How Assessment in Final Year Project Can Threatening the Validity?

Abstract

An undergraduate project evaluation is measured using a rubric assessment. The tendency for the examiner to be biased is high. Otherwise, these subjective evaluations will threaten the validity and reliability of the assessment. Therefore, this study was conducted to investigate the effect of halo in the evaluation of undergraduate projects using analysis software program Multi Facet Rasch Measurement (MFRM) version 3.82.2. The methodology of this study is by using a design approach quantitative form. A total number of 98 students and 43 lecturers who serve as examiners of the Faculty of Electrical and Electronic Engineering of Tun Hussein Onn Malaysia University were involved in this study using purposive sampling. Halo effects were analyzed based on the objective of the study and detected using two methods, as a group and individually. In the analysis group there were four indicators examined (i) A Fixed Chi- Squared Test (Fixed) (ii) The trait separation ratio (iii) The trait separation index and (iv) The Reliability of the trait separation index. Meanwhile, in the analysis of individual level was examined by identifying significant bias interactions. From the study, it was found that studies show that there are halo effects at the individual level while halo effects cannot be detected at the group level. The findings of the study are to create that measurement is an important element in interpreting student performance. Although halo effects can be identified individually among examiners. But it does not affect the assessment because it does not exist in the group. Training and coordination need to be done to further develop the same level of understanding to improve student's ability to be translated to actual ability. Quality assessment will produce excellent and valuable human capital in the future.