Hello, I run my data using Smart PLS. I have dimensions for my construct.. may I avoid using second order?
Second question.. For one construct, I have 3 dimensions... for each dimension I have few item.. may I average the score item (likert scale score) at each dimension become one value.. Now I have 3 value for 3 dimensions... This 3 value as if item for my construct. My question is.. my I use this way to avoid second order..
Thank you
Hello
Could you provide a screen capture of your model to help you better?
Regards
Hello Mr. Alejandro.. kindly please see my model from attachment.. tqvm
Hello
On one side, I think you haven't loaded correctly your input data in SmarPLS. A1 to A8 and B1 to B8 are your "indicators" or items, that is to say, the answers of your respondents. These are the columns you have to import in SmartPLS, not the average score (SmartPLS automatically computes the latter, for example, to give you the mean, std. dev., etc.
On the other side, I do not understand when you refer to second order construct. In your model you have two IV (first order constructs) and a last third construct which is the DV. I guess that you want to know if it is possible to join both IVs in only one. However, this dosn't mean that it is a second order construct. You have to analyze how other researchers have built the dimensions in your model.
My recommendation is that you include A1 to A4 (HD1), A5 to A8 (HD2), B1 to B4 (AS1) and B5 to B8 (AS2), and the two "indicators" (C1 and C2) for the DV.
Regards
Hello Mr. Alejandro.. may I have your email? Maybe I can contact/ email you personally. tq
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running