Bootstrapping procedure: Significance level

Frequently asked questions about PLS path modeling.
Post Reply
wassimbenchikha
PLS Junior User
Posts: 3
Joined: Thu Mar 26, 2015 3:25 pm
Real name and title: Wassim Ben Chikha

Bootstrapping procedure: Significance level

Post by wassimbenchikha » Thu Mar 26, 2015 3:30 pm

Dear all,
Please, I did not understand the utility of significance level in bootstrapping procedure. In fact, in the previous version of PLS (i.e., 2.0) this option does not exist and the validity of the model’s relationships is done using T Statistics test. Please, can you clarify this point.

With Kind regards
Wassim

User avatar
Hengkov
PLS Super-Expert
Posts: 1618
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Re: Bootstrapping procedure: Significance level

Post by Hengkov » Sun Mar 29, 2015 3:43 pm

Hi,

In SmartPLS 3, has provided the output P-value with a choice of a one-tailed and two-tailed depends on hypothesis you make.

Greetings,

wassimbenchikha
PLS Junior User
Posts: 3
Joined: Thu Mar 26, 2015 3:25 pm
Real name and title: Wassim Ben Chikha

Re: Bootstrapping procedure: Significance level

Post by wassimbenchikha » Sun Mar 29, 2015 7:37 pm

Hi,

Thank you for your response.

Please, how can I interpret the results from P values? In fact, In PLS 2, T-statistics are larger than 2.58, 1.96 and 1.65 for significance level of 1%, 5% and 10%, respectively.

Many thanks for your help and look forward for your response.

With best regards
Wassim

User avatar
Hengkov
PLS Super-Expert
Posts: 1618
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Re: Bootstrapping procedure: Significance level

Post by Hengkov » Tue Mar 31, 2015 9:28 am

Hi,

It is easy to interpret with the P-value, depending on the level of significance that you are using and the direction of the hypothesis (eg 5%), means that if the result is < 0.05 then H0 is rejected (two-tailed test).

Greetings,

wassimbenchikha
PLS Junior User
Posts: 3
Joined: Thu Mar 26, 2015 3:25 pm
Real name and title: Wassim Ben Chikha

Re: Bootstrapping procedure: Significance level

Post by wassimbenchikha » Tue Mar 31, 2015 7:06 pm

Hi,

Thank you for your response,

It is not clear for me.

Please Find attached an example. Here, I have used a significance level of 0.01. The P values of these three hypothesis is <0.01, then should I reject these hypothesis?

Many thanks for your help and look forward for your response.

With best regards
Wassim
Attachments
example.png
example.png (37.24 KiB) Viewed 6544 times

jmbecker
SmartPLS Developer
Posts: 1129
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Bootstrapping procedure: Significance level

Post by jmbecker » Tue Apr 07, 2015 7:33 pm

The setting of the significance level only affects the confidence intervals.
For T and p-values the setting you have done for the significance level is irrelevant (not the direction of the hypothesis one or two-tailed)
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

Post Reply