{"id":804,"date":"2024-04-10T15:15:59","date_gmt":"2024-04-10T08:15:59","guid":{"rendered":"https:\/\/cattleyapublicationservices.com\/?p=804"},"modified":"2024-04-10T21:26:40","modified_gmt":"2024-04-10T14:26:40","slug":"perbedaan-sem-pls-dengan-sem-konvensional","status":"publish","type":"post","link":"https:\/\/cattleyapublicationservices.com\/?p=804","title":{"rendered":"Perbedaan SEM PLS dengan SEM Konvensional"},"content":{"rendered":"\n<p>SEM (Structural Equation Modeling) adalah teknik analisis data yang digunakan untuk menguji hubungan antar variabel dalam suatu model teoritis. SEM konvensional (CB-SEM) dan SEM Partial Least Squares (PLS) adalah dua varian SEM yang populer dengan beberapa perbedaan fundamental.<\/p>\n\n\n\n<p><strong>Perbedaan Utama:<\/strong><\/p>\n\n\n\n<p><strong>1. Asumsi Distribusi Data:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Membutuhkan data yang terdistribusi normal dan multivariat normal. Asumsi ini penting untuk memastikan validitas statistik dari estimasi parameter dan uji hipotesis.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Lebih fleksibel dalam hal asumsi distribusi data. PLS-SEM dapat digunakan dengan data non-normal dan bahkan data kategorikal.<\/li>\n<\/ul>\n\n\n\n<p><strong>2. Ukuran Sampel:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Membutuhkan ukuran sampel yang relatif besar (minimum 100-150) untuk mencapai kekuatan statistik yang memadai.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Berkinerja lebih baik dengan ukuran sampel yang lebih kecil (minimum 30-50) dibandingkan CB-SEM.<\/li>\n<\/ul>\n\n\n\n<p><strong>3. Estimasi Parameter:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Estimasi parameter didasarkan pada metode Maximum Likelihood (ML) yang meminimalkan fungsi likelihood.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Estimasi parameter menggunakan metode PLS yang memaksimalkan kovariansi antara variabel laten dan indikatornya.<\/li>\n<\/ul>\n\n\n\n<p><strong>4. Uji Hipotesis:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Uji hipotesis didasarkan pada statistik chi-square, t-test, dan Wald test.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Uji hipotesis didasarkan pada bootstrapping dan resampling untuk menghasilkan nilai p dan confidence interval.<\/li>\n<\/ul>\n\n\n\n<p><strong>5. Kemampuan Multikolusi:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Sensitif terhadap multikolusi antar variabel, yang dapat menyebabkan bias dan hasil yang tidak akurat.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Lebih toleran terhadap multikolusi dibandingkan CB-SEM, sehingga dapat digunakan dengan model yang memiliki variabel berkorelasi tinggi.<\/li>\n<\/ul>\n\n\n\n<p><strong>6. Fokus Analisis:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Lebih fokus pada pengujian hipotesis dan konfirmasi model teoritis.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Lebih fokus pada prediksi dan eksplorasi hubungan antar variabel.<\/li>\n<\/ul>\n\n\n\n<p><strong>7. Interpretasi Hasil:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Interpretasi hasil didasarkan pada nilai p, coefficient of determination (R^2), dan standar error.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Interpretasi hasil didasarkan pada path coefficient, loadings, dan variance explained.<\/li>\n<\/ul>\n\n\n\n<p><strong>8. Penggunaan Software:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CB-SEM:<\/strong>&nbsp;Software populer: AMOS, Mplus, LISREL, EQS.<\/li>\n\n\n\n<li><strong>PLS-SEM:<\/strong>&nbsp;Software populer: SmartPLS, WarpPLS, PLS-Graph, R packages (pls, semPLS).<\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-table\"><table><tbody><tr><td><strong>Aspek<\/strong><\/td><td><strong>CB-SEM<\/strong><\/td><td><strong>PLS-SEM<\/strong><\/td><\/tr><tr><td>Distribusi Data<\/td><td>Normal, multivariat normal<\/td><td>Fleksibel (non-normal, kategorikal)<\/td><\/tr><tr><td>Ukuran Sampel<\/td><td>Besar (minimum 100-150)<\/td><td>Kecil (minimum 30-50)<\/td><\/tr><tr><td>Estimasi Parameter<\/td><td>Maximum Likelihood (ML)<\/td><td>Partial Least Squares (PLS)<\/td><\/tr><tr><td>Uji Hipotesis<\/td><td>Chi-square, t-test, Wald test<\/td><td>Bootstrapping, resampling<\/td><\/tr><tr><td>Multikolusi<\/td><td>Sensitif<\/td><td>Toleran<\/td><\/tr><tr><td>Fokus Analisis<\/td><td>Pengujian hipotesis, konfirmasi model<\/td><td>Prediksi, eksplorasi hubungan<\/td><\/tr><tr><td>Interpretasi Hasil<\/td><td>Nilai p, R^2, standar error<\/td><td>Path coefficient, loadings, variance explained<\/td><\/tr><tr><td>Software<\/td><td>AMOS, Mplus, LISREL, EQS<\/td><td>SmartPLS, WarpPLS, PLS-Graph, R packages (pls, semPLS)<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p><strong>Referensi:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Hair,&nbsp;J. F., Hult, G. T. M., Ringle, C. M., &amp; Sarstedt, M. (2017). A primer on partial least squares structural equation modeling (PLS-SEM). Sage publications.<\/li>\n\n\n\n<li>Chin, W. W. (1998). The partial least squares approach to&nbsp;structural equation modeling. In G. A. Marcoulides (Ed.), Modern methods for business research&nbsp;(pp. 295-336). Lawrence Erlbaum Associates.<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>SEM (Structural Equation Modeling) adalah teknik analisis data yang digunakan untuk menguji hubungan antar variabel dalam suatu model<\/p>\n","protected":false},"author":1,"featured_media":827,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[25],"tags":[],"class_list":["post-804","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-analisis-data-sem-pls"],"aioseo_notices":[],"jetpack_featured_media_url":"https:\/\/cattleyapublicationservices.com\/wp-content\/uploads\/2024\/04\/2fee449d-8a24-41f7-9992-aaacef9ad9e4.jpg","jetpack_sharing_enabled":true,"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=\/wp\/v2\/posts\/804","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=804"}],"version-history":[{"count":2,"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=\/wp\/v2\/posts\/804\/revisions"}],"predecessor-version":[{"id":828,"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=\/wp\/v2\/posts\/804\/revisions\/828"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=\/wp\/v2\/media\/827"}],"wp:attachment":[{"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=804"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=804"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cattleyapublicationservices.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=804"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}