Page 1 of 1

Comparing the strength of relationships

Posted: Wed Apr 10, 2019 9:45 am
by ngn999
Hello everyone,

I have a model in which 2 IVs (A and B) ---> DV (C), running SmartPLS. A, B, and C are reflective constructs measured by 7-point Likert items. It is hypothesised that the positive effect of A->C is stronger than the positive effect of B->C. Is it enough if I use path coefficients and effect sizes to compare the effects of A and B on C and then support this hypothesis or I need other analyses?

Thanks in advance.

Re: Comparing the strength of relationships

Posted: Sat Apr 13, 2019 12:40 pm
by jmbecker
Of course a descriptive analysis of the path coefficients provides a first indication, but it would be better to really test if they are different.

Rodríguez-Entrena, M., Schuberth, F., & Gelhard, C. (2018). Assessing statistical differences between parameters estimates in Partial Least Squares path modeling. Quality & quantity, 52(1), 57-69.