It is Not All Statistics – Problem solving and Decision Making

It is Not All Statistics to Achieve Good Understanding for Problem solving and Decision Making.

As a manager, you are expected to act as an expert on what will happen in the future.  You need to solve problems, forecast, budget, and calculate risk; make decisions and also communicate information and projections to other decision makers.  Your main job is to manage – to solve problems and make decisions.  Putting aside luck and talent for a while, to solve problems and make good decisions you need two primary things: information and decision making tools/methods.

Information also covers two key elements – information and knowledge (experience) that you already have and information you should get (research).

Decision making tools and decision making methods are what is needed to use the information to make decisions.  The overall method is called understanding.  The most common method of which is analysis.  (Analysis means literally to break a complex problem down into smaller, more manageable “independent” parts for the purposes of examination — with the hope that solving these smaller parts will lead to a solution of the more complex problem as well.  There are many types of analysis, for which statistical analysis is used in many as a tool) Most managers tend to be fairly poor at this.   Understanding may alternatively use synthesis instead of analysis The term synthesis means using a process which combines two or more pre-existing elements and results in something new.

Over some period of time, one of the primary tools of analysis,  statistics (the mathematical science pertaining to the analysis, interpretation or explanation of data, and the presentation of information), has become not only preeminent in analysis but in the majority of minds it is the only acceptable method – totally synonymous with analysis.  This is probably due to the belief that the only worthwhile analysis must be numeric – despite the fact that many variables (more on this later) cannot adequately or effectively be measured.

The high priests of statistics have of course a vested interest as do the leading acolytes who believe their status is improved by appearing to be more scientific, objective and analytic.  Statistics as the only method of analysis is not only theoretically incorrect it is also practically dangerous.  You may have the numbers but they might not refer directly to your problem

Analysis is a method for approaching and solving problems, making decisions, and conveying information to other people.  Other decision making methods include wild gambling, guesswork; assumptions etc.

This blog does not seek to denounce or reject statistics (eg. statistics are mendacious truths; lies, dam lies and statistics).  Far from it. Statistics are an excellent tool in the toolbox of analysis – but it is not the first one we should use and perhaps we should not use it at all (even allowing for the fact that most folk cannot use stats very well anyway)

Before using statistics for decision making analysis a manager should understand:

  • The need to understand the whole problem and apply logic
  • The basis and nature of the problem(s) at hand  (What will the value of the analysis and decision?  Will the cost of the process be too high?
  • The basis and nature of the outcomes of the analysis and decision making (is a solution likely, desirable  or possible?
  • The wider implications (and future implications) of the problem/decision making
  • All the possible variables involved (daunting – but necessary)
  • The nature of those variables
  • The type and extent of the data collection methods and the sampling to be used
  • The options available (and what resources will be needed to implement those options)

Having done this, then you then decide on the best tools to use.  After all a mechanic does not first approach the engine with a wrench!  The nature of the problem and its variables need to be understood first.

Another factor to consider about statistics is that they provide solutions to mathematical methods – the answer to an equation is not the same thing as the solution to the problem.  It needs interpretation.  Without the prior work in understand the problem etc, the interpretation is likely to be well off the mark.

Try this experiment: Give several people an identical problem and series of facts about it and ask them to draw a conclusion.  You will be amazed at the variety of different answers you receive.  People are creative in their interpretations.  They will perceive both the problem itself and the facts they receive in dissimilar ways.  It is inaccurate to assume that a problem has one “right” solution, or that a question can have only one “right” answer.  The spectrum of possible interpretations may be infinite.  That’s what makes people so interesting-each of us has our own approach to problems.  But this same condition also adds to the challenge we all face in business, every day and every time we interact with someone else.

Why write a blog about this?  Well because we constantly see or hear of research that was not properly conceived, but very well analysed statistically, which of course provided great answers, just not the right ones for the problem.  Is it any wonder managers seem to have less faith in research?

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s