r^2 adjusted

[jimjohn]

i have a data set witha bout 34,000 cases. if i have a regression model with R^2 adjusted = .2 and p-val <0, can i conclude this si a decent model? most of my tests seem to have a p-val <0 because the data set is so big, but what should the minimum R^2 adjusted be before i can conclude its a good model. thanks!

[Ornelas, Fermin-2]

Note that this question is not well structured. No information is given about the model. In general a high r square is preferred .7 or higher (in a multiple linear regression), but one has to consider the purpose of the model.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of jimjohn
Sent: Wednesday, February 27, 2008 2:46 PM
To: [hidden email]
Subject: r^2 adjusted

i have a data set witha bout 34,000 cases. if i have a regression model with
R^2 adjusted = .2 and p-val <0, can i conclude this si a decent model? most
of my tests seem to have a p-val <0 because the data set is so big, but what
should the minimum R^2 adjusted be before i can conclude its a good model.
thanks!
--
View this message in context: http://www.nabble.com/r%5E2-adjusted-tp15724142p15724142.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

NOTICE: This e-mail (and any attachments) may contain PRIVILEGED OR CONFIDENTIAL information and is intended only for the use of the specific individual(s) to whom it is addressed.  It may contain information that is privileged and confidential under state and federal law.  This information may be used or disclosed only in accordance with law, and you may be subject to penalties under law for improper use or further disclosure of the information in this e-mail and its attachments. If you have received this e-mail in error, please immediately notify the person named above by reply e-mail, and then delete the original e-mail.  Thank you.

=====================
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

[Hector Maletta]

Jimjohn,
The p-value is the probability that the actual population value of the
correlation coefficient is zero. According to your results, that probability
is very low (p<0.0005 or something like that). This is because R2=0.20 with
a sample as large as N=34,000 is enough to get reasonably certain that the
population R2 is not zero.
Now, being not zero does not mean that your R2 is sufficiently high as to
justify a theory. R2=0.20 means that your independent variables explain 20%
of the variance in the dependent variable, while other (unspecified) factors
explain the remaining 80%. For some purposes this might be rewarding enough,
but for other purposes it would be not satisfactory.
You must separate the two aspects in your mind: the p value answers to the
question whether your sample result for R2 is a good representation of the
population value of R2. The R2 value itself is an indication of the strength
of the relationship between predictor and predicted variables. Even with
R2=0.001, you may have a low p value if the sample is large enough. The
result would be statistically significant (i.e. significantly different from
zero) but perhaps not substantively meaningful (most of the variance is left
unexplained).

Hector

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
jimjohn
Sent: 27 February 2008 19:46
To: [hidden email]
Subject: r^2 adjusted

i have a data set witha bout 34,000 cases. if i have a regression model with
R^2 adjusted = .2 and p-val <0, can i conclude this si a decent model? most
of my tests seem to have a p-val <0 because the data set is so big, but what
should the minimum R^2 adjusted be before i can conclude its a good model.
thanks!
--
View this message in context:
http://www.nabble.com/r%5E2-adjusted-tp15724142p15724142.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