I have been using an adapted UTAUT model in a secondary school setting and have attempted to investigate the moderators of gender, experience etc. I have also tested the possible moderating affects of some policy documents a school may have (technology plan, curriculum guide that includes technology curriculum, and an annual teacher evaluation that may include a technology integration evaluation).
I have analyzed this using SmartPLS and because it does not give significance levels, I calculated T scores to arrive at the significance levels. My problem is in my case only one of all the moderating effect tests was significant. Here is an example of a typical report of one of the moderator section.
The presence of a curriculum guide that included technology use the average variance extracted decrease for social influence (AVE=.66; β= -.029; α=.87, ρ=.59); attitude (AVE=.11; β= -.001; α=.83, ρ=.98); and anxiety (AVE=.64; β= .011; α=.82, ρ=.85); performance expectancy (AVE=.49; β=.057; α=.81, ρ=.45); and effort expectancy (AVE=.52; β= -.037; α=.33, ρ=.67); with the inclusion of a curriculum guide as a moderator. The presence of curriculum guide that includes technology use had a statistically significant influence on the relationship between social influences and behavioral intention (ρ=.034). While the relationship is statistically significant the relationship accounts for only 2.9% of the variance.
Basically, do you think I followed the correct procedure (T tests for significance) and that it is reasonable that on one (of 30 moderators) was statistically significant. If they are not statistically significant does the data have any value at all? Is there a better way to determine significance us SmartPLS?
I would appreciate any guidance you could offer.
Regards,
Significance levels of moderators
-
- PLS Junior User
- Posts: 4
- Joined: Wed Nov 10, 2010 10:46 am
- Real name and title:
Hi
I am doing multi-group research whereby I am examining how different/similar two groups of teams are. I have found the following article to be very useful in explaining a procedure to follow. I found this article from previous discussion forums concerning moderation. It might help you too. The article is as follows:
EBERL, M. (2010) An application of PLS in Multi-Group Analysis: The need for differentiated Corporate-Level Marketing in the Mobile Communications industry. IN VINZI, V., ESPOSITO; CHIN, W., W.; HENSELER, J. & WANG, H. (Eds.) Handbook of Partial Least Squares: Concepts, Methods and Applications. Springer Heidelberg Dordrecht, London.
In addition to the article above, page 25 of the following article might add some more to your investigations, Sosik et. el., specifically cover group moderation:
SOSIK, J., J.; KAHAI, S., S. & PIOVOSO, M., J. (2009) Silver bullet or Voodoo statistics?: A primer for using the Partial Least Squares data analytic technique in group and organizational research. Group & Organization Management, 34, 5 - 36.
Good luck
Howard
I am doing multi-group research whereby I am examining how different/similar two groups of teams are. I have found the following article to be very useful in explaining a procedure to follow. I found this article from previous discussion forums concerning moderation. It might help you too. The article is as follows:
EBERL, M. (2010) An application of PLS in Multi-Group Analysis: The need for differentiated Corporate-Level Marketing in the Mobile Communications industry. IN VINZI, V., ESPOSITO; CHIN, W., W.; HENSELER, J. & WANG, H. (Eds.) Handbook of Partial Least Squares: Concepts, Methods and Applications. Springer Heidelberg Dordrecht, London.
In addition to the article above, page 25 of the following article might add some more to your investigations, Sosik et. el., specifically cover group moderation:
SOSIK, J., J.; KAHAI, S., S. & PIOVOSO, M., J. (2009) Silver bullet or Voodoo statistics?: A primer for using the Partial Least Squares data analytic technique in group and organizational research. Group & Organization Management, 34, 5 - 36.
Good luck
Howard