Feasible, Adaptive Web-Based Testing System

Abstract

Adaptive testing systems generate tests for assessment that are tailored to each student. In these tests, students are assessed through a process that uses Item Response Theory (IRT), a well-founded psychometric theory. This theory is responsible for estimating student knowledge, determining the next question that must be posed at each moment, and deciding test finalization. System of Intelligent Evaluation Using Tests for Teleeducation (SIETTE) is a Web-based environment for generating and constructing adaptive tests. In SIETTE, teachers can create tests for self-assessment. In this kind of test, questions are posed one by one, and the correction of each question is shown immediately after the student's answer. Along with this correction, and in terms of the student's answer, feedback is provided. Feedback consists of pieces of knowledge that help students detect misconceptions or reinforce concepts correctly learned. Furthermore, hints can be included when questions are posed to supply students with some kind of help or explanation about the stem. As a result, this kind of test can be used not just for assessment, but also for instructional purposes. The first goal of this paper is to show how SIETTE can be used for instructional purposes, by combining adaptive student self-assessment test questions with feedback and hints. This paper also shows that the Web is a feasible platform for the generation of adaptive tests, supporting the use of SIETTE for this purpose.