so my model in SEM has an good fit, and I let R plot my model including path co-efficients (it is automated so I figure the numbers on all paths are path coefficients?).

I cant get a grip on what they mean though... can anyone shed a light on it? Is there a cut-off value like with RSMEA? When are they significant?

Thank you so much


A parameter estimate in a structural equation model is either a covariance (curved arrow) or a regression parameter (straight arrow). They are interpreted in the same way as you interpret these parameters in any more. It's a little easier to explain if you gave an example.

  • $\begingroup$ I measure the influence of three Latent variables (A,B,C) on one Mediator (D) and one final outcome (E). All are straight arrows and I have to write a page of interpretation on what the values of the arrows mean. All arrows from A have negative numbers ranging from -0.001 to -0,23. B has 0,67 to D but 1,9 to E. C has 0.7 to both D and E. The straight error between D and E has a value of 0.09. $\endgroup$ – visiteur123 Nov 6 '16 at 12:46
  • $\begingroup$ Straight error? $\endgroup$ – Jeremy Miles Nov 7 '16 at 5:54
  • $\begingroup$ So this is homework? Please tag with self-study. $\endgroup$ – Jeremy Miles Nov 7 '16 at 5:55
  • $\begingroup$ *straight arrow + edited the tag. Am new to the forum so pardon me please. $\endgroup$ – visiteur123 Nov 7 '16 at 10:41
  • $\begingroup$ Do you have a path diagram you can post? It's easier to explain if you do. $\endgroup$ – Jeremy Miles Nov 8 '16 at 21:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.