Rodliyah, Millatur (2016) Estimasi Score Factor Dengan Partial Least Square (PLS) Pada Measurement Model (Studi Kasus : Remunerasi Tenaga Kependidikan di Lingkungan ITS). Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1314201039-Master-Thesis.pdf - Published Version Download (2MB) | Preview |
Abstract
Birokrasi di Indonesia merupakan sistem administrasi negara yang bertujuan untuk
melayani kepentingan rakyat. Kondisi birokrasi di Indonesia sampai saat ini masih memiliki
banyak kekurangan. Salah satu bentuk reformasi birokrasi yang dilakukan pemerintah
Indonesia adalah dengan pemberian remunerasi bagi Pegawai Negeri Sipil (PNS).
Remunerasi merupakan bagian kesejahteraan yang diterima oleh pegawai yang dapat
dijadikan sebagai unsur motivasi bagi pegawai untuk berprestasi dan kinerja menjadi lebih
baik. Variabel-variabel yang saling berhubungan dan digunakan pada penelitian ini adalah
variabel kinerja, remunerasi, motivasi berprestasi, karakteristik lingkungan kerja, dan transfer
pelatihan. Dengan kerangka pemikiran variabel laten motivasi berprestasi, karakteristik
lingkungan kerja, dan transfer pelatihan, berpengaruh terhadap kinerja, dan variabel laten
kinerja mempengaruhi variabel laten remunerasi. Variabel-variabel tersebut merupakan
construct atau variabel laten yang tidak dapat diukur secara langsung, sehingga metode SEM
dirasa mampu menyelesaikan permasalahan tersebut. Namun SEM memiliki beberapa asumsi
yang harus dipenuhi dan seringkali jika menggunakan data real di lapangan asumsi-asumsi
itu terlanggar, sehingga diperlukan suatu metode yang bebas asumsi, bebas distribusi (free
distribution) dan fleksibel yaitu metode SEM alternatif yang berbasis varians atau sering
disebut dengan PLS. PLS merupakan suatu metode estimasi yang berfokus untuk
memaksimumkan varians diantara variabel laten dan menjadi alternatif untuk OLS regresi.
Penelitian ini dilakukan untuk mengestimasi score factor pada measurement model dan
pemodelan dalam kasus remunerasi tenaga kependidikan di lingkungan ITS menggunakan
pendekatan Partial Least Square (PLS) menggunakan skema jalur (path scheme), skema
sentroid (centroid scheme), dan skema faktor (factor scheme). Hasil penelitian diperoleh
bahwa pemodelan remunerasi tenaga kependidikan di lingkungan ITS dengan skema faktor
PLS adalah yang terbaik dengan nilai Q-square sebesar 0,7262, R-square kinerja sebesar
67,69 persen dan R-square remunerasi sebesar 15,28 persen
==================================================================================================================
Bureaucracy in Indonesia is a state administrative system that aims to serve the interests of
the people. Conditions of bureaucracy in Indonesia still has many shortcomings until now.
One form of bureaucratic reforms made by the government of Indonesia is the remuneration
for civil servants (PNS). Remuneration is part of welfare received by employees which can
be used as an element of motivation for employees to excel and better performance. The
variables are interrelated and used in this study are performance, remuneration, achievement
motivation, characteristics of the work environment, and training transfer. With latent
variable framework achievement motivation, characteristics of the work environment, and
transfer training, affect the performance, and the latent variables affect performance latent
variable remuneration. These variables are constructs or latent variables that can not be
measured directly, so the SEM method deemed able to resolve these problems. However
SEM has some assumptions that must be met and often when using real data in the field
assumptions were violated, so we need a method that is free of assumptions, free distribution
and flexible that is SEM methods of alternative-based variance is often called the PLS , PLS
is an estimation method that focuses on maximizing the variance between latent variables and
an alternative to OLS regression. This study was conducted to estimate the factor scores in
the measurement model of remuneration and modelling in the case of educational staff in the
ITS with Partial Least Square (PLS) approach using path, centroid, and factor scheme. The
result is remuneration of educational staff in the ITS modelling with factor schemes is the
best method, outcome the Q-square value is 0,7262, R-square value of performance is 67,69
percent and R-square value of remuneration is 15,28 percent.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTSt 519.535 Rod e |
| Uncontrolled Keywords: | Estimasi score factor, partial least square (PLS), kinerja, dan remunerasi |
| Subjects: | H Social Sciences > HA Statistics |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis |
| Depositing User: | Users 13 not found. |
| Date Deposited: | 14 Jun 2017 07:11 |
| Last Modified: | 27 Dec 2018 06:51 |
| URI: | http://repository.its.ac.id/id/eprint/41666 |
Actions (login required)
 |
View Item |