Using your Powers of Inference

“I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.”—Abraham H. Maslow, Toward a Psychology of Being

It is a forgone conclusion; that poject professionals must be sharp critical thinkers, able to infer causality based on inaccurate, incomplete or sometimes out of date data. As a people, we have been refining this technique for more than two thousand years, since the pre-Socratics, but what does it mean to be imperative? Critical Thinking, as defined by the National Council for Excellence in Critical Thinking, 1987 is as follows:

"Critical thinking is the intellectually disciplined process of actively and skillfully conceptualizing, applying, analyzing, synthesizing, and/or evaluating information gathered from, or generated by, observation, experience, reflection, reasoning, or communication, as a guide to belief and action. In its exemplary form, it is based on universal intellectual values that transcend subject matter divisions: clarity, accuracy, precision, consistency, relevance, sound evidence, good reasons, depth, breadth, and fairness"[1].

This statement by Michael Scriven & Richard Paul presented at the 8th Annual International Conference on Critical Thinking and Education Reform, 1987, actually breaks down into a few very simple, yet incredibly hard concepts to master. psychological techniques notwithstanding, these concepts form the basis of this article. Causal analysis begins with five basic interrogatives, known as The 5ws:

– what has happened? (the effect or effects),

– how was the effect experienced? (sequential placement),

– who experienced it? (the Stakeholder or Stakeholders),

– when did/does the effect happen? (temporal placement), and

– where did/does it happen (spatial placement).

Moreover, we apply them to people (including ourselves), places, and things. We are interested in only one fundamental question—what is/are the cause or causes of this/these effects or effect. In answering this, we discover if the cause is entirely correct; mostly right; likely to be true, given the right circumstances; not correct; or whichever causes the Stakeholders think are true? Below are the five reasoning types that correspond.

– Deductive Reasoning: What is absolutely true?

– Inductive Reasoning: What is observably, mostly true?

– Abductive Reasoning: What is most likely to be true?

– Reductive Reasoning: What is not true?

– Fallacious Reasoning: What we or the Stakeholders think is/are true?

Although I listed Fallacious Reasoning as a tool, I would not call it an effective one. Quite the opposite, in fact. Fallacies lead to something I call Monological Belief Systems. I cover this in more detail in my soon to be released book.

Whenever someone accuses me of taking too much time on a root-cause analysis, I always remind them that Einstein once said, “If I had an hour to solve a problem I'd spend 55 minutes thinking about the problem and 5 minutes thinking about solutions.” ― Albert Einstein (1875-1955). I would never presume to compare myself to the man who developed the General Theory of Relativity, one of the two columns of modern physics (the other is Quantum Mechanics), but as he did, and I believe, we must spend more time focused on understanding problems and less time just fixing them. A business analyst’s most cherished word is why.

The key to effective problem solving is superior communication, outstanding communication, and even exceptional communication. To relate well with others, we must understand them as well as ourselves. Thus, any problem-solving technique must be rooted in psychological frameworks. A requisite for almost any endeavor is a simple recipe of general intelligence and the far more complicated and difficult to master, emotional intelligence. Concepts so critical to our success, I have set aside another chapter in my book to focus solely on communication and its essential emotional elements. Other factors equally important to problem solving are tenacity, concentration and intense focus, high-functioning comprehension, and a robust Working Memory.

When I first started my BA career my Working Memory, a system for temporarily storing and managing the information required to carry out complex cognitive tasks such as learning, reasoning, and comprehension[2], was just average. I could only hold between five to seven ideas in my head at any one time and they had to be related for me to remember them. I used to carry around a digital recorder and ask my Stakeholders to recite their issues. It was not a good solution, as it made people uncomfortable. To become a better analyst, I knew I had to improve my Short-term or recent memory system. Having the ability to hold many thoughts in our heads at one time is critical to effective reasoning and problem-solving (Cowan 2005). Working Memory is not a single process controlled by one brain region; it is an array of transitive methods working in concert. The figure at the begining of this article[3] illustrates the cognitive areas involved in our Working Memory. Improving these areas helps us problem solve efficiently.

As a general problem-solving technique, Inductive Reasoning is a useful way to generalize when we only have a limited number of observations or raw data with which to work. Used judiciously, it can be astonishingly persuasive. However, if we use insufficient, incomplete or unrepresentative data, our conclusions would be flawed; and so would our subsequent actions. Spending millions of dollars and wasting thousands of hours building solutions based on a false premise, in my early years, was a recurring theme for the organizations I contacted too.

Their initial arguments on the observed effects may have seemed logically sound, but demonstrably wrong because their first premise was false; thus each subsequent premise would have been false, also. The first five years of my business analysis and project management career did not afford me the luxury of discovering causation or shaping a correct and measured response. By the time I was on board, decisions made, and costs sunk, no one was willing to listen to alternatives—not even when it was too late. Much later on in my career, I became known as a person who could bring projects back to life and identify actual causality. My tendency is to lean towards induction as a means to identify causation.

Inductive reasoning relies on our ability to identify patterns observed from the events that surround the effects and, for all practical purposes, reach a unified conclusive cause—if and only if our argument is convincing, evidently. Consequently, training ourselves and our requirements team to induce conclusions using logical inference is not an easy task. It takes a lot of mental training and always starts with a sound premise:

1. If 1, 2, and, three are thought to be true, then P may also be true.

2. The assertion of 1, 2, and three are effects, which we observe in some circumstance.

3. The possible proposition of P must test successfully using a reasonable range of different conditions, usually our experiences and other observed phenomena in and around the problem space.

4. Because this theory of evidence is inherently uncertain, we must identify as much credible data as we can.

5. Once the results are convincing enough, we proceed, with caution, towards a potential solution, though;

we must always retest the theory of proof when circumstances shift. Otherwise, we risk scoping out a solution that will neither meet the intended business goals and objectives nor the Stakeholders’ needs. This translates, to me, as “by the time we finished our analysis the requirements changed”.

You may be wondering why I prefer Inductive Reasoning if it is inherently ambiguous. Shouldn’t we proceed with something more reliable, such as a syllogistic approach, or Deductive Reasoning? Ideally—yes; but recall, Deductive Reasoning, the opposite of Induction, starts with a rule, which proves beyond all reasonable doubt. This top-down approach, in which a determination, based on the concordance of multiple premises, assumes the data to be accurate.

This is an almost impossible task in organizations that deal principally with knowledge work. People, process and technology interact in myriad ways. Even if we could deduce beyond a reasonable doubt, it would take too long to assemble and test all the data. Our circumstances, changed, would require us to start all over again. In most cases, we are left with Inductive and Abductive Reasoning. As a general approach I prefer to follow the following process:

– In a squeeze, when time is of the essence, my data or access to data is limited is restricted, respectively; and I have been told to proceed without all or even some of the facts, then I will use Abductive Reasoning, which is a best-fit argument. What is mostly true, if you will.

– Depending on the situation, I may begin my cause-and-effect analysis with Reductive Reasoning. I do not use it as intended, however. Strictly speaking, this type of inference demonstrates a correct premise by showing that a false or absurd result follows from its denial. My approach is the opposite. I will attempt to disprove a premise that something is causing the circumstance by proving the absurdity of the claim. This is a mixture of deductive and inductive reasoning because we want to show how likely something is to be true or in my case, what is liable to be not correct.

– As a default, I begin my analysis trying to induce from analogies, examples, observations, and Stakeholder experiences to form a certain proposition. I will help the Stakeholders identify testable patterns. If our physical or logical tests hold up, then the premise is sound. Frequently, my thinking will take the form of a matrix. This is where I might have to model my thoughts. Whenever it becomes necessary to correlate and compare similarities between different logical paths, I will create a decision tree or similar type of model. This helps Stakeholders understand complex causalities such as spirals or cycles.

– With a solid premise, my preference is to complete the causal analysis with Deductive Reasoning, building a conclusion that follows logically and coherently from the assumption that we induced earlier. Because we may not be dealing with a pre-established fact, we cannot formally call this Deductive Reasoning, but it is close enough. Remember, inductive inferences do not establish conclusions with absolute certainty, but through observable and predictive certainty, formally, there is a difference.

Critical thinking, as a general method, is applied to a wide variety of problem-solving techniques. There are many different types of inductive reasoning, including simple induction, casual inference, an argument from analogy, generalization, and statistical syllogism. As would a surgeon, it is important we use the right tool for the task. Other than an argument from analogy, I have not had many opportunities to induce with the tools mentioned above.

The formality and rigor I do apply are directly proportional to the scale of the problem that I am trying to resolve. I am not insensitive to how costly and time consuming this process can be, nor do I want to be caught up in paralysis by analysis. Nevertheless, I have seen too many project’s cost increase ten-fold because the proper rigor, not applied during the initial discovery phase, substantially affects its life-cycle with disastrous results—spend one thousand dollars now to save $10,000 later.

Endnotes

[1]    Defining Critical Thinking, http://www.criticalthinking.org/pages/defining-critical-thinking/766 (accessed May 16, 2016).

[2]    Working memory definition - MedicineNet - Health and .., http://www.medicinenet.com/script/main/art.asp?articlekey=7143 (accessed May 15, 2016).

[3]    How can we enhance working memory? | BrainFacts.org Blog, http://blog.brainfacts.org/2013/05/how-can-we-enhance-working-memory/ (accessed May 15, 2016).

References

Cowan N. 2005. Working memory capacity limits in a theoretical context. Human learning and memory: Advances in theory and application. The 4th Tsukuba international conference on memory: 155-175.