Drawing Good, Generalized Conclusions
Inductive reasoning involves making useful generalizations about the environment as a whole, based upon a necessarily limited number of observations.
Scientific method, which has done so much to advance humanity in the last 500 years, is inductive reasoning in its purest form. Used appropriately, it can be incredibly powerful. But, if you use faulty or unrepresentative data, your conclusions can be flawed.
At the core of inductive reasoning is the ability to look at outcomes, events, ideas and observations, and draw these together to reach a unified conclusion. Considering this, an experienced business person can use his or her own experiences to draw conclusions about current situations and solve problems based on what he or she has known to work in the past in similar situations.
By accepting conclusions derived from inductive reasoning as "true" (in a practical sense), good managers can build on these conclusions and move forward effectively and successfully.
How to Use the Tool
Much inductive reasoning happens intuitively and automatically. Without it we just couldn't function: Everything we did would have to be subject to so much analysis and consideration that we'd "grind to a halt".
At the other extreme from this intuitive inductive reasoning, we have the formal scientific method we're taught at school:
- Formal statement of a hypothesis, a suggested generalized "truth.
- Design of an experiment – a series of tests or observations designed to test the hypothesis in a fair and reasonable range of different circumstances, and with complicating factors "controlled" as far as possible.
- Proper documentation of the tests carried out (the "Method") and the "Results" of the experiment, so that these can be subjected to later scrutiny.
- Drawing of "Conclusions", technically establishing whether the hypothesis has been shown to be false or not, but in practice establishing whether the hypothesis is likely to be true.
(This is also where ideas of statistical significance become important, in mathematically evaluating how good conclusions are likely to be.)
And in between these two extremes, we have different levels of formality of inductive reasoning that circumstances allow and demand.
The formality of the process used to reach conclusions largely depends on the scale of the problem being solved.
Where an issue affects very many people significantly and reasonable time is available, then reasoning had better be thoroughly researched and checked – otherwise, huge errors can be made.
However, this research and checking can be slow and costly, and can cause "paralysis by analysis". Small decisions, or those needing quick resolution, often will not allow detailed research. This is where a less formal (and necessarily more error-prone) approach is necessary, and is why experience is important in good business decision making.
This is not, however, an easy get-out for lazy decision making: You need to do as much research as you can with the resources and time available.
Experience speeds inductive decision making, and makes it more robust and more likely to be correct. On the other hand, people relying on experience too much can miss important new factors in a changing environment. This is where businesspeople must keep at least a certain amount of open-mindedness, and need to keep scanning the environment for change.
There are many different types of inductive reasoning, including simple induction, causal inference, prediction, argument from analogy, generalization and statistical syllogism. These are beyond the scope of this article.
In the past, when we offer customers the chance of winning a prize if they fill in a questionnaire, we get a 9 percent response rate. If we don't offer a prize, we get a 3 percent response rate. We're going to tell customers who respond to our latest survey that respondents will go into a free prize draw, so we'll get a response of around 9 percent.
Using inductive reasoning, past experience is being applied to a future situation to predict an outcome. The past experience that is being applied seems to be appropriate, but there is no guarantee that it will definitely apply in the future. It may be that customers have got bored of filling in questionnaires and won't reply at the same rate this time.
An alternative to inductive reasoning is deductive reasoning. While inductive reasoning starts with facts and induces a rule from them, which is then used to make an estimate of what will happen in the future, deductive reasoning works the other way. It starts with a rule, that has been proven beyond all reasonable doubt, and calculates an outcome from that. For example, if a bank offers a 4 percent interest rate, and you deposit $100 for a year, you will get $104 at the end of the year. Using inductive reasoning to calculate what you investment would be worth after a year, you would ignore the currently advertised interest rate and say, "Last year, I got $6 interest on my $100 investment, so I expect to get the same this year."
Inductive reasoning is the process by which we make a necessarily limited number of observations and seek to draw generalized conclusions from them.
The most rigorous inductive reasoning is conducted through formal experiments, much as we're taught at school.
However, decisions which are small in scope or need to be made quickly often do not allow the overheads of extensive method and experimentation: Here we have to accept that the "truths" we seek to establish may be faulty, and generalizations may be misleading. This is where management of uncertainty becomes important.