Percent Change Crosstabs

[Bruce Weaver]

Oops, someone pointed out off-list that I had it backwards below.  I meant to say:

  RR - 1 = .873 - 1 = -.127 = -12.7%

I hope this did not cause undue anguish for anyone.  ;-)

Bruce Weaver wrote

Ryan did not say RR = -12.7.  He said:

   % change in risk  = (RR - 1)*100=[(41.9/48.0) - 1]*100 = -12.7

RR = 41.9%/48.0% = 0.873
1 - RR = 1 - .873 = -.127 = -12.7%


Re the GENLIN syntax I gave (see below), the coefficient for Group gives the risk difference (with SE and CI).  Tony pointed out that one can get the same thing from the Somer's d table generated by CROSSTABS.  But as he said, and my example showed, you have to know which of the 3 results you want.  That's one reason why I prefer the GENLIN approach.  (Another reason is that you can add other explanatory variables.)


GENLIN Recid BY Group
   /MODEL Group INTERCEPT=YES
  DISTRIBUTION=NORMAL LINK=IDENTITY
   /CRITERIA SCALE=MLE COVB=ROBUST PCONVERGE=1E-006(ABSOLUTE) SINGULAR=1E-012
 ANALYSISTYPE=3(LR)
     CILEVEL=95 CITYPE=WALD LIKELIHOOD=FULL
   /MISSING CLASSMISSING=EXCLUDE
   /PRINT CPS DESCRIPTIVES MODELINFO FIT SUMMARY SOLUTION.

HTH.

DKUKEC wrote

Thank you all for your contributions.  

Ryan, with respect to RR=-12.7; how would I report this.  Would I report that the treatment group reduced the risk of recidivism by 12.7% when compared to the control group?

Bruce/Tony, I am still unsure how GLM and Crosstab (Somer's d) computes the difference in proportions...  what does the output mean in both examples and how does it compute the difference?  

I apologize in advance for my ignorance and thank you all for your imput.

Sincerely,
Damir