Dynamic reliability assessment of multiple choice tests in distributed e-leaning environments: A case study
Standard Error of Measurement
Full recordShow full item record
AbstractThe development of high-quality e-learning products is one of the most demanding areas in the field of educational research. Reliability of the students' grading mechanisms especially in the case of virtual classrooms, which lack in physical student-instructor interaction, is extremely important. In this paper, based on real data, we utilize two reliability estimation methods to calculate several multiple choice tests' reliability. Moreover, since multiple choice tests are an imperfect measure of students' knowledge, we also estimate the students' true ability of scoring using the tests' standard error of measurement. Concluding this study embeds reliability assessment methods in the e-learning process and then carefully analyzes the produced data to provide the strengths and weaknesses of the analyzed course's multiple choice tests. © 2007 International Federation for Information Processing.