1
$\begingroup$

enter image description hereI conducted a path analysis for my dissertation, I am not sure how should I interpret the follow result?

  • Non significant direct or indirect paths between IV and DV1,DV2,DV3.
  • Significant path between IV and M.
  • Significant path between M and DV1.
  • Non significant paths between M and DV2 and DV3.
  • A significant path between DV1 and DV3.
  • Non significant path between DV2 and DV3, and between DV1 and DV2.

I understand there is no mediation but I am confused about how to refer to the significant paths in the context of the current model.

$\endgroup$
  • $\begingroup$ Welcome to CrossValidated. Have you heard the line "The difference between significant and non-significant is itself not significant"? You are better off telling the magnitude of the path coefficients you have found. Uploading a path diagram would also help. And finally, could you be more specific than "how should I interpret"? It's so broad that I'm afraid this question is likely to be closed as unanswerable in this format. $\endgroup$ – rolando2 Apr 11 '18 at 14:15
  • $\begingroup$ Ok, thank you. I added a diagram and i tried to be more specific. I hope it is better now. $\endgroup$ – weird smile Apr 11 '18 at 14:50
  • $\begingroup$ Don't try to put the indirect paths on the path diagram. That makes it too confusing to understand. $\endgroup$ – Jeremy Miles Apr 11 '18 at 20:43
0
$\begingroup$

One suggestion is to (1) rerun the model with all the non-significant direct paths removed, (2) confirm that the parameters of the other models do not substantively change, and (3) redraw the model with only direct paths drawn in.

If you do so, and if the results are comparable, you will have a much simpler model to interpret. From what you have presented, it sounds as though you have a true mediation between IV and DV1 thru M. It appears there may be a correlation with DV1 and DV3, and this correlation exists separate from any influence from IV or M. Furthermore, DV2 and DV3 are not related to either IV or M (thus, could for all practical purposes, be dropped from the model).

$\endgroup$

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.