Journal of Engineering Education Transformations

Abstract

Abstract—In the present pandemic condition, challenge as a teacher, is not only to learn new tools for online teaching but also conducting online assessment and to assure that the assessment curves are not flat but shall be a natural response which has bell shaped distribution. We all know the impact of flat evaluation is that the industry will lose trust on the quality of graduate and their knowledge based on marks/grades. This will put the quality of technical education under doubt when high scorers perform poorly in the recruitment process (including job interviews) where students are not even able to pass the basic aptitude clearance test. Online education may not lack the quality and integrity of faceto- face instruction, but it is easy to cheat in online assessment. So, there is a need to develop some innovative and best practices which will help the sincere students to have academic edge over the insincere students and the students' marks or grades shall be the replica of their intellectual ability. It will enable the deserving students to get their due credits and not bring all at the same level. Current problem of flat evaluation which has become more significant in the current pandemic can be overcome. This is mainly because of online examinations in the home environment. This paper suggests and tests some techniques to remove flatness in evaluation and make it more authentic such as, use of uniqueness and uncertainty in question paper and assessment, conducting multiple activities as part of evaluation, giving weightage to various activities as per the efforts, making assessment a multilevel activity etc. The results show the evaluation curves which were flat when no innovative assessment technique was used for assessment became bell shaped when innovative assessment technique was used. Moreover, it was found to be in alignment with testing based on classroom testing under a proctored environment.

Keywords—Innovation, Evaluation, Assessment, Normal distribution