Which model to use?
Locked
 
|
Hi all
I need to construct a model in SPSS. The dependant variable is categorical - (3 categories - ordinal) The independant variables are categorical (dichotomous, nominal and ordinal - 2-6 levels) I thought I would have to use a multinomial logistic model - is this correct - because the output did not contain dummy variable results - and didn't make a lot of sense Many thanks to the wonderful people who help me |
|
|
Charlene,
Not multinomial logistic. Ordinal logistic. Look at the the (cryptically named) PLUM command. I think there are now other alternatives as well. Perhaps others can comment on the relative strengths of PLUM and other commands. Gene Maguin -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of charlene thornton Sent: Saturday, September 24, 2011 6:57 AM To: [hidden email] Subject: Which model to use? Hi all I need to construct a model in SPSS. The dependant variable is categorical - (3 categories - ordinal) The independant variables are categorical (dichotomous, nominal and ordinal - 2-6 levels) I thought I would have to use a multinomial logistic model - is this correct - because the output did not contain dummy variable results - and didn't make a lot of sense Many thanks to the wonderful people who help me -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Which-model-to-use-tp4836365p4 836365.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 |
Administrator
|
Re PLUM, here's a chapter you might find useful:
http://www.norusis.com/pdf/ASPC_v13.pdf If the "parallel lines assumption" is violated too grossly, multinomial logistic regression would be one alternative you could use.
--
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 Maguin, Eugene
CATREG Categorical Regression would be another alternative.
Art Kendall Social Research Consultants On 9/25/2011 1:23 PM, Gene Maguin wrote: > Charlene, > > Not multinomial logistic. Ordinal logistic. Look at the the (cryptically > named) PLUM command. I think there are now other alternatives as well. > Perhaps others can comment on the relative strengths of PLUM and other > commands. > > Gene Maguin > > -----Original Message----- > From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of > charlene thornton > Sent: Saturday, September 24, 2011 6:57 AM > To: [hidden email] > Subject: Which model to use? > > Hi all > > I need to construct a model in SPSS. > > The dependant variable is categorical - (3 categories - ordinal) > The independant variables are categorical (dichotomous, nominal and ordinal > - 2-6 levels) > > I thought I would have to use a multinomial logistic model - is this correct > - because the output did not contain dummy variable results - and didn't > make a lot of sense > > Many thanks to the wonderful people who help me > > -- > View this message in context: > http://spssx-discussion.1045642.n5.nabble.com/Which-model-to-use-tp4836365p4 > 836365.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 > ===================== 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
Art Kendall
Social Research Consultants |
|
|
In reply to this post by Bruce Weaver
Hello list,
I have a file with a number of very long string variables (around 200 to 400 characters). I wanted to use the SPSS MODIFY TABLES extension command in order to print them. This is the syntax I tried: SUMMARIZE FORMAT LIST NOTOTAL NOCASENUM /VARIABLES v07. SPSSINC MODIFY TABLES SUBTYPE = 'Report' PROCESS = ALL SELECT = -1 /WIDTHS WIDTHS = <any number>. The width of the column does indeed change, but whatever I substitute for <any number>; the width changes to the same length, which is to broad to be printed (Message: Cannot fit row labels on one page - try "Rescale Wide Table to Fit Page" in Format ! Table Properties"). Changing SELECT = -1 into SELECT = "v01" doesn't help. There is no problem if I change the width "manually" in the pivot editor Can someone tell me what I'm doing wrong? And for the wish list... could there be an option to set "Rescale All Wide Tables to Fit Page" as a default. Regards, Antoon Smulders ===================== 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 |