PEMILIHAN ATRIBUT EVALUASI KINERJA DOSEN YANG BERPENGARUH TERHADAP TINGKAT KEPUASAN MAHASISWA DALAM MENGIKUTI MATA KULIAH
The level of student satisfaction on a course is determined by many factors, one of them is lecturer performance. Evaluation of lecturer�s performance can be done by student as respondent. Course evaluation dataset from 5280 students of Gazi University was used in this research. From 28 attributes...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Theses and Dissertations NonPeerReviewed |
| Published: |
[Yogyakarta] : Universitas Gadjah Mada
2014
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/132892/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=73437 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The level of student satisfaction on a course is determined by many
factors, one of them is lecturer performance. Evaluation of lecturer�s performance
can be done by student as respondent. Course evaluation dataset from 5280
students of Gazi University was used in this research. From 28 attributes in the
dataset, there were 12 attributes related to the level of student satisfaction in
enrolling courses, while 16 other attributes indicated the students� evaluation of
lecturer performance. The aim of this research was to obtain subset containing
lecturer performance attributes which had significant effect on the level of student
satisfaction in enrolling courses.
Clustering process was done to get class attribute that indicated the level of
student satisfaction in enrolling courses. Attribute selection methods such as CFS,
Wrapper, Gain Ratio, and RELIEF were used to determine the input attributes
(students� evaluation of lecturer performance) which had significant effect on the
level of student satisfaction in enrolling courses. Classification process was used
as an evaluation method, while the accuracy value, ROC, TP rate, and FP rate
were used as evaluation parameter to determine the best subset.
The evaluation showed that the classification using the subset attribute was
better than the classification using all attributes. Subset from Wrapper Attribute
Selection was the best subset made in this research, generating 88.02% accuracy,
0.928 ROC, 0.880 TP rate, and 0.079 FP rate. There were 7 attributes that
significantly affect on the level of student satisfaction in enrolling courses, such
as: lecturer's knowledge was relevant and up to date, lecturer taught in accordance
with the announced lesson plan, lecturer was committed to the course and was
understandable, lecturer arrived on time for classes, lecturer was open and
respectful to student�s view about the course, lecturer gave relevant homework,
assignments/projects, and helped/guided students, and lecturer responded to
questions about the course inside and outside of the course |
|---|