Comparing 2 regression models with same DV and different IVs

[John D Crawford]

I would like to know how to obtain the statistical significance of the difference in R-squareds obtained from 2 regression equations, in the same sample, and with the same dependent variable, but with different sets of predictor variables. I would also like to know the answer to the similar question with logistic regression. That is, how to obtain the statistical significance of the differences in predictive ability for two models, with the same binary dependend variable, but with a different set of predictors. Help much appreciated.

[Matthias Spörrle]

Hi John,

for regression analysis you might want to use
- the corresponding commands implemented in MBESS (e.g., ci.R2) or
- the program R2 (Steiger & Fouladi, 1992)
in order to compute the CI of R-squared.

HTH
M


On Wed, Mar 16, 2011 at 6:21 AM, Johncrawford <[hidden email]> wrote:

> I would like to know how to obtain the statistical significance of the
> difference in R-squareds obtained from 2 regression equations, in the same
> sample, and with the same dependent variable, but with different sets of
> predictor variables. I would also like to know the answer to the similar
> question with logistic regression. That is, how to obtain the statistical
> significance of the differences in predictive ability for two models, with
> the same binary dependend variable, but with a different set of predictors.
> Help much appreciated.
>
>
>
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Comparing-2-regression-models-with-same-DV-and-different-IVs-tp3759374p3759374.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

I was also thinking that one of Jim Steiger's programs might be useful for comparing R-squared values from the two linear regressions.  You can find them here:

   http://www.statpower.net/Software.html

The same question was posed in the SPSS newsgroup, where Ray Koopman suggested an article by Clarke (2007).

  http://groups.google.com/group/comp.soft-sys.stat.spss/browse_frm/thread/4604725282e27bc3#

HTH.

Matthias Spörrle wrote

Hi John,

for regression analysis you might want to use
- the corresponding commands implemented in MBESS (e.g., ci.R2) or
- the program R2 (Steiger & Fouladi, 1992)
in order to compute the CI of R-squared.

HTH
M


On Wed, Mar 16, 2011 at 6:21 AM, Johncrawford <[e-mail]> wrote:
> I would like to know how to obtain the statistical significance of the
> difference in R-squareds obtained from 2 regression equations, in the same
> sample, and with the same dependent variable, but with different sets of
> predictor variables. I would also like to know the answer to the similar
> question with logistic regression. That is, how to obtain the statistical
> significance of the differences in predictive ability for two models, with
> the same binary dependend variable, but with a different set of predictors.
> Help much appreciated.
>
>
>
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Comparing-2-regression-models-with-same-DV-and-different-IVs-tp3759374p3759374.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [e-mail] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>

=====================
To manage your subscription to SPSSX-L, send a message to
[e-mail] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD