Hi all;
I use the Smart PLS program for my model analysis.
I have created a path model among 8 of the Usability criteria and assigned "Optimism" as the moderating effect from the Technology Readiness Index.
Whichever any independent variable that I assign the Optimism as the moderating effect, the path coefficient turns out to be negative (example of -0,079), even when the p value (example of 0.03) is significant (as you see on the snipping). Always negative when I change the path lines in the program.
(Path coefficient value = -0.079, t=2.173)
I couldn't find a solution. What could your suggestions be?
I have sought the effect of suppression, but currently no way out.
Thank you so much for your reponses.
Hello Mehmet,
What is this mysterious independent variable that you show in your sample model as influencing Satisfaction? How is it quantified?
With just the information given in your query, it's hard to know what the actual concern might be. The small and negative coefficient associated with the unnamed IV (in green in the display) might simply be due to: (a) collinearity with Optimism; (b) little real relationship with Satisfaction scores; (c) poor technical quality (reliability, validity) of the IV as quantified; or (d) some combination of factors.
Good luck with your work.
The effect size is too small. It doesn't matter whether it is significant or not. It is a very weak effect. You can also check effect size in the results of smartpls sheet. Therefore, in my opinion you need to re-consider your data validity or apply bootstrap in different variation.
When the effect is negative, it indicates that the higher the moderator level the affected relationships are reduced in magnitude.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling