Hi,
I'm using a single item construct (dependent variable) in my model. Does it make any difference if I model it as reflective or as formative?
I have an impression that there is only one item mathematically modeling it as reflective or as formative would be equivalent. Is this really the case, or is there a difference?
Single item construct (formative vs. reflective)
- Hengkov
- PLS Super-Expert
- Posts: 1599
- Joined: Sun Apr 24, 2011 10:13 am
- Real name and title: Hengky Latan
- Location: AMQ, Indonesia
- Contact:
Hengky Latan, SE Pattimura University
Hi Haslinda,
Please see article
Jarvis, C.B., Mackenzie, S.B.,and Podsakoff, P.M. 2003. “A critical review of construct indicators and measurement model misspecification in marketing and consumer research,” Journal of Consumer Research, (30:2), pp. 199–218.
for difference reflective or formative model.
But if you used single item construct (dependent variable) in your model = reflective model
Regards,
Hengky
Please see article
Jarvis, C.B., Mackenzie, S.B.,and Podsakoff, P.M. 2003. “A critical review of construct indicators and measurement model misspecification in marketing and consumer research,” Journal of Consumer Research, (30:2), pp. 199–218.
for difference reflective or formative model.
But if you used single item construct (dependent variable) in your model = reflective model
Regards,
Hengky
- Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi,
PLS-PM is “partial” in the sense that the measurement model is computed in one step and the structural model in another step of the iteration procedure.
Then, with just one indicator, it doesn’t matter if it is formative or reflective:
Formative --> measurement model computed as a simple regression
Reflective --> measurement model computed as a correlation
And the correlation = standardized regression coefficient = 1.0 --> LV = indicator
Best regards,
Bido
PLS-PM is “partial” in the sense that the measurement model is computed in one step and the structural model in another step of the iteration procedure.
Then, with just one indicator, it doesn’t matter if it is formative or reflective:
Formative --> measurement model computed as a simple regression
Reflective --> measurement model computed as a correlation
And the correlation = standardized regression coefficient = 1.0 --> LV = indicator
Best regards,
Bido