Mengapa Harus Analisis SEM PLS

Sumber:
Jorg Henseler, Christian M. Ringle and Rudolf R. Sinkovics (2009),
The Use Of Partial Least Squares Path Modeling In International Marketing, New Challenges to International Marketing, Advances in International Marketing, Volume 20, 277–319
---------------------------------------------------------------------------------------
Many researchers argue that the goal of their studies is in line with particular strengths of PLS path modeling. The most important motivations are exploration and prediction, as PLS path modeling is recommended in an early stage of theoretical development in order to test and validate exploratory models. Another powerful feature of PLS path modeling is that it is suitable for prediction-oriented research. Thereby, the methodology assists researchers who focus on the explanation of endogenous constructs.
-------------------------------------------------------------------------------------
Motivasi penggunaan analisis PLS adalah studi eksplorasi dan prediksi. SEM PLS juga direkomendasikan sebagai analisis yang dapat digunakan untuk model pengembangan teori tahap awal sebagai analisis untuk pengujian atau model validasi eksplorasi. Analisis ini memfokuskan pada penjelasan varians variable laten endogen.

------------------------------------------------------------------------------------
The characteristics of PLS path modeling, which researchers regard as relevant for their studies on international marketing, can be summarized as follows:

1. PLS delivers latent variable scores, i.e. proxies of the constructs, which are measured by one or several indicators (manifest variables).
2. PLS path modeling avoids small sample size problems and can therefore be applied in some situations when other methods cannot.
3. PLS path modeling can estimate very complex models with many latent and manifest variables.
4. PLS path modeling has less stringent assumptions about the distribution of variables and error terms.
5. PLS can handle both reflective and formative measurement models.

-----------------------------------------------------------------------------------------------
Adapun karakterisrik model SEM PLS adalah:

1. SEM PLS menghasilkan skor variable laten. skor ini merupakan proksi skor variable laten yang dapat diukur oleh 1 atau beberapa indikator pengukur. 
2. SEM PLS dapat digunakan untuk ukuran sample penelitian yang terbatas/ kecil
3. SEM PLS dapat bekerja untuk model kompleks dengan bantak variabel laten dan  indikator
4. SEM PLS mempunyai  asumsi yang lebih longgar dari distribusi data dan distribusi variable error
5. SEM PLS dapat menangani model pengukuran reflektif dan formatif.

----------------------------------------------------------------------------------------------

Salam,
Sofyan Yamin
@ Desember 2017
di Bogor.