Research Article
Automatic Programming Assessment to Measure Programming Problem-Solving Skills
@INPROCEEDINGS{10.4108/eai.20-10-2022.2328840, author={Rina Harimurti and Yeni Anistyasari and Puput Wanarti Rusimamto and Subuh Isnur Haryudo}, title={Automatic Programming Assessment to Measure Programming Problem-Solving Skills}, proceedings={Proceedings of the 4th Annual Conference of Engineering and Implementation on Vocational Education, ACEIVE 2022, 20 October 2022, Medan, North Sumatra, Indonesia}, publisher={EAI}, proceedings_a={ACEIVE}, year={2023}, month={5}, keywords={automatic scoring system problem solving skills computer programming}, doi={10.4108/eai.20-10-2022.2328840} }- Rina Harimurti
Yeni Anistyasari
Puput Wanarti Rusimamto
Subuh Isnur Haryudo
Year: 2023
Automatic Programming Assessment to Measure Programming Problem-Solving Skills
ACEIVE
EAI
DOI: 10.4108/eai.20-10-2022.2328840
Abstract
Problem-solving skills today form an integral part of the curriculum and meet student needs because they have become a problem in the dominant field of education. Computer programming is a subject that requires problem-solving strategies and involves a large amount of logical programming activity. Therefore, one alternative method of introducing problem-solving skills is through computer programming. With the development of programming software and industrial needs, which are supported by operations in the industry, progressing so fast, the existence of software that is able to automatically assess the results of student work and assignments is needed, especially in vocational schools. This research generally aims to implement an automatic assessment tool that can help teachers make corrections on student work related to computer programming. This type of research is quantitative and descriptive; the results are presented in the form of a research report with data analysis using SEM. The results showed that the automatic assessment system used was reliable, and the instrument for measuring the problem-solving skills of subjects had good construct validity, with the LF of each indicator worth 0.50.