The Decision Bonsai are a hybrid of concept maps and decision trees. They were originally developed to give introductory statistics students a map to inference procedures but have evolved to be used for other topics. The tree is 'grown' during the semester so that students build a picture of the relationships in their mind. Recent work is moving toward the development of more complete concept maps for introductory statistics, statistical quality methods and probability and stochastic processes courses. These Decision Bonsai would be then pointed to at appropriate points in the concept maps.
Content Quality Concerns:
The Inference Bonsai describes the explanatory variable as "Explanatory Type" and labels them with parameters (p or Œµ) rather than categorical or quantitative. Explanatory variables generally do not have a parameter of interest. Portraying one-sample procedures as being an explanatory variable with one category is a stretch. The Inference Bonsai also does not consider model assumptions, as if choosing the correct inference procedure does not depend at all on the validity of the model assumptions. The first branch in the Probability Bonsai isn't labeled Yes/No. It uses some verbiage I was not familiar with, specifically listing "Compare Means" under a heading of "Proportion", which confused me.
Content Quality Strengths:
This material describes a good teaching method and a way for instructors to help their students review and conceptualize the major topics in a statistics course. It covers most inference procedures in an introductory course.
Ease of Use Concerns:
An instructor would need to take this idea and re-construct it pretty much from scratch in order to use it in a class. That's not to say it wouldn't be worth it, but this is not a ready-to-use material. Very visually unappealing. Some of the annotations of the Probability Bonsai are particularly hard to understand.
Ease of Use Strengths:
It's a good idea, and a good way to conceptualize a statistics course. The pdf introduction makes it clear that these Bonsais are developed throughout the course and do not magically appear in their final form at the end.
Potential Effectiveness Concerns:
An instructor would have to build their course around this framework for it to be helpful for students. While it makes students think systematically to follow the decision tree, it discourages understanding of why a particular test or probability model is used in a particular situation.
Potential Effectiveness Strengths:
I think this could be a very valuable resource for instructors to use as they create lectures and learning materials for a course. It gives students a framework to think about the decisions they have to make.
Potential Effectiveness Rating:
Source Code Available:
Source Code Available
Intended User Role:
Free for All