Interpreting stepwise regression results

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

natura
Posts: 5
Joined: Wed Aug 12, 2015 7:12 pm

Interpreting stepwise regression results

Postby natura » Wed Aug 12, 2015 7:20 pm

Hi!
I'm having problems in understanding the results of my analysis. If my model selects an indipendent variable with a negative t coefficient,does it means that there's a negative correlation? For example: my dipendent value is "salary" and the indipendent value selected by SPSS is "BMI", with beta coeff. -0,16, t -2,5 and p-value 0,00... Does it means that less money are predictors for higher BMI?
I hope my question is enough clear
natura
Posts: 5
Joined: Wed Aug 12, 2015 7:12 pm

Re: Interpreting stepwise regression results

Postby natura » Thu Aug 13, 2015 6:36 am

This is an exemple!Please can you help me to understand how to explain this tab?
You do not have the required permissions to view the files attached to this post.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Interpreting stepwise regression results

Postby GerineL » Thu Aug 13, 2015 10:35 am

You can look at the beta coefficient: negative means a negative relation, so indeed, higher values on the predictor predict lower values on the outcome.
natura
Posts: 5
Joined: Wed Aug 12, 2015 7:12 pm

Re: Interpreting stepwise regression results

Postby natura » Thu Aug 13, 2015 10:49 am

Thank you for your reply!!! :D You mean standardized beta ? :?
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Interpreting stepwise regression results

Postby GerineL » Thu Aug 13, 2015 1:27 pm

beta is always standardized, therefore it is under the heading " standardized coefficients" . However looking at your table again, it is very surprising: you ahve no b and no se b, but a sig beta?
I don't get it, are you sure there are no problems with your variable?

please post the entire table (not just this part of it) including notes below the table, just in case...
natura
Posts: 5
Joined: Wed Aug 12, 2015 7:12 pm

Re: Interpreting stepwise regression results

Postby natura » Thu Aug 13, 2015 2:06 pm

OMG sorry! I messed up while copying&pasting so values changed. Sorry.
This is the complete table
You do not have the required permissions to view the files attached to this post.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Interpreting stepwise regression results

Postby GerineL » Fri Aug 14, 2015 8:52 am

aah good, this looks like it makes sense :-)

so as you see, the - sign is equal between b, beta and t.

if you look for "interpreting spss output regression" or something comparable, you can find great tutorials online explaining how to interpret b vs beta etc.
natura
Posts: 5
Joined: Wed Aug 12, 2015 7:12 pm

Re: Interpreting stepwise regression results

Postby natura » Fri Aug 14, 2015 10:37 am

Ok! Thank you very much!!! 8)

Who is online

Users browsing this forum: No registered users and 1 guest

cron