Date:         Mon, 19 Jan 1998 16:18:40 -0300
Reply-To:     hmaletta@overnet.com.ar
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         "Hector E. Maletta" <hmaletta@OVERNET.COM.AR>
Subject:      Re: seeking CHAID alternatives
Comments: To: Thomas J Welander <tomwelander@JUNO.COM>
Content-Type: text/plain; charset=us-ascii

Thomas J Welander wrote:
>
> Hey there,
> I'm a CHAID user needing to evaluate what else is available in the
> current crop of decision tree software.  I like CHAID and don't have any
> big desire to switch; the company needs me to do a routine software
> evaluation.
>
> We primarily use CHAID for customer sat. applications, namely, examining
> customer service attributes in survey data to see what predicts
> satisfaction, customer loyalty, and value.  For these purposes, the
> important features of CHAID are 1) its ability to compare predictors of
> mixed data types (nominal to ratio), 2) the predictor list output, and 3)
> segmentation tree output.
>
> I'm mainly looking for testimonials from users of CHAID and its
> alternatives.  Also, any and all mentions of program names and other
> places to find advice would be appreciated.  And please steer me toward
> the archives if y'all have covered this topic previously.
>
> I recognize some of your names from lurking sproradically in the past &
> look forward to getting acquainted.
>
> Thanks,
> Tom Welander
> Elrick & Lavidge
> Atlanta
 
 
Answer Tree is said to be the newly released SPSS alternative to CHAID.
In fact, the CHAID chi-square-based method is but one of the approaches
available with Answer Tree. But I haven't tasted it yet.
 
Hector Maletta
Universidad del Salvador
Buenos Aires, Argentina