Pemodelan Survival Stratified Cox dan Extended Cox pada Lama Studi Doktoral ITS

Nurmalasari, Selly (2024) Pemodelan Survival Stratified Cox dan Extended Cox pada Lama Studi Doktoral ITS. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043211112-Undergraduate-Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (14MB) | Request a copy

Abstract

Peraturan Rektor ITS No. 18 Tahun 2023 menetapkan bahwa mahasiswa Program Doktor dianggap lulus tepat waktu jika menyelesaikan studinya dalam waktu 3 tahun. Namun, selama 4 periode wisuda terakhir di ITS, sekitar 73% mahasiswa Doktoral tercatat menyelesaikan studinya dalam waktu lebih dari 3 tahun. Hal ini sejalan dengan data dari PDDikti Perguruan Tinggi yang menunjukkan bahwa rata-rata masa studi Doktoral ITS adalah 4,7 tahun, yang berarti sebagian besar mahasiswa tidak lulus dengan tepat waktu. Permasalahan tersebut akan berdampak pada efisiensi akademik dan kualitas lulusan, sehingga diperlukan upaya untuk menduga faktor-faktor yang menjadi penyebab mahasiswa mengalami keterlambatan dalam menyelesaikan studinya. Mahasiswa Doktoral cenderung menghadapi tantangan yang lebih berat seperti bekerja, tuntutan akademik maupun tanggung jawab pribadi dan profesional lainnya. Hal tersebut kemungkinan menjadi salah satu penyebab atas keterlambatan kelulusan studi. Oleh karena itu, penelitian ini dilakukan menggunakan pendekatan analisis survival dengan metode Regresi Cox yang memerlukan asumsi Proportional Hazard (PH). Apabila asumsi tidak terpenuhi, maka digunakan pemodelan alternatif yaitu Stratified Cox dan Extended Cox. Sebagian besar mahasiswa doktoral (70,3%) tidak lulus tepat waktu, dengan rata-rata lama studi 10-11 semester. Kemudian model Stratified Cox dengan interaksi status pekerjaan (X3) memperoleh nilai AIC terendah dan jumlah variabel signifikan yang lebih banyak daripada model lainnya. Model ini dapat menangkap pola tambahan yang diinteraksikan dengan status pekerjaan, sehingga dukungan sosial, tekanan eksternal, dan kondisi kesehatan berperan terhadap mempercepat atau memperlambat durasi studi. Mahasiswa yang bekerja cenderung memiliki waktu studi yang lebih lama, karena terbagi antara kewajiban akademik dan pekerjaan. Penelitian ini diharapkan dapat memberikan wawasan untuk merancang kebijakan yang lebih efektif dalam meningkatkan kualitas dan efisiensi studi di Program Doktoral ITS.
===================================================================================================================================
ITS Chancellor's Regulation No. 18 of 2023 stipulates that Doctoral Program students are considered to graduate on time if they complete their studies within 3 years. However, during the last 4 graduation periods at ITS, around 73% of Doctoral students were recorded to complete their studies in more than 3 years. This is in line with data from PDDikti Higher Education which shows that the average ITS Doctoral study period is 4.7 years, which means that most students do not graduate on time. These problems will have an impact on academic efficiency and the quality of graduates, so efforts are needed to estimate the factors that cause students to experience delays in completing their studies. Doctoral students tend to face tougher challenges such as work, academic demands as well as other personal and professional responsibilities. This may be one of the causes of the delay in graduation of studies. Therefore, this study was conducted using a survival analysis approach with the Cox Regression method which requires the assumption of Proportional Hazard (PH). If the assumption is not met, alternative modeling, namely Stratified Cox and Extended Cox, is used. Most doctoral students (70.3%) do not graduate on time, with an average length of study of 10-11 semesters. Then the Stratified Cox model with job status interaction (X3) obtained the lowest AIC value and a greater number of significant variables than other models. This model can capture additional patterns that interact with employment status, so that social support, external pressure, and health conditions play a role in accelerating or slowing down the duration of the study. Working students tend to have a longer study time, because they are divided between academic and work obligations. This research is expected to provide insight to design more effective policies in improving the quality and efficiency of studies in the ITS Doctoral Program.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Doktoral ITS, Extended Cox, Masa Studi, Proportional Hazard, Stratified Cox. Doctoral ITS, Extended Cox, Length of Study, Proportional Hazard, Stratified Cox.
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis. L Education > L Education (General) Q Science
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Selly Nurmalasari
Date Deposited:	11 Feb 2025 04:06
Last Modified:	11 Feb 2025 04:06
URI:	http://repository.its.ac.id/id/eprint/118632

Actions (login required)

View Item