Changing the significance level

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
ywwblue
PLS Junior User
Posts: 1
Joined: Fri May 19, 2006 8:41 am
Real name and title:

Changing the significance level

Post by ywwblue »

Hi,

I'd like to know the default value of the significance level, but
can't find it.
Could someone please tell me what the vaule is and how to change it?

Another question is that the t-statistics in the bootstrapping output are based on the 2-sided t test?

Thanks.
nicolevanveghel
PLS Junior User
Posts: 1
Joined: Sun Jun 11, 2006 12:02 pm
Real name and title:

Post by nicolevanveghel »

I am interested in this question as well?
Is there a possibility to change the sign. level?
Moreover, is there a possibility to view the sign. level?
It is not in the output at the moment.

Thanks
Nicole
stefanbehrens
PLS Expert User
Posts: 54
Joined: Wed Oct 19, 2005 5:53 pm
Real name and title:

Post by stefanbehrens »

Hi Aden et al.

Sorry, but I'm not sure I fully understand your question(s).

The bootstrapping procedure in SmartPLS produces t-values for many of the parameters/ estimates of your model (e.g., loadings, weights or path coefficients). These t-values are your raw input for running significance tests. However, PLS doesn't run those for you. You have to do it yourself.

Testing a path coefficient for significance works like this:
1) Decide on your desired alpha-level (e.g., 0.05)
2) Decide on a one- or two-sided test (depending on your hypothesis for the path)
3) Look up the critical t-value in a table using the number of boostrap runs (e.g., 1000) as the degrees of freedom

Examples:
a) For a one-sided test on significance at the 0.05-level with 1000 bootstrap runs your bootstrapped t-value should exceed 1.646 (or 2.330 at 0.01-level).
b) For a two-sided test with the same parameters the bootstrapped t-value has to exceed 1.962

Hope this helps.

Cheers,
Stefan
Post Reply