From the course: Machine Learning and AI Foundations: Decision Trees with SPSS

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

How CHAID handles ordinal variables

How CHAID handles ordinal variables

From the course: Machine Learning and AI Foundations: Decision Trees with SPSS

Start my 1-month free trial

How CHAID handles ordinal variables

“

- [Instructor] Our ordinal variable will be passenger class. CHAIDs combined first class with second class indicated in model or with the notation less than or equal to two. Actually I can force it to break into three groups. So why did it form those particular groups? Why did it combine first and second and not second and third? Well first and second aren't terribly close. They're closer to each other than they are to third class at 62 and 48 percent survival. to third class at 62 and 48 percent survival. Third class is at 26 percent survival, but also CHAID is running a statistical test, which as we've seen is effected by factors like sample size. First and second class are not significantly different. So they've been combined. If the sample size were 10,000 the notes seven and eight would almost certainly have been kept apart. What's the difference between this example and the last one? Well with embarked, anything could be combined with anything. With first class, first could not…

Contents