Investigation of the Use of Electronic Portfolios in the Determination of Student Achievement in Higher Education Using the Many-Facet Rasch Measurement Model

This study aimed to determine the rater behavior in the evaluation process of student electronic portfolios used to measure student achievement in higher education, and thus to evaluate the usability of the electronic portfolio system. Considering that rater behavior adversely affects both validity and reliability in determining the performance of individuals, it is important to identify the effect of this factor and evaluate the related results in line with this effect. The data of the study were collected from the students enrolled in an English language teaching program at Gazi University Gazi Education Faculty within the scope of the measurement and assessment course in the fall semester of 2017-2018. An analytic rubric developed by the researchers was used in the evaluation of the student electronic portfolios. The study included two participants groups consisting of three raters and 61 students (11 male, 50 female). In the analysis of the data, the many-facet Rasch measurement model was used as an analysis method since it was appropriate for the nature of the current data set. When the findings of the study were examined, it was found that one or more rater behaviors interfered with the performance of the individual in the use of non-objective measurement tools, and consequently negatively affected the validity and reliability of the measurements. In conclusion, it can be stated that the individual's performance related to electronic portfolios in higher education is generally affected by the rater behavior in the evaluation process independent of the measurement tool. In addition, it has been confirmed that electronic portfolios can be used to determine individual performance in higher education.