All entries for Saturday 19 March 2011
March 19, 2011
We used the same material, and some of the tools we selected were the same, and we were in the same class and listened to the same lectures and seminars. Almost all the things turned out to be the same, however, the decision we made for Waverider were different from each other. This definitely was an interesting topic. Even though a slight different in our thinking can lead to a big different in the final outcome. Even in one group, we had the arguments about choosing tools. Also, after the analysis, we had a lot of discussion about our decision while we got the same results in our analysis. Thus, I can imagine the different opinions in a company or even outside a company. Also, even one person, it is impossible for him to make the same decision in different situation or different time though he faces the same problem. Therefore, we make a decision today, but perhaps we want to change our opinion tomorrow. So I can see the importance of the word ¡°robust¡± and of course, it is fairly difficult to achieve robust.
The recommendations of each group are different, either slightly or significantly because of different focuses. Some pay more attention on to the tangible costs and profits in the material while some groups stress more on potential costs and benefits. Even all the groups used decision tree as one of their decision tools, the results turn out to be different. One of the possible reasons could be the uncertainty information, for example, the probability of available chance in Llymingtom. Different solutions and ideas generate different approaches in analyzing this issue. Therefore, bias can¡¯t never be avoid no matter which tool is chosen. Everyone can use the same too in different ways and even in the same way, possibly the focuses are distinct, thus the results can¡¯t be same. Also, there was an interesting question raised yesterday, since we used more than one tools in analyze one decision, sometimes the results were totally incompatible. So can we say in this sense, bias can be reduced to a certain degree. However, as far as I see, if the results are confronted, bias may be reduced, for example, one kind of bias exists in one tool and a different kind of bias exists in the other tool. When the distinct results appear, people may start to think the reason or sometimes, they try to balance each other. In this case, bias could be reduced. But if different tools indicate the same results, that may not mean the outcomes are correct, perhaps in that situation, bias is strengthened. Every tool has the same bias and shows the same results. Thus, the above bias is easy to be ignored.