What is BETA
Locked
 
|
Hello all,
In the following MANOVA output what is BETA: Regression analysis for WITHIN+RESIDUAL error term --- Individual Univariate .9500 confidence intervals Dependent variable .. bp.syst.upper.arm BP Systolic (Upper Arm) (mmHg) COVARIATE B Beta Std. Err. t-Value Sig. of t Lower -95% CL- Upper amlodipi -.8520235365 -.2138057332 .07923 -10.75336 .000 -1.00736 -.69669 ramipril -.0186825348 -.0045037579 .10296 -.18145 .856 -.22053 .18317 indapami -3.8671405370 -.3176786994 .47517 -8.13840 .000 -4.79870 -2.93558 ===================== 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 |
|
Hi Max,
BETA is the standardized coefficient. On Fri, Jul 10, 2009 at 9:06 AM, MaxJasper<[hidden email]> wrote: > Hello all, > > In the following MANOVA output what is BETA: > > Regression analysis for WITHIN+RESIDUAL error term > --- Individual Univariate .9500 confidence intervals > Dependent variable .. bp.syst.upper.arm BP Systolic (Upper Arm) (mmHg) > > COVARIATE B Beta Std. Err. t-Value Sig. of t Lower -95% CL- Upper > > amlodipi -.8520235365 -.2138057332 .07923 -10.75336 .000 -1.00736 -.69669 > ramipril -.0186825348 -.0045037579 .10296 -.18145 .856 -.22053 .18317 > indapami -3.8671405370 -.3176786994 .47517 -8.13840 .000 -4.79870 -2.93558 > > ===================== > 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 |
|
|
In reply to this post by MaxJasper
On Fri, 10 Jul 2009 00:06:33 -0600, MaxJasper <[hidden email]> wrote:
> Hello all, > > In the following MANOVA output what is BETA: > > Regression analysis for WITHIN+RESIDUAL error term > --- Individual Univariate .9500 confidence intervals > Dependent variable .. bp.syst.upper.arm BP Systolic (Upper Arm) > (mmHg) > > COVARIATE B Beta Std. Err. t-Value The ordinary regression coefficient, B, depends on the units of measurement, so if we do the same research, but you measure time in hours and I measure it in days, then our results will differ by a factor of 24. It is therefore difficult to compare the importance of one variable with another in the same study, if the variables are measured in different units (and how could you measure time and weight in the same units?) If you consider the variables as standardised, i.e. measured in standard deviation units, then the unit of measurement drops out and the equation is expressed in absolute numbers, that is in terms of BETA. David Hitchin ===================== 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 |
Administrator
|
I have two comments on what David has said. First, I'm not very comfortable with the idea that the standardized regression equation is "expressed in absolute numbers". What is absolute about them? They are relative to the standard deviation. Second, well much of what David says could be taken straight from any of a number of textbooks, I think that many of those books present a somewhat simplistic view of this topic. I recently read a bit about standardized regression coefficients in John Fox's book (Applied Regression Analysis and Generalized Linear Models, 2008, Sage), and he is very cautious about their use. He gives this interesting example. When two variables are measured on the same scale (e.g., years of education, and years of employment), then relative impact of the two can be compared directly. But suppose those two variables differ substantially in the amount of spread. In that case, comparison of the standardized regression coefficients would likely yield a very different story than comparison of the raw regression coefficients. He then says: "If expressing coefficients relative to a measure of spread potentially distorts their comparison when two explanatory variables are commensurable [i.e., measured on the same scale], then why should the procedure magically allow to compare coefficients [for variables] that are measured in different units?" (p. 95) I like that a lot. Cheers, Bruce
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
|
|
In reply to this post by MaxJasper
You are all so good. Thanks a lot for your comments.
===================== 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 |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |