grouping Independent variables Multiple regression analysis.

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

Posts: 1
Joined: Mon Jun 04, 2012 7:32 am

grouping Independent variables Multiple regression analysis.

Postby Ali.Aziz1984 » Mon Jun 04, 2012 7:43 am

Dear all,

I want to perform a Multiple regression analysis on my dataset. My data set has approximately 10 Independent variables which I want to use to predict the variation in my dependent variable. In the Multiple regression analysis i want to perform the backward selection.

My question/problem is as follows:

Is it possible to group a couple of the independent variables as one? so that in backward selections SPSS see is as one variable?
here an example...

a couple of Independent variables are: weight, age, sex, length. Is it possible to group these IV to one? say a group variable name with: "clinical facotrs" so that SPSS takes these four 'of the model' at once during backward selection? (excuse me for my bad english :) )

Thanks in advance,
Posts: 473
Joined: Thu Apr 05, 2012 5:58 pm

Re: grouping Independent variables Multiple regression analy

Postby Penguin_Knight » Tue Jun 05, 2012 1:56 am

Yes and no.

In the panel where you put your independent variable, you can see there is something called "Block 1 of 1", if you click the "Next Button", then you'll get to Block 2 of 2, in which you can put in another set of variables. If you do that, SPSS will fit the regression with your variable in Block 1, and then add the variables in Block 2. The beauty is that it will also provide you statistics on change of R-squared and F-test for the added variables from each block. You can then manually remove the whole block if they don't contribute anything.

See here for more details: ... /mlt07.htm

Who is online

Users browsing this forum: No registered users and 1 guest