Logistic regression error message

#1
Hi everyone,

I am currently investigating the bivariate association between educational level and political participation. For this I am using a logistic regression but I keep getting the error message " Logistic Regression - The dependent variable has more than two non-missing values. For logistic regression, the dependent value must assume exactly two values on the cases being processed." And I know this assumption is true, and I thought I had coded my dependent variable in the right way...but apparently not.

Here's what I did:
compute polparti= MEAN(contactpt, workedpt, workorg, wearbadge, signpet, lawdemo, onlinept).
EXECUTE.
VALUE LABELS polparti 0 'no participate' 1 'participate'.

And all the items I used (contactpt, workedpt, working, etc.) were coded in this way to get rid of missing values:
recode contplt (1=1) (2=0) (ELSE=SYSMIS) INTO contactpt.
recode wrkprty (1=1) (2=0) (ELSE=SYSMIS) INTO workedpt.
recode wrkorg (1=1) (2=0) (ELSE=SYSMIS) INTO workorg.
recode badge (1=1) (2=0) (ELSE=SYSMIS) into wearbadge.
recode sgnptit (1=1) (2=0) (ELSE=SYSMIS) into signpet.
recode pbldmn (1=1) (2=0) (ELSE=SYSMIS) into lawdemo.
recode pstplonl (1=1) (2=0) (ELSE=SYSMIS) into onlinept.
execute.

When I look at the frequency distribution of the polparti variable I see the issue because it basically goes from 0 to 0.014 to 0.018 to 0.029 etc. until it reaches 1 (participate), but I don't know where it went wrong and how I can correct this. Any help would be very much appreciated!!!

Thanks in advance.
 

Attachments

Dason

Ambassador to the humans
#2
I don't use spss but it seems the issue is that what you really want to do is recode so anything above 0 gives you 'participate' but what you've written only recodes to participate if the value equals 1.