How do you determine which customer, or citizen group, best matches your offerings or programs? What if you had a way to avoid spending a lot of money and time trying to reach people and instead could more efficiently reach them—and increase your return on investment?

The segmentation capabilities in IBM SPSS AnswerTree enable you to more accurately target people and eliminate the guesswork that many direct mail software solutions impose. Results from IBM SPSS AnswerTree can complement your direct mail software and eliminate your need to send offers to arbitrary groups of people—or everyone—on your mailing list.

IBM SPSS AnswerTree enables you to more efficiently target the right groups of people. Use scalable decision trees to reveal segments and predict how groups will respond to your promotions and programs. IBM SPSS AnswerTree gives you the widest choice of decision trees available through four powerful algorithms, including CHAID, so you can choose the best fit for your data.

IBM SPSS AnswerTree visually displays decision tree models—so you can easily see the groups that matter. At-a-glance evaluation graphs enable you to act on results quickly. Answer Tree's unrivaled analytical power, scalability, and presentation features give you a smarter, more effective way to discover significant groups and create profiles.

IBM SPSS AnswerTree 3.1 is available in English, Japanese, French, German, Italian, and Spanish. Contact your local office to find out more.

Turn data into strategic information

With IBM SPSS AnswerTree, you can easily detect segments and patterns, such as "high-profit customers who are likely to respond to Web offerings" or "students who miss more than 45 days of school a year are twice as likely to drop out."

How decision tree results are different in IBM SPSS AnswerTree 3.1/3.0

IBM SPSS AnswerTree 3.1/3.0 has implemented consistent rules for handling ties between predictors when determining splits. In the past, IBM SPSS AnswerTree did not always follow consistent rules about what to do in case of a tie. The most obvious result of these rules is that some trees developed in previous versions of IBM SPSS AnswerTree will look different if built in IBM SPSS AnswerTree 3.1/3.0.

Because these tie-breaking rules affect more extreme situations and are often arbitrary decisions, the results in any version are equally valid. For example, in C&RT, the order that the predictor was added to the model-an arbitrary rule-determines the split. The difference is that now you can understand and explain how a split is made in these situations. It is also worth noting that in these situations it may be worth using IBM SPSS AnswerTree's interactive tree building features to choose the predictor, since you may be more interested in including one attribute instead of another equally significant predictor.

The algorithms use different rules to determine which predictor to use. For example, with a continuous target, CHAID will choose the predictor whose split gives the most significantly different groups of the target. For a categorical target, CHAID selects the predictor whose measure of independence with the target is smallest. (See Appendix E in the IBM SPSS AnswerTree User's Guide for details.) If there is a "tie" between two predictors, then IBM SPSS AnswerTree needs to choose one predictor over another. In 3.1/3.0 the rules are always consistent:

For CHAID & Exhaustive CHAID:

1. The split rule with lower p value is listed first
2. In case of tie, the rule with higher F (or Chi-Square) is listed first
3. In case of tie, the rule with lower DF is listed first
4. In case of tie, the rule whose associated predictor variable was added first to the model is listed first

For QUEST:

1. The split rule with lower p value is listed first
2. In case of tie,
1. Order among all the tied Chi-Square. The larger value listed first.
2. If there are ties in the item above, order by DF. The smaller DF listed first.
3. Order among all the tied F (F or Levene's F). The larger value listed first.
4. If there are ties in the item above, order by DF1+DF2 with the larger sum listed first. If there still is a tie, order by DF1 with the smaller value listed first. (note: sort by DF1+DF2 will put the predictor with fewer missing values before predictor with more missing values)
5. To set ordering among these two groups (Chi-Square group and F group), list first the predictor added first to the model

C&RT has not changed. The rules are as follows:

1. The split rule with higher improvement is listed first
2. In case of tie, the rule whose associated predictor variable was added first to the model is listed first

• Operating system: Windows 98, 2000 or Windows NT 4.0 with Service Pack 5 or higher
• Hardware: Pentium-class processor, SVGA monitor and CD-ROM drive for installation
• Minimum free drive space: 70MB for software
• Minimum RAM: 64MB or more
• Microsoft Internet Explorer 5.0 for reading help documents

• Windows NT Server, Windows 2000 Server or Windows 2000 Advanced Server:
  • Hardware: Pentium-class processor, SVGA monitor and CD-ROM drive for installation
  • Minimum free drive space: 70MB
  • Minimum RAM: 64MB or more
• Solaris 2.6, 7 and 8:
  • Hardware: Ultra Sparc 2 (or better) and CD-ROM drive for installation
  • Minimum free drive space: 70MB
  • Minimum RAM: 256 MB