Please don't judge me for asking this question, I am still learning statistics and come from a non-mathematical background...

I created a Conceptual Model using lavaan in R and got the following results:

enter image description here

In order to analyze the results and interpret which elements are the most important (rank them by significance effect), do I look at bets (estimate) or z-value?

Thank you for your help!

  • 1
    $\begingroup$ Could you clarify what you mean by "important"? $\endgroup$ – whuber May 6 '20 at 10:52
  • $\begingroup$ @whuber by important I meant significant. From p-values I can tell that they are all statistically significant under 95% confidence level. As I am doing a path analysis, I would like to rank the different variables based on their significance. And to do so, I used the "estimate" column $\endgroup$ – cookiemonster1697 May 6 '20 at 12:44
  • 1
    $\begingroup$ There doesn't seem to be anything left to ask, then: if you want to rank variables by their "significance," then you need to sort your table by the p-values, not the coefficient estimates. $\endgroup$ – whuber May 6 '20 at 12:52
  • $\begingroup$ @whuber thank you very much for your valuable help! If I just may ask, could you please tell me what the estimate column refers to then? $\endgroup$ – cookiemonster1697 May 6 '20 at 13:00
  • 1
    $\begingroup$ Those are the estimated coefficients in the model. $\endgroup$ – whuber May 6 '20 at 15:44

Your Answer

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

Browse other questions tagged or ask your own question.