Interpret Chi Squared

Feel free to post your requests for new SmartPLS features and other suggestions.
Post Reply
Newbie456
PLS Junior User
Posts: 4
Joined: Sat Mar 04, 2017 1:16 pm
Real name and title: Mr G Redman

Interpret Chi Squared

Post by Newbie456 » Sun Mar 19, 2017 10:10 am

Hi all,

So I seem to have a very 'high' Chi² value of 357.47 and I am struggling to interpret this. As per the SmartPLS Documentation, Chi² value is calculated using the following formula (N-1)*L (N is the number of observations and L is the maximum likelihood function). I know that Chi² is used to test the significance of my findings and see if this matches whether the distribution of values in a statistical sample conforms to a known theoretical distribution.

I have 184 respondents and I know that degrees of freedom is used in the calculation however I do not know how to interpret the value. If I don't know the df then how will I know what p value is associated to the Chi² value and know if the value 357.47 means I can reject the null hypothesis? I was under the impression that a low Chi value means that the null hypothese can be rejected. Can 357.47 be converted to a percentage?

Is anyone able to shed more light on this?

Thanks!

User avatar
cringle
SmartPLS Developer
Posts: 796
Joined: Tue Sep 20, 2005 9:13 am
Real name and title: Prof. Dr. Christian M. Ringle
Location: Hamburg (Germany)
Contact:

Re: Interpret Chi Squared

Post by cringle » Sun Mar 19, 2017 10:49 am

You find more info about Chi² and other fit measures here: https://www.smartpls.com/documentation/fit

You may not want to interpret the value (if the size depends on the model complexity) but criteria such as NFI that use Chi² but take the model complexity into account.

Best regards
Christian

Newbie456
PLS Junior User
Posts: 4
Joined: Sat Mar 04, 2017 1:16 pm
Real name and title: Mr G Redman

Re: Interpret Chi Squared

Post by Newbie456 » Sun Mar 19, 2017 11:20 am

Hi Christian,

Thank you for your response. I have read through the entire section on Smart PLS documentation.

Could you please elaborate on "if the size depends on the model complexity" ?

So if I have NFI could ignore the Chi Squared as this is taken into account in NFI?

Thanks,

User avatar
cringle
SmartPLS Developer
Posts: 796
Joined: Tue Sep 20, 2005 9:13 am
Real name and title: Prof. Dr. Christian M. Ringle
Location: Hamburg (Germany)
Contact:

Re: Interpret Chi Squared

Post by cringle » Sun Mar 19, 2017 3:35 pm

Could you please elaborate on "if the size depends on the model complexity" ?
-->Don't know how to better explain it: Smaller model = lower Chi² value; more complex model = higher Chi² value.
So if I have NFI could ignore the Chi Squared as this is taken into account in NFI?
--> But the NFI does not ignore Chi². Please check the explanation here, again: https://www.smartpls.com/documentation/fit

Newbie456
PLS Junior User
Posts: 4
Joined: Sat Mar 04, 2017 1:16 pm
Real name and title: Mr G Redman

Re: Interpret Chi Squared

Post by Newbie456 » Sun Mar 19, 2017 9:30 pm

Thank you, that is helpful.

Post Reply