Causes and Correlations

I have not posed an article since 2018. You can expect a few in rapid succision as a catch up.

There are no accidents in my philosophy. Every effect must have its cause. The past is the cause of the present, and the present will be the cause of the future. All of these are links in an endless chain stretching from the finite to the infinite. —Abraham Lincoln

Business Analysis, for me, was mostly a hunt-and-peck affair. Having come from the IT world as a UNIX administrator and Oracle DBA, my experience with various business units was limited. Feeling downcast, early in my career, never knowing what conversations I should have, when, and with whom, I knew there had to be a better way. How should I identify requirements? Where do I start? In the early 1990s, there was little to no reference material available.

That is when I began to create my materials and techniques, which I slowly developed and refined throughout my career. My ideas have never been original, but my approach is. The words I use, the combinations, the mixing and matching of the old and the new, make it my own. My philosophy has always been that business analysis is primarily rooted in Organizational and Behavioral Physiology.

Nevertheless, at the root of it, all are the Stakeholders. We spend most of our time with people, listening to their professional issues, problems, dislikes, anything, and everything that prevents them from reaching a sustainable productivity level within their work environment.

I have identified many of the issues over the years' source back to cultural shortfalls- organizational habits that just seem too impossible to resolve. Organizational and Behavioral Physiology has always seemed to be an excellent way to understand the needs of my Stakeholders.

Admittedly, I may have to employ traditional techniques in the process, but the emphasis is not on tradition, preferably on a more obscure way of thinking. We are not looking for a particular cause so much as we are looking for Causality. As Abraham Lincoln (1861–1865) reminds us, "All of these are links in an endless chain…". It is our duty as professionals to identify Causality. We must not be satisfied with mere Band-Aid solutions, casual symptoms, and easy fixes. Then, even if we think that there is something bigger going on in the background, we are still reluctant to take action. We must have a deeper understanding of causation and its close cousin, correlation.

Causality, Agency, and correlation are at the heart of Source identification. Correlation does not always imply causation. This concept is a questionable cause fallacy. The reason is known as cause and effect, whereas correlation infers relationships. Both require independent and dependent variables. An independent variable can affect while a dependent variable is controlled or influenced by outside factors.

Agency is the actual thing that caused the action or intervention that produced the effect. For example, a book or article can sometimes change the way you think about a particular subject. In this case, the book itself played the role of the agent. Agents cause things to happen, and cause always has one or more effects. Sometimes we believe in unseen agents acting negatively. This statement is a false correlation known as a false-pattern—more on this in an upcoming chapter. Our first step in identifying a Source for our REPAC set is determining if we are dealing with causation or correlation.

Correlation is a relationship between two sets of variables used to describe or predict information. If there is a correlation, we can sometimes assume the dependent variables change solely because the independent variables change. We may see a correlation and find Causality, but not always. Correlations can be positive or negative. For example, a positive correlation would be the number of Helpdesk tickets and the frequency of change requests for a particular system.

The dependent and independent variables increase or decrease together. Nonetheless, this does not imply Causality. Conversely, in a negative correlation, the dependent and independent variables increase or drop opposite each other. In another example, a business analyst might notice the more Requirements there are on a project, the less time she seems to have to analyze and package them, but the more time she seems to spend on them. Once again, correlation does not imply causation. Upon closer examination, we may discover that although these issues correlate to one another, they neither cause nor affect each other. This idea seems counterintuitive since the examples given become cited as causal issues for project lateness, cost overruns, and the like. Misunderstanding the differences between correlative events and actual causation and Agency are systemic in many projects I work on today.

Causation, on the other hand, is much harder to see. As a species, we tend to apply the Agency to events. In the social sciences, Agency is our ability to make choices. When an Agent is perceived, unknown, or unseen, we tend to see Causality where none exists. Take the concept of time, for example. When we think about time, we tend to think about a belief known as The Law of Averages. In fallacious terms, this is the Gamblers Fallacy. 

We did not evolve to understand randomness; we tend to believe if two events happen in sequence, the first must have caused the second. This belief system can be very challenging when we are trying to predict how long something will take. When event Q directly follows event P, our brains, by default, perceive a direct correlation to a third, often unseen agent. We then draw causation between the agent and the events. Is the relationship between poorly written Requirements and defective solutions correlative or Causality? Is there an unseen agent controlling the outcomes?

As you explore the root cause of your business problems, be sure to keep an open mind and don't allow yourself to think fallaciously.