3. Outline
•What is CMV?
•Is CMV really exist?
•How to prevent CMV happen?
•How to detect CMV?
–Harman single-factor test
–CFA Factor test
–ULMC (Unmeasured Latent Method Construct) Technique
•How to correct CMV?
–ULMC
–Correlational Marker Technique
–CFA Marker Technique
•Conclusion
4. What is CMV?
•Common method variance, may cause systematic measurement error and further bias the estimates of the true relationship among theoretical constructs.
•Method variance can either inflate or deflate observed relationships between constructs, thus leading to both Type I and Type II errors.
5. What is CMV?
•The variance of every measured variable can be partitioned into three components
–trait variance
–method variance (one of systematic error),
–error variance (random error of measurement, nonsystematic influences on measured variables)
6. What is CMV?
•Total variance= true variance + systematic variance (common method variance)+ random variance (measurement error)
•method variance is referred to as systematic bias
13. CMV?
INFLUENCE?
CONTROL
EXIST
EXIST
GAME OVER
NOT EXIST
NOT EXIST
1.Harman’s single factor
2.CFA計算CMV
3.CFA單因子及多因子比較
1.ULMC
2.CFA MarkerVariable
1.ULMC
2.CFA Marker as Control Variable
CMV 檢測流程
14. How to detect CMV?
•A priori 事前準備
–Marker variable
•Post Hoc 事後檢查
–Harman’s one factor
–CFA
–Unmeasured Latent Method Construct ( ULMC )
–Correlational Marker Technique
–CFA Marker Technique
15. Selection of Marker Variable
•事前準備A prior
–marker variable (Ideal marker)是一個在理 論上與潛在構面無關的變數,一般期望marker variable與模型其他的構面相關為0。
•事後偵側Post Hoc
–選擇與所有構面相關“最低”的構面來當 marker variable (Nonideal marker)。
32. CFA Marker Technique ApproachWilliams, Hartman, & Cavazotte, (2010).
•五大模型
1.Model 1:CFA Model
2.Model 2:Baseline Model
3.Model 3:Method C Model
4.Model 4:Method U Model
5.Model 5:Method R Model
33. Model 1
•CFA Model是將所有構 面含Marker Variable 設成完全有相關CFA