One of the benefits of being someone who works in data science is that I get a lot of unusual requests for information in order to help the decision-making of others. As would come to no surprise to those who know me personally, I have a particular logic about the way I go about this task. I tend to be fairly explicit about my assumptions, and I also tend to be generous with the benefit of the doubt when it comes to calculations. At times the data is frightening enough already, one does not want to prejudice decisions by being unkind. I am aware that this sort of approach to handling data, being very communicative about the inner logic of how I go about solving a problem as well as being kind and generous-minded in regards to the conclusions I draw, is probably unusual. Yet even someone who has an orderly and disciplined mind is still a human being, and there are always subjective decisions to be made. Why not show oneself to be generous to others where there is a doubt, in the hope at least that others may learn and profit from the example, and that one may receive a similar generosity from others during our own times of judgment. This hope is not always realized, and others may not consider me a particularly generous-minded person, but that is at least the standard I seek to attain.
One of the consistent areas of interest that makes this sort of work more exciting, and that is relevant in other areas of life as well, is trying to figure out the logic of the problem. There is usually far more data to look at from the beginning than is really useful, and so one seeks to make the best decisions by aggregating the data together. Since most of the data is far too complicated and massive for any one person to master completely, there are various means, like pivot tables or aggregation queries, by which one summarizes the data in some fashion so as to figure out the essential part of the data that tells the story one wants. The same data can tell many different stories, if you only know how to ask it the right questions. I have found in life that people are the same way. There are some stories data will tell over and over again, but which no one pays attention to, and there are other stories that require very delicate finesse in finding out because it requires looking at the data in a way that one may be unfamiliar with. Usually, if I am trying to convey a point to someone else, I try not to make it too hard to figure out. To be sure, data is full of layers and intricate correlations that can be made, but at the same time each layer and each story should be readily comprehensible if one is simply literate to what the data says and how one goes about reading it.
A proper understanding of the data itself then makes the logic of the problem easier to understand. Sometimes we may not fully understand what we want. Sometimes the data itself is bad—subject to error and corruption like all areas of life. Sometimes we may only have a vague and inchoate idea of what it is that we want to know, and so we may not be able to wrangle the data correctly so as to tell us the answer we want, even if the answer is not something we want to hear. Generally speaking I try to treat data the way I treat other people; I ask questions and want to know the answers, and I will not take any action to prejudice the answer to the best of my abilities, because what I want is an honest answer, and presumably a relationship of some sort with the data that will allow for much greater ease of use and familiarity. To be sure, sometimes data can be difficult to work with, and sometimes it may require formatting and it may be fairly sensitive and temperamental, but being the sort of person who is fairly sensitive myself, I tend to be reasonably patient (at least in my own estimation) with the data I find myself working with. Again, I figure that someone who requires patience ought to show patience with others, even if they are numbers in a computer program that one is working with.
What is the payoff for this effort and this patience? The hope is that one is able to come to an accurate answer that allows people to make better decisions and, if necessary, change behaviors so as to achieve better results in the future, to figure out what went down and why, and how we can grow and improve. Life is the same way; our experiences are complicated pieces of data, subject to our interpretation and the interpretation of others, subject to human frailty and error, and also with the aim of helping to inform our decision-making skills so that we behave more wisely in the future. There is far more data than anyone ever looks at and tries to make sense of, and there are far more opportunities we have in our lives to learn than we ever take advantage of either. The real question is, do we understand the logic of the problems in our own lives and in our own contexts, and are we successfully able to wrestle with our experiences, and to deal patiently and generously with other people as we seek a better life not only for ourselves, but for all of those we happen to come into contact with. This is as tall order, but just about everything worth doing in life is a task of considerable challenge and ambition. After all, that’s what makes life fun, even for someone like myself.

Pingback: Reducto Ad Caiaphas | Edge Induced Cohesion