Partial Least Squares-SEM
1. Partial
Least Squares is just like Principal Component (PC) Regression except in how
the component scores are comput
2. PC
regression = weights are calculated from the covariance matrix of the
predictors- PLS =
weights reflect the covariance structure between predictors and response
3. While conceptually not too much
of a stretch, it requires a more complicated iterative algorithm
4. Like in regression, the goal is to maximize
the correlation between the response(s) and component scores
ภายใต้กฎที่ว่า
กล่าวโดยสรุปการประเมินผล PLS-SEM
Reflective Measurement Models
|
Formative Measurement Models
|
1.
Internal consistency (composite reliability) ร่วมกับ Cronbachs Alpha)
2.
Indicator reliability
3.
Convergent validity (AVE > 0.5)
4.
Discriminant validity (Fornell-Lacker Analysis)
|
1.
Convergent validity (R2 แต่ละตัว)
2. Collinearity among indicators (VIF)
3.
Significance and Relavance od outer weights
|
1.
Coefficients of determination (R2 = จำนวนตัวแปรตาม)
2.
Predictive relevance (Q2)
3.
Size and significance of path coefficients
4.
f2 effect sizes
5.
q2 effect sizes (if necessary)
|
Source: Hair, J. F. Jr., Hult, G. T. M., Ringle,
C.M.,Sarstedt, M. (2013) A Premier Partial Least Squares Structural Equation
Modeling (PLS-SEM). London: Sagepublications.com
ไม่มีความคิดเห็น:
แสดงความคิดเห็น