Dear all,
Liang's method is widely used in PLS research. However, one recent paper in MISQ said it does not work. so my questions are:
(1) Shall we never use this method at all?
(2) If not use this method, how can we test the common method bias "AFTER" we have collected the data (so no marker variable available)??
Single factor analysis is too weak as well.
Thank you all, and looking forward to hearing from u!
LIANG, H.; SARAF, N.; HU, Q.; XUE, Y. Assimilation of enterprise systems: the effect of institutional pressures and the mediating role of top management. MIS Quarterly. v.31, n.1, p.59-87, mar/2007.
Common method bias, never use Liang et al.'s method????
-
jialin3105
- PLS Junior User
- Posts: 8
- Joined: Sun Mar 27, 2011 2:28 am
- Real name and title:
-
Hengkov
- PLS Super-Expert
- Posts: 1599
- Joined: Sun Apr 24, 2011 10:13 am
- Real name and title: Hengky Latan
- Location: AMQ, Indonesia
- Contact:
Hi,
Liang et al. (2007) method in MISQ not good for controlling CMV. Based on Chin et al. (2012a) simulation ignore this method. After that, Chin et al. (2012b) develop new method for dealing CMV with Measured Latent Merker Variable (MLMV). But, MLVM can only we used before collect data with choose item/indicators for MLMV. So, if you say "after" (you already done collect data) this method cannot used (may be it's disadvantages this approach). For alternative, I recommend used CFA Marker by Williams et al. (2010).
Regards,
References:
Chin, W.W., Thathcer, J.B., and Wright, R.T. 2012a. “Assessing Common Method Bias: Problem with the ULMC Technique," MIS Quarterly (36:3), pp. 1003-1019.
Chin, W.W., Thathcer, J.B., Wright, R.T., Steel, D.J. 2012b. “Controlling for Common Method Variance in PLS Analysis: The Measured Latent Marker Variable Approach,” Proceedings: 7th International Conference on Partial Least Squares and Related Methods, Houston, Texas USA, pp. 1-8.
Williams, L. J., Hartman, N., and Cavazotte, F. 2010. “Method Variance and Marker Variables: A Review and Comprehensive CFA Marker Technique,” Organizational Research Methods (13:3), pp. 477-514.
Liang et al. (2007) method in MISQ not good for controlling CMV. Based on Chin et al. (2012a) simulation ignore this method. After that, Chin et al. (2012b) develop new method for dealing CMV with Measured Latent Merker Variable (MLMV). But, MLVM can only we used before collect data with choose item/indicators for MLMV. So, if you say "after" (you already done collect data) this method cannot used (may be it's disadvantages this approach). For alternative, I recommend used CFA Marker by Williams et al. (2010).
Regards,
References:
Chin, W.W., Thathcer, J.B., and Wright, R.T. 2012a. “Assessing Common Method Bias: Problem with the ULMC Technique," MIS Quarterly (36:3), pp. 1003-1019.
Chin, W.W., Thathcer, J.B., Wright, R.T., Steel, D.J. 2012b. “Controlling for Common Method Variance in PLS Analysis: The Measured Latent Marker Variable Approach,” Proceedings: 7th International Conference on Partial Least Squares and Related Methods, Houston, Texas USA, pp. 1-8.
Williams, L. J., Hartman, N., and Cavazotte, F. 2010. “Method Variance and Marker Variables: A Review and Comprehensive CFA Marker Technique,” Organizational Research Methods (13:3), pp. 477-514.
Hi Hengkov,
I am thinking what if I take the average means of all formative indicators and make it as global single-item (marker or Measured Latent Merker Variable-MLMV)? Does it make sense?
How can I decide my global single-item measure as I have almost in the last stage of analysis.
BR, Murad
I am thinking what if I take the average means of all formative indicators and make it as global single-item (marker or Measured Latent Merker Variable-MLMV)? Does it make sense?
How can I decide my global single-item measure as I have almost in the last stage of analysis.
BR, Murad
MURAD ALI, Ph.D