Ideas

As a psychologist (almost!), we do a lot of work with people making decisions. If the technical jargon is removed;

"Decision tree analysis is a statistical tool for determining the mathematically correct decision in a situation with cascading events and multiple levels of uncertainty"

what you are essentially left with is a mind-hack - a guided efficient decision making process. I can visually imagine these decision trees unfolding in therapy, and the discussions that could focus around the different elements of the decision tree. Often visual cues/anchors are required in therapy to help drive home points.

Simply put - if this is printable and interactive, I would use this in therapy. If this projects gets moving, I would love to be involved

PS - ill get you that damn doughnut!!!

sweet!! i like devil's food and powdered doughnuts the best... ;-)

you can try out the desktop-based decision tree analysis products out there right now on either treeage.com or vanguardsw.com. You actually don't need software at all to map out a decision tree but it makes it easier and solving the tree is a trivial task on a computer.

Porting this type of app to the web and then enabling the collaborative and component library aspects plus adding a wizard-driven quickstart interface are the key differentiators. The incumbent players are not going to pursue this because pushing an alternate low-dollar subscription-based model would cannibalize their current high-margin desktop software products. It will be a disruptive play from a new entrant that makes this happen.

definitely try it with your patients/clients - seeing the visual tree of consequences is useful in itself even without the probabilities and weights added.

sean

Using a decision tree is one way of sorting out what matters from what doesn't.

It has the capacity (as PsychSplash pointed out) to be used elsewhere, not just in financial decisions.

You do have the garbage in-garbage out problem to contend with. The reputation of the product will suffer if it is not robust enough to properly serve the users in the target market, if it is going to a mass market.

I know there is an Excel add-in for decision tree analysis (sold inexpensively by a third party as shareware). What that type of product needs to give it mass appeal is a GUI and user-friendly hand-holding to walk people through the process. (If the customers are not into math & stats).

BTW when n=1, I'm not convinced that the result of using a decision tree is necessarily the best answer, but I think that going through the analytical process forced by the tree leads to a more considered decision.

Voting it up, cheers.

Jill,
thanks. clearly the ease of use of the interface and an effective video tutorial would be the determinant whether it achieves mainstream usage or not. I could see doing a wizard interface that asks a series of questions to generate the tree. Once there is a decent library of commonly-made decisions like whether or not to refinance a house or take on a new rental property, you could make it so the user chooses from an existing component to work with and modify to their needs.

your assessment of the "garbage in/ garbage out" phenomena is correct - the trees are only as good as the info that's fed into them so confidence in the accuracy of probabilities and weighting on the part of the user is imperative.

thanks for the feedback
sean