A reader of the column I write on tutoring seniors wrote in recently to point out that an incident I described in that column really was an example of how to use (or misuse) rational thought in making decisions. The incident involved installing a wireless repeater in my client’s house so they could get coverage from one end to the other. I have set up several wireless networks, but never had occasion to install a repeater. Getting it working correctly involved a series of decisions that were based on totally inadequate data, but like most decisions, it worked.
The adventure started when they complained about the limited range of their access point, and I had happened to see an advertisement for on off-brand repeater on sale at Fry’s. We discussed possibilities, and they said they would like to try it. So I bought a cheapie unit and, sure enough, the software that came with it was minimal, and the documentation almost written in English. I could not get it to talk to their system. At this point the decision making paradigms come into play. I could have returned it or kept playing with settings. But you have to fold into the decision-making matrix the cost of my time and inconvenience to the clients weighed against such things as my desire to learn how to do what should have been a simple task. Oh, yes, there is also the little matter of not wanting to leave the scene with egg on my face.
I decided that performing a new function in front of customers was an error. We all make that mistake. So to recover, I took the repeater home to try it on my system. That worked. I did everything as I did before, but now it worked. Here is another decision point. Did I do something wrong the first time, or was there a basic incompatibility? At this point, I would have been justified in trying again, but instead of risking further flailing in front of customers, I returned the cheapie unit and purchased a repeater of the same brand as my client’s router.
Having learned my lesson about performing in public, I installed their new repeater on my home system first. This was another decision. What was the rationale for this action? After all, the bad unit worked on my system, so what did I expect of the more expensive one? I buttoned it back up and took it to their house where it came up immediately and everyone seems to be happy.
Why did I exchange units? The replacement one cost almost three times as much. Class costs – right? Well, no. I could give you counter-examples of the lack of correlation between the price of an object and its utility. So cost was a factor, but not a driver. That is, I thought the probability of getting a working unit would correlate with price, but not guarantee success. What really swayed me was the idea of sticking with the same brand. Standard protocols and interchangeability are great ideas, but not all manufacturers stick to the standards. I thought that the probability of units from the same manufacturer being compatible was higher than the probability of compatibility of units from competing manufacturers even though both claim to meet industry standards. This conjecture was partially confirmed at Fry’s when I exchanged units. The salesperson said that even sticking within brands is risky because they tend to be finicky – and this came from a person who was trying to sell a unit! That was more data to feed into my internal decision-making process.
Now throughout this process did I mentally made a decision tree and apply Bayesian theory? Did I make a mental payoff matrix and optimize my return? Of course not. But on a more informal basis, the lessons of a lifetime have been assimilated such that I was indeed considering the various tradeoffs without even bothering. That is one benefit of at least having a nodding acquaintance with the proper way to analyze input.
In response to the interest my original tutorial generated, I have completely rewritten and expanded it. Check out the tutorial availability through Lockergnome. The new version is over 100 pages long with chapters that alternate between discussion of the theoretical aspects and puzzles just for the fun of it. Puzzle lovers will be glad to know that I included an answers section that includes discussions as to why the answer is correct and how it was obtained. Most of the material has appeared in these columns, but some is new. Most of the discussions are expanded compared to what they were in the original column format.
[tags]experience, decision theory, analysis, data[/tags]