<xml> </xml><![endif]--><!--[if gte mso 9]><xml> Normal 0 false false false EN-ID X-NONE X-NONE </xml><![endif]--><!--[if gte mso 9]><xml> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin-top:0cm; mso-para-margin-right:0cm; mso-para-margin-bottom:8.0pt; mso-para-margin-left:0cm; line-height:107%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-fareast-language:EN-US;} </style> <![endif]--> Consistent Partial Least Squares (PLSc) is an improvement upon PLS and an alternative to Variance-Based Covariance Structural Equation Modeling (CB-SEM). PLSc can overcome the weaknesses of PLS, which often results in inconsistent estimates, as well as address the limitations of CB-SEM. PLSc can predict models with both reflective and formative indicator designs, small sample sizes (30- 100), assumption-free and distribution-free characteristics, thus not requiring multivariate normal distribution of data. Therefore, statistical significance testing is addressed using bootstrap methods. Multigroup SEM with the PLSc approach can be used to test whether the relationships between variables in the model are significant and whether there are differences in the relationships between variables in different population subgroups. A case study on factors influencing human life quality with 187 samples divided into two groups based on island differences was analyzed using the PLSc approach for parameter estimation and further Multigroup analysis to test for differences in variable relationships between different groups.In this study, data were obtained from the Indonesian Central Bureau of Statistics (BPS) website and analyzed using smartPLS software. The variables used in the case study consist of exogenous latent variables, namely successful of economic(?), intermediary latent variables such as poverty (?1), and endogenous latent variables, namely human life quality (?2). The results of the case study indicate that after model modification, all indicators are significant for their respective latent variables. The structural model shows that in each region, economic success has a negative and significant effect on poverty, economy success has a positive and significant effect on human life quality, and poverty has a negative and significant effect on human life quality. The saturated model fit values indicate that the model is able to explain the data variation well. Meanwhile, multigroup analysis shows that the differences in variable relationships between the Java and Sumatra island are not significant "> <xml> </xml><![endif]--><!--[if gte mso 9]><xml> Normal 0 false false false EN-ID X-NONE X-NONE </xml><![endif]--><!--[if gte mso 9]><xml> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin-top:0cm; mso-para-margin-right:0cm; mso-para-margin-bottom:8.0pt; mso-para-margin-left:0cm; line-height:107%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-fareast-language:EN-US;} </style> <![endif]--> Consistent Partial Least Squares (PLSc) is an improvement upon PLS and an alternative to Variance-Based Covariance Structural Equation Modeling (CB-SEM). PLSc can overcome the weaknesses of PLS, which often results in inconsistent estimates, as well as address the limitations of CB-SEM. PLSc can predict models with both reflective and formative indicator designs, small sample sizes (30- 100), assumption-free and distribution-free characteristics, thus not requiring multivariate normal distribution of data. Therefore, statistical significance testing is addressed using bootstrap methods. Multigroup SEM with the PLSc approach can be used to test whether the relationships between variables in the model are significant and whether there are differences in the relationships between variables in different population subgroups. A case study on factors influencing human life quality with 187 samples divided into two groups based on island differences was analyzed using the PLSc approach for parameter estimation and further Multigroup analysis to test for differences in variable relationships between different groups.In this study, data were obtained from the Indonesian Central Bureau of Statistics (BPS) website and analyzed using smartPLS software. The variables used in the case study consist of exogenous latent variables, namely successful of economic(?), intermediary latent variables such as poverty (?1), and endogenous latent variables, namely human life quality (?2). The results of the case study indicate that after model modification, all indicators are significant for their respective latent variables. The structural model shows that in each region, economic success has a negative and significant effect on poverty, economy success has a positive and significant effect on human life quality, and poverty has a negative and significant effect on human life quality. The saturated model fit values indicate that the model is able to explain the data variation well. Meanwhile, multigroup analysis shows that the differences in variable relationships between the Java and Sumatra island are not significant ">
Laporkan Masalah

Multigroup Structural Equation Modelling dengan Pendekatan Consistent Partial Least Square

Ni Kadek Sudarti, Dr. Abdurakhman, M.Si

2024 | Tesis | S2 Matematika

<!--[if gte mso 9]><xml> </xml><![endif]--><!--[if gte mso 9]><xml> Normal 0 false false false EN-ID X-NONE X-NONE </xml><![endif]--><!--[if gte mso 9]><xml> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin-top:0cm; mso-para-margin-right:0cm; mso-para-margin-bottom:8.0pt; mso-para-margin-left:0cm; line-height:107%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-fareast-language:EN-US;} </style> <![endif]-->

Consistent Partial Least Square (PLSC) merupakan penyempurnaan PLS dan alternatif dari Covariance Based Structural Equation Modeling (CB-SEM) yang berbasis varians. PLSc dapat mengatasi kelemahan dari PLS yang sering menghasilkan estimasi yang tidak konsisten serta mengatasi kelemahan CB-SEM yaitu PLSc dapat memprediksi model dengan desain indikator yang bersifat reflektif maupun formatif, ukuran sampel kecil (30 ? 100), bebas asumsi, dan bebas distribusi sehingga datanya tidak harus berdistribusi normal multivariat. Maka dari itu, uji signifikansi statistik diatasi menggunakan metode bootstrap. Multigroup SEM dengan pendekatan PLSc dapat digunakan untuk memprediksi apakah hubungan antara variabel pada model signifikan serta apakah terdapat perbedaan hubungan antar variabel pada subkelompok yang berbeda. Studi kasus faktor-faktor yang mempengaruhi kualitas hidup manusia dengan 187 sampel yang terbagi menjadi dua kelompok berdasarkan perbedaan pulau dianalis menggunakan pendekatan PLSc untuk estimasi parameter dan selanjutnya analisis Multigroup untuk menguji apakah terdapat perbedaan hubungan variabel antar grup yang berbeda. Dalam penelitian ini, data diperoleh dari website BPS Indonesia dan di analisis dengan bantuan aplikasi smartPLS. Variabel yang digunakan terdiri dari laten eksogen yaitu keberhasilan ekonomi (?), laten perantara yaitu kemiskinan (?1) dan laten endogen yaitu kualitas hidup manusia (?2). Hasil studi kasus menunjukkan setelah modifikasi model, semua indikator signifikan terhadap masing-masing variabel latennya. Model struktural menunjukkan, pada setiap wilayah, keberhasilan ekonomi berpengaruh negatif dan signifikan terhadap kemiskinan,keberhasilan ekonomi berpengaruh positif dan signifikan terhadap kualitas hidup manusia serta kemiskinan berpengaruh negatif dan signifikan terhadap kualitas hidup manusia. Nilai saturated model fit menunjukkan bahwa model mampu menjelaskan variasi data dengan baik. Sementara itu, analisis multigroup menunjukkan perbedaan hubungan antar variabel pada pulau Jawa dan Sumatra tidak signifikan.

<!--[if gte mso 9]><xml> </xml><![endif]--><!--[if gte mso 9]><xml> Normal 0 false false false EN-ID X-NONE X-NONE </xml><![endif]--><!--[if gte mso 9]><xml> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin-top:0cm; mso-para-margin-right:0cm; mso-para-margin-bottom:8.0pt; mso-para-margin-left:0cm; line-height:107%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-fareast-language:EN-US;} </style> <![endif]-->

Consistent Partial Least Squares (PLSc) is an improvement upon PLS and an alternative to Variance-Based Covariance Structural Equation Modeling (CB-SEM). PLSc can overcome the weaknesses of PLS, which often results in inconsistent estimates, as well as address the limitations of CB-SEM. PLSc can predict models with both reflective and formative indicator designs, small sample sizes (30- 100), assumption-free and distribution-free characteristics, thus not requiring multivariate normal distribution of data. Therefore, statistical significance testing is addressed using bootstrap methods. Multigroup SEM with the PLSc approach can be used to test whether the relationships between variables in the model are significant and whether there are differences in the relationships between variables in different population subgroups. A case study on factors influencing human life quality with 187 samples divided into two groups based on island differences was analyzed using the PLSc approach for parameter estimation and further Multigroup analysis to test for differences in variable relationships between different groups.In this study, data were obtained from the Indonesian Central Bureau of Statistics (BPS) website and analyzed using smartPLS software. The variables used in the case study consist of exogenous latent variables, namely successful of economic(?), intermediary latent variables such as poverty (?1), and endogenous latent variables, namely human life quality (?2). The results of the case study indicate that after model modification, all indicators are significant for their respective latent variables. The structural model shows that in each region, economic success has a negative and significant effect on poverty, economy success has a positive and significant effect on human life quality, and poverty has a negative and significant effect on human life quality. The saturated model fit values indicate that the model is able to explain the data variation well. Meanwhile, multigroup analysis shows that the differences in variable relationships between the Java and Sumatra island are not significant


Kata Kunci : Multigroup, SEM, PLSc, smartPLS

  1. S2-2024-502284-abstract.pdf  
  2. S2-2024-502284-bibliography.pdf  
  3. S2-2024-502284-tableofcontent.pdf  
  4. S2-2024-502284-title.pdf